Complement System: Pathways and Function
A patient with a genetic deficiency in a single small plasma protein, C3, suffers recurrent, life-threatening bacterial infections. This happens despite having a fully normal antibody repertoire and normal phagocytes. Why should one missing protein cripple host defense so broadly? The answer is that C3 sits at the convergence point of three separate surveillance pathways. Its cleavage is the hinge on which almost all of complement's power turns. Complement is not one molecule but a cascade of roughly 30 plasma and cell-surface proteins. These circulate in inactive form and detect danger within seconds of exposure. This happens long before adaptive immunity has mounted a response. Complement matters because it is the fastest humoral arm of Innate Immunity. It acts as a biochemical alarm system that tags microbes for destruction. It also recruits phagocytes and can punch holes directly in microbial membranes. It bridges innate and adaptive immunity: complement fragments coat antigens for B-cell recognition. Antibodies in turn activate complement, creating a feedback loop between the two systems. Deficiencies in complement components are among the most instructive natural experiments in immunology. Each one reveals exactly which pathway or effector function that protein normally supplies. The big picture is one of convergent, amplifying enzymatic cascades. Three distinct initiation routes exist: classical, lectin, and alternative. These are triggered by different molecular cues. The cues are antibody-antigen complexes, microbial sugar patterns, and spontaneous low-level activation, respectively. All three routes funnel into cleavage of the central protein C3. This generates opsonins, inflammatory mediators, and the building blocks of a pore-forming membrane attack complex (MAC). The system is a proteolytic cascade in the same design logic as blood clotting. Each enzyme cleaves and activates the next. This amplifies a small initial signal into a massive, localized response. This article works through the molecular mechanism of each pathway step by step. It covers the classical experiments (Bordet, Ehrlich, Nelson, and the modern deficiency studies). These experiments established how complement functions. The article also presents two worked quantitative examples and common conceptual traps. It covers the history of the discovery and its modern clinical relevance. This sets up later coverage of how complement activity is restrained and regulated.
- Complement is a cascade of ~30 plasma proteins, activated via three pathways (classical, lectin, alternative) that all converge on cleaving C3.
- C3 convertase is the central enzymatic event; its formation determines whether opsonization, inflammation, and lysis proceed.
- C3b opsonizes microbes for phagocytosis; C3a and C5a are anaphylatoxins that drive inflammation; C5b initiates the membrane attack complex (MAC).
- The alternative pathway is antibody-independent and provides constant low-level innate surveillance via spontaneous C3 "tick-over."
- Jules Bordet first showed serum contains a heat-labile lytic factor distinct from heat-stable antibody; this factor is complement.
- C3 deficiency causes severe pyogenic infections; terminal component (C5-C9) deficiency causes selective susceptibility to Neisseria.
All three complement pathways exist to generate C3 convertase, because cleaving C3 is the single step that converts a silent plasma protein into an opsonin, an inflammatory signal, and the trigger for membrane-destroying pore formation.
The core idea
The complement system is a set of soluble and membrane-bound proteins, made mostly by the liver and circulating in the blood and tissue fluid as inactive zymogens. "Complement" refers to the fact that this system complements — enhances — the ability of antibodies and phagocytes to clear microbes, a naming choice that reflects its historical discovery as an accessory to antibody-mediated killing. In modern terms it is better understood as an independent innate defense system that antibody merely recruits, rather than a mere adjunct to adaptive immunity. Three activation pathways converge on a shared trunk. The classical pathway is triggered when C1q binds antibody (IgM or clustered IgG) already bound to an antigen surface. The lectin pathway is triggered when mannose-binding lectin (MBL) or ficolins bind repeating sugar patterns on microbial surfaces — a pattern-recognition mechanism structurally analogous to Toll-like receptor recognition of Pattern Recognition and PAMPs. The alternative pathway is triggered continuously, at low level, by spontaneous hydrolysis of C3 in plasma, and it is amplified specifically on surfaces that lack the regulatory proteins found on host cells — meaning it discriminates self from non-self by the absence of protection rather than the presence of a specific ligand. All three pathways generate an enzyme complex called C3 convertase, which cleaves the abundant plasma protein C3 into two fragments: C3b (which covalently attaches to the microbial surface) and C3a (a small soluble peptide). C3b deposition triggers formation of C5 convertase, which cleaves C5 into C5b and C5a. C5b nucleates assembly of the membrane attack complex (MAC), a ring-shaped pore of C5b-6-7-8-9 that lyses susceptible cells directly. This convergence — three distinct recognition strategies feeding one shared effector machinery — is the conceptual core of the whole system.
Mechanism
Step 1 — Classical pathway initiation. IgM (pentameric, so a single molecule bound to antigen can activate complement) or two or more clustered IgG molecules bind antigen on a microbial surface. The C1 complex — C1q bound to two C1r and two C1s molecules — recognizes the Fc regions of the bound antibody. C1q binding induces a conformational change that activates C1r, which then cleaves and activates C1s, a serine protease.
Step 2 — Lectin pathway initiation. Independent of antibody, mannose-binding lectin (MBL) or ficolins bind repeating mannose, fucose, or N-acetylglucosamine residues arrayed on microbial surfaces (a pattern rarely presented by host glycoproteins in that geometry). MBL is structurally and functionally analogous to C1q. Bound MBL associates with MBL-associated serine proteases (MASP-1 and MASP-2), which perform the same proteolytic role as C1r/C1s.
Step 3 — C4 and C2 cleavage. Activated C1s (classical) or MASP-2 (lectin) cleaves C4 into C4a and C4b, and cleaves C2 into C2a and C2b. C4b attaches covalently to the microbial surface via an exposed thioester bond, and C2a binds C4b to form the C4b2a complex — the classical/lectin pathway C3 convertase.
Step 4 — Alternative pathway initiation and amplification. C3 undergoes slow spontaneous hydrolysis in plasma to C3(H2O), which binds Factor B. Factor D cleaves bound Factor B to Bb, generating a fluid-phase C3 convertase, C3(H2O)Bb, that cleaves a small amount of C3 to C3b continuously. This C3b can attach to any nearby surface. On host cells, regulatory proteins (Factor H, DAF, MCP — covered in depth under complement regulation) rapidly inactivate surface-bound C3b. On microbial surfaces, which lack these regulators, C3b persists, binds Factor B, and Factor D cleaves it to form C3bBb, the alternative pathway C3 convertase, stabilized by properdin (Factor P). This creates a self-amplifying feedback loop: C3b generates more C3 convertase, which generates more C3b.
Step 5 — C3 cleavage: the convergence point. Both C3 convertases (C4b2a from classical/lectin; C3bBb from alternative) cleave C3 into C3a and C3b at a massive rate — a single convertase molecule can cleave hundreds of C3 molecules, providing enormous amplification from a small initiating signal. C3b coats the microbial surface (opsonization) and also feeds back into more convertase formation.
Step 6 — C5 convertase and MAC formation. Accumulating C3b binds to the existing C3 convertase, converting it into a C5 convertase (C4b2a3b for classical/lectin, C3bBbC3b for alternative). C5 convertase cleaves C5 into C5a (released into fluid phase) and C5b, which remains anchored to the surface. C5b sequentially recruits C6, C7, C8, and multiple copies of C9, which polymerize into a ring that inserts into the lipid bilayer, forming the membrane attack complex — a transmembrane pore roughly 10 nm in diameter that permits free flow of ions and water, causing osmotic lysis, most effective against Gram-negative bacteria and some enveloped viruses.
Step 7 — Effector outputs. Three biologically distinct outputs radiate from this cascade:
- Opsonization: C3b and its degradation fragment iC3b coat microbial surfaces and are recognized by complement receptors (CR1, CR3) on phagocytes, dramatically enhancing Phagocytosis: Mechanism and Killing.
- Inflammation: C3a and C5a are anaphylatoxins — small peptides that bind G-protein-coupled receptors on mast cells, basophils, and endothelium, triggering degranulation, vasodilation, increased vascular permeability, and chemotaxis of neutrophils, reinforcing the events of Inflammation: The Innate Response to Tissue Damage. C5a is by far the more potent anaphylatoxin and also the most powerful chemoattractant known for neutrophils.
- Lysis: the MAC directly kills susceptible cells, most importantly Gram-negative bacteria such as Neisseria, whose thin peptidoglycan and exposed outer membrane make them uniquely vulnerable.
The evidence
Bordet's discovery of "alexine" (1895–1898). Jules Bordet, working at the Pasteur Institute, showed that fresh serum from an animal immunized against Vibrio cholerae could lyse the bacteria, but serum that had been heated to 56°C for 30 minutes lost this lytic ability even though it still agglutinated the bacteria. Crucially, adding fresh, non-immune serum (heated or not, containing no specific antibody) back to the heat-inactivated immune serum restored lytic activity. This proved two things: (1) killing requires two distinct components — a heat-stable, specific factor (later shown to be antibody) and a heat-labile, non-specific factor present in any normal serum; (2) the heat-labile factor, which Bordet called "alexine" (Ehrlich later renamed it "complement"), is not itself specific for any antigen and can be supplied by unrelated serum. This experiment is the founding proof that antibody-mediated killing depends on an accessory, heat-labile serum system distinct from antibody itself.
Ehrlich's side-chain theory and terminology. Paul Ehrlich, working somewhat independently, proposed that antibody ("amboceptor") binds antigen and then a separate serum substance — "complement" — completes the lytic reaction, coining the name that persists today. The Bordet-Ehrlich priority dispute over interpretation was intense, but the operational fact both established — heat lability of the lytic factor, restoration by fresh normal serum — has been reproduced innumerable times and remains the textbook demonstration that complement is a separate, heat-sensitive serum system.
The sheep red blood cell hemolysis assay (Complement Fixation Test). Building on Bordet's findings, the classical CH50 (50% hemolytic complement) assay became the standard quantitative test: serial dilutions of test serum are mixed with antibody-sensitized sheep erythrocytes, and the dilution producing 50% lysis defines one CH50 unit. This assay, still used clinically to screen for complement pathway defects, directly demonstrates the antibody-dependence of the classical pathway and the strict requirement for an intact, unheated serum complement source — heating serum abolishes CH50 activity, exactly reproducing Bordet's original observation biochemically.
Pillemer's discovery of the alternative pathway (1954). Louis Pillemer and colleagues found that zymosan (a yeast cell wall polysaccharide) could deplete serum complement activity and activate C3 without any antibody present, via a serum factor he called "properdin." This was initially controversial — critics argued Pillemer had merely detected trace natural antibody — but subsequent work using antibody-deficient and genetically agammaglobulinemic sera confirmed that C3 activation and microbial killing could proceed via a properdin/Factor B/Factor D route independent of antigen-specific antibody. This established the alternative pathway as a bona fide antibody-independent innate defense mechanism, decades before pattern-recognition theory gave it a conceptual home.
Natural "experiments": human complement deficiencies. Genetic deficiencies in individual complement components, discovered from the 1960s onward, function as clean loss-of-function experiments in humans. C2 deficiency (the most common classical pathway deficiency) causes increased susceptibility to encapsulated bacteria and lupus-like autoimmunity, revealing complement's role in clearing immune complexes. C3 deficiency causes severe, recurrent pyogenic infections, confirming C3's non-redundant, convergent role in opsonization. Deficiencies in the terminal components C5–C9 cause a strikingly narrow phenotype — recurrent Neisseria infections (meningitis, gonorrhea) with otherwise normal health — proving that the MAC's physiological importance is largely restricted to killing this one genus of thin-walled, complement-sensitive bacteria, since these patients handle nearly all other pathogens normally via opsonization and phagocytosis alone.
Worked examples
Example 1 — Predicting the phenotype of a Factor D-deficient patient. Factor D is required only for cleaving Factor B in the alternative pathway amplification loop; it plays no role in the classical or lectin pathways, which use C1s/MASP-2 to cleave C4 and C2 instead. Reasoning through consequences:
- Classical pathway: unaffected — antibody-mediated complement activation and CH50 assay (which measures the classical pathway) will be normal.
- Lectin pathway: unaffected — MBL-driven activation proceeds normally.
- Alternative pathway amplification loop: abolished — no Factor B cleavage, no C3bBb convertase, no properdin-stabilized amplification.
- Net consequence: initial C3b generation via classical/lectin pathways still occurs when antibody or MBL is present, but the powerful antibody-independent amplification loop is lost.
- Predicted clinical phenotype: recurrent Neisseria infections, similar to terminal pathway deficiencies, because the alternative pathway's amplification loop is especially important for generating enough MAC to kill Neisseria efficiently, and antibody responses to encapsulated Neisserial polysaccharides are often too slow or too weak (especially in young children) to drive sufficient classical-pathway C3b deposition on their own.
This matches the actual clinical observation: Factor D and properdin deficiencies present with a Neisseria-restricted susceptibility pattern much like terminal component deficiencies, distinguishing them from C3 deficiency (which is far more broadly catastrophic) and from C1/C2/C4 deficiency (which presents more with immune-complex disease than infection).
Example 2 — Quantifying amplification through the cascade. Suppose a single antibody-antigen complex activates one C1 complex, and assume (for illustrative simplicity) the following turnover numbers, roughly consistent with the enzymatic amplification known to occur at each step:
| Step | Molecules activated per enzyme per unit time |
|---|---|
| One C1s cleaves C4 | ~30 C4 molecules |
| One C4b2a convertase cleaves C3 | ~200 C3 molecules |
| Fraction of C3b that nucleates C5 convertase and progresses to MAC | ~5% |
Starting from 1 activated C1 complex:
- C4 molecules cleaved: 1 × 30 = 30 C4b (each potentially forming a convertase, but take a conservative estimate that 10 of these successfully form stable C4b2a convertases)
- C3 molecules cleaved by 10 convertases: 10 × 200 = 2,000 C3b molecules deposited on the microbial surface
- C5 convertases formed and progressing toward MAC: 2,000 × 0.05 = 100 sites of MAC assembly
From a single initiating antibody-binding event, the cascade generates on the order of hundreds to thousands of opsonizing C3b molecules and roughly 100 potential lytic pores — a striking illustration of why complement is described as an amplifying cascade rather than a linear signaling pathway, and why even trace antibody binding (or a single IgM pentamer) can trigger a disproportionately large, locally destructive response. (These specific multipliers are illustrative approximations to teach the logic of enzymatic amplification, not measured biochemical constants — the qualitative conclusion, massive amplification from few initiating events, is the well-established finding.)
Common misconceptions & traps
| Misconception | Correction |
|---|---|
| "Complement is just part of the antibody response." | The alternative and lectin pathways act without any antibody; complement is an independent innate system that antibody can recruit but does not require. |
| "C3a and C5a are the opsonins." | C3b (and its fragment iC3b) is the opsonin; C3a and C5a are anaphylatoxins driving inflammation, a distinct function. |
| "The membrane attack complex is the main way complement clears most bacteria." | Most Gram-positive bacteria are resistant to MAC lysis (thick peptidoglycan blocks pore insertion); opsonization and phagocytosis are the dominant clearance mechanism for most pathogens. MAC-mediated lysis is disproportionately important mainly against Neisseria and a few other Gram-negatives. |
| "All complement pathways use the same convertase." | Classical/lectin pathways form C4b2a; the alternative pathway forms C3bBb — different subunit composition, same catalytic function (cleaving C3). |
| "Complement activation is triggered only on microbes." | Activation (especially alternative pathway tick-over) occurs continuously and non-specifically in plasma; discrimination from host cells depends on regulatory proteins on the host surface, not on the initiating chemistry itself. |
The scientist behind it
Jules Bordet (1870–1961) was a Belgian physician-immunologist working at the Institut Pasteur in Paris under Élie Metchnikoff before returning to Brussels to found the Pasteur Institute of Brabant. Trained in bacteriology at a moment when the field was consumed by the humoral-versus-cellular immunity debate (antibody theorists versus Metchnikoff's phagocyte theorists), Bordet occupied an unusually synthetic position: he took antibody-mediated bacteriolysis seriously while insisting a second, non-specific serum factor was mechanistically indispensable. His meticulous heat-inactivation and reconstitution experiments with cholera vibrios were technically demanding for the era — requiring precise temperature control, fresh serum handling, and quantitative assessment of bacterial lysis — and his insistence on distinguishing the heat-stable specific factor from the heat-labile non-specific one directly anticipated the modern antibody/complement distinction. He also co-discovered the causative agent of whooping cough (Bordetella pertussis, named for him) and developed the complement fixation test into a diagnostic tool, later adapted by August von Wassermann for syphilis serology.
Origins · 1 minJules Bordet and how this was discovered
Where this matters today
Complement deficiencies are screened clinically using the CH50 (classical pathway) and AH50 (alternative pathway) assays, both direct descendants of Bordet's original hemolysis experiments; an abnormal CH50 with normal AH50 localizes a defect to C1, C4, or C2, while the reverse pattern points to Factor B, D, or properdin. Patients with terminal-pathway (C5-C9) or properdin/Factor D deficiencies are specifically vaccinated against meningococcus and often given prophylactic antibiotics, a direct clinical application of the Neisseria-restricted phenotype described above. Complement is now a validated drug target. Eculizumab, a monoclonal antibody that blocks C5 cleavage, prevents MAC formation and is used to treat paroxysmal nocturnal hemoglobinuria (PNH), a disease in which red cells lack the GPI-anchored complement regulators DAF and CD59 and are destroyed by uncontrolled complement activation — a striking real-world illustration of what happens when the regulatory checks on this same cascade fail (the subject of complement regulation more broadly). Complement receptor engagement is also exploited pharmaceutically and diagnostically: opsonization with C3b is used to enhance antibody-based cancer immunotherapies, and complement fixation tests remain in use for diagnosing certain viral and fungal infections. In transplantation and autoimmune disease, complement inhibitors (targeting C1, C3, or C5) are active areas of drug development for conditions ranging from age-related macular degeneration (driven partly by alternative pathway dysregulation, notably variants in Factor H) to atypical hemolytic uremic syndrome. Agriculturally and in vaccine design, understanding lectin-pathway activation by microbial glycan patterns informs adjuvant design intended to boost innate recognition of subunit vaccines.
Ask the tutor: How do regulatory proteins like Factor H, DAF, and CD59 keep the amplifying alternative pathway from attacking host cells?
Why does C3 deficiency cause more severe disease than deficiency in any single pathway-specific protein like C1q or MBL?
Because C3 is the shared convergence point downstream of all three pathways, its loss abolishes opsonization, anaphylatoxin generation, and MAC formation regardless of which recognition route (antibody, lectin, or spontaneous) initially detected the microbe. A deficiency in C1q or MBL removes only one of three parallel entry routes into the cascade, and the remaining two pathways can often still generate enough C3b for adequate host defense — which is why isolated MBL deficiency is common in the population and often clinically silent. C3 deficiency, by contrast, is functionally equivalent to losing the entire complement system at once, producing severe, recurrent pyogenic bacterial infections similar to those seen in antibody deficiency syndromes.
Why are Gram-negative bacteria like Neisseria especially vulnerable to the membrane attack complex, while most Gram-positive bacteria are not?
The MAC kills by inserting a ring-shaped protein pore directly into a lipid bilayer, allowing uncontrolled ion and water flux that osmotically lyses the cell. Gram-negative bacteria have a thin peptidoglycan layer and an exposed outer membrane that the MAC can access and insert into. Gram-positive bacteria have a thick peptidoglycan cell wall (tens of nanometers) that physically prevents MAC components from reaching the underlying plasma membrane, even though C3b opsonization and phagocytosis still work normally against them. This explains why terminal-pathway complement deficiencies produce a strikingly narrow infection susceptibility limited mainly to Neisseria species rather than broad bacterial vulnerability.
How does the lectin pathway distinguish microbial surfaces from host cells if it does not use antibody?
Mannose-binding lectin and ficolins recognize specific carbohydrate patterns — repeating mannose, fucose, or N-acetylglucosamine arrays presented at high density and particular spacing — that are common on microbial cell walls and capsules but rare in that dense, repetitive arrangement on mature host cell glycoproteins, which are typically capped with sialic acid. This is pattern recognition in the same conceptual family as Toll-like receptor recognition of PAMPs: the ligand itself is not exclusively microbial, but its density, geometry, and terminal exposure on host cells is masked by sialylation, giving the lectin pathway an approximate but effective self/non-self discrimination.
What is the biological difference between C3a and C5a if both are anaphylatoxins?
Both are small peptide fragments released during complement activation that bind G-protein-coupled receptors on mast cells, basophils, and endothelial and smooth muscle cells, triggering degranulation, vasodilation, and increased vascular permeability. C5a, however, is substantially more potent on a molar basis and additionally functions as one of the most powerful known chemoattractants for neutrophils, actively recruiting phagocytes to the site of complement activation. C3a is generated in far greater quantity (because C3 is more abundant and is cleaved by convertases far more frequently than C5), so the two fragments differ in both potency and the relative scale of local production.
Why did Pillemer's discovery of the alternative pathway face so much skepticism initially?
Because at the time, complement activation was understood almost exclusively through the classical, antibody-dependent lens established by Bordet and Ehrlich decades earlier, and critics argued that Pillemer's zymosan-activation experiments were simply detecting trace natural antibodies already present in normal serum rather than a genuinely antibody-independent pathway. The controversy was resolved only when researchers demonstrated C3 activation and complement-mediated effects using serum from antibody-deficient (agammaglobulinemic) sources, proving the reaction could proceed with no antibody present at all. This vindicated properdin and the alternative pathway as a distinct, physiologically important route, decades before the conceptual framework of innate pattern recognition made such antibody-independent detection intuitive.
How does the CH50 assay work, and what does an abnormal result tell a clinician?
The CH50 assay measures the classical pathway's ability to lyse antibody-coated sheep red blood cells: serial dilutions of a patient's serum are added to a fixed number of sensitized erythrocytes, and the dilution causing 50% lysis defines the CH50 titer. A normal CH50 indicates the classical pathway components (C1 through C9) are functionally intact and present in adequate amounts. An abnormally low or absent CH50 points to a deficiency somewhere along that pathway and prompts follow-up testing (individual component assays) to localize exactly which protein is missing, exactly as Bordet's original heat-inactivation and serum-restoration experiments first demonstrated that complement activity could be selectively abolished and restored.
Does complement act only against bacteria, or does it have broader roles?
Complement's classical role is antibacterial, but its functions extend well beyond that. C3b opsonization helps clear immune complexes from circulation, preventing their deposition in tissues (a function whose failure underlies lupus-like disease in early classical-pathway deficiencies). Complement fragments also help shuttle antigen to follicular dendritic cells in lymph nodes, lower the threshold for B-cell activation via the CR2/CD19 co-receptor complex, and can lyse certain enveloped viruses and parasites directly. Dysregulated complement activation, rather than absence, is now understood to drive several diseases, including paroxysmal nocturnal hemoglobinuria, atypical hemolytic uremic syndrome, and age-related macular degeneration, motivating an entire modern drug class of complement inhibitors.