Carbamylation of immunoglobulin abrogates activation of the classical complement pathway

Post‐translational modifications of proteins significantly affect their structure and function. The carbamylation of positively charged lysine residues to form neutral homoitrulline occurs primarily under inflammatory conditions through myeloperoxidase‐dependent cyanate (CNO^?) formation. We analyzed the pattern of human IgG₁ carbamylation under inflammatory conditions and the effects that this modification has on the ability of antibodies to trigger complement activation via the classical pathway. We found that the lysine residues of IgG₁ are rapidly modified after brief exposure to CNO^?. Interestingly, modifications were not random, but instead limited to only few lysines within the hinge area and the N‐terminal fragment of the CH2 domain. A complement activation assay combined with mass spectrometry analysis revealed a highly significant inverse correlation between carbamylation of several key lysine residues within the hinge region and N‐terminus of the CH2 domain and the proper binding of C1q to human IgG₁ followed by subsequent complement activation. This severely hindered complement‐dependent cytotoxicity of therapeutic IgG₁. The reaction can apparently occur in vivo, as we found carbamylated antibodies in synovial fluid from rheumatoid arthritis patients. Taken together, our data suggest that carbamylation has a profound impact on the complement‐activating ability of IgG₁ and reveals a pivotal role for previously uncharacterized lysine residues in this process.     

Human astrovirus coat protein binds C1q and MBL and inhibits the classical and lectin pathways of complement activation

Human astroviruses (HAstVs) constitute a family of non-enveloped, RNA viruses which cause infantile gastroenteritis. We have previously demonstrated that purified HAstV coat protein (CP), multiple copies of which compose the viral capsid, bind C1q resulting in inhibition of classical complement pathway activity. The objective of this study was to further analyze the mechanism by which CP inhibits C1 activation. CP inhibited C1 activation, preventing cleavage of C1s to its active form in the presence of heat-aggregated IgG, a potent classical pathway activator. CP also inhibited generation of the potent anaphylatoxin C5a. CP dose-dependently bound to C1q, the isolated globular heads and the collagen-like regions of the C1q molecule. When CP was added to C1, C1s dissociated from C1q suggesting that CP functionally displaces the protease tetramer (C1s–C1r–C1r–C1s). Given the structural and functional relatedness of C1q and MBL, we subsequently investigated the interactions between CP and MBL. CP bound to purified MBL and was able to inhibit mannan-mediated activation of the lectin pathway. Interestingly, CP did not bind to a variant of MBL that replaces a lysine residue (Lys55) critical for binding to MASP-2, a functional homolog of C1s. Finally, CP was shown to cross the species barrier to inhibit C3 activation and MAC formation in rat serum. These findings suggest CP inhibits C1 and MBL activation via a novel mechanism of interference with the normal interaction of the recognition molecule with its cognate serine proteases.     

Role of immunoglobulin G in killing of Borrelia burgdorferi by the classical complement pathway

The antibody and complement requirements for killing of Borrelia burgdorferi 297 by normal human serum (NHS) and NHS plus immunoglobulin G (IgG) were examined. B. burgdorferi activated both the alternative and classical complement pathways in NHS. In NHS chelated with 10 mM ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid plus 4 mM MgCl2 (Mg-EGTA) to block classical pathway activation, consumption (activation) of total hemolytic complement, complement component 3 (C3), and C9 by B. burgdorferi was observed. Furthermore, challenge of unchelated NHS with 297 cells resulted in the consumption of C4, in addition to an increase in C3 and C9 consumption over that observed in chelated serum. In spite of complement activation, B. burgdorferi was resistant to the nonspecific bactericidal activity of NHS. The addition of human anti-B. burgdorferi IgG to NHS, however, resulted in the complete killing of 297 cells. Bactericidal activity of this serum was abrogated if NHS was immunochemically depleted of C1, indicating that killing was mediated by the classical pathway. The manifestation of bactericidal activity was accompanied by a large increase in total complement and C3 consumption over that observed in NHS alone. Under similar conditions, only a minimal increase in C9 consumption was observed. No increase in total complement consumption was observed if NHS plus anti-B. burgdorferi IgG was treated with Mg-EGTA prior to challenge. The results of these experiments demonstrate that B. burgdorferi is resistant to the nonspecific bactericidal activity of NHS, in spite of classical and alternative complement pathway activation. B. burgdorferi is sensitive to serum, however, in the presence of IgG, which mediates bacterial killing through the classical complement pathway.     

IgG deposition and activation of the classical complement pathway involvement in the activation of human granulocytes by decellularized porcine heart valve tissue

Decellularization treatment of heart valves has been thought to eliminate tissue immunogenicity. Early failure of tissue-engineered xenogeneic heart valves was seen in children and has been a major drawback in this promising field of research. This study was designed to characterize the effects of acellular porcine heart valve tissue on immune activation in vitro. Incubation of decellularized porcine tissue with human plasma led to adsorption of IgG, activation of the classical complement pathway and adhesion of activated polymorphonuclear leukocytes (PMN). This inflammatory response was strongly inhibited by proteins extracted from native porcine tissue which might indicate that inhibitors of PMN activation present in the extracellular matrix (ECM) are lost during the decellularization process.     

Human neutrophil peptide-1 inhibits both the classical and the lectin pathway of complement activation

Human neutrophil peptide-1 (HNP-1) is a member of the alpha-defensin family. Defensins are cationic antimicrobial peptides, which play an important role in the antimicrobial response to microorganisms. In addition, recent studies have revealed the involvement of defensins in inflammation, immunity and wound repair. Defensins are present in the azurophilic granules of neutrophils and are released upon neutrophil stimulation. Previous studies showed that HNP-1 binds to C1q and inhibits the classical complement pathway. In view of the structural and functional similarity between C1q and MBL, we have now examined the interactions between HNP-1 and MBL. We observed a dose-dependent binding of HNP-1 to MBL in calcium-free buffer, indicating that HNP-1 binds to MBL most likely via the collagenous domains. To identify the binding sites in HNP-1 involved in the binding to C1q and MBL, we used a series of overlapping synthetic linear peptides that spanned the entire HNP-1 sequence. Both MBL and C1q showed a dose-dependent binding to the same set of peptides, suggesting a similar binding site in HNP-1 for both MBL and C1q. Strongest binding was observed to peptides containing the C- or N-terminal part of the HNP-1 molecule. Using an ELISA based system, we demonstrated that HNP-1 inhibits activation of both the classical pathway and lectin pathway of complement. Furthermore, we demonstrated that C1q and MBL can form complexes with HNP-1 in solution. Together, the data indicate that HNP-1 interacts with both C1q and MBL efficiently resulting in inhibition of both the classical and the lectin pathway of complement. We conclude that HNP-1 may play a role in protection against tissue injury during inflammatory conditions by inhibiting the early phase of complement activation.     

Control of the classical and the MBL pathway of complement activation

The activation of complement via the mannan-binding lectin (MBL) pathway is initiated by the MBL complex consisting of the carbohydrate binding molecule, MBL, two associated serine proteases, MASP-1 and MASP-2, and a third protein, MAp19. In the present report we used an assay of complement activation specifically reflecting the physiological activity of the MBL complex to identify biological and synthetic inhibitors. Inhibitor activity towards the MBL complex was compared to the inhibition of the classical pathway C1 complex and to a complex of MBL and recombinant MASP-2. A number of synthetic inhibitors were found to differ in their activities towards complement activation via the MBL pathway and the classical pathway. C1 inhibitor inhibited both pathways whereas alpha2-macroglobulin (alpha2M) inhibited neither. C1 inhibitor and alpha2M were found to be associated with the MBL complex. Upon incubation at 37 degrees C in physiological buffer, the associated inhibitors as well as MASP-1, MASP-2, and MAp19 dissociated from MBL, whereas only little dissociation of the complex occurred in buffer with high ionic strength (1 M NaCl). The difference in sensitivity to various inhibitors and the influence of high ionic strength on the complexes indicate that the activation and control of the MBL pathway differ from that of the classical pathway. MBL deficiency is linked to various clinical manifestations such as recurrent infections, severe diarrhoea, and recurrent miscarriage. On the other hand, impaired control of complement activation may lead to severe and often chronically disabling diseases. The results in the present report suggests the possibility of specifically inhibiting of the MBL pathway of complement activation.     

BrkA protein of Bordetella pertussis inhibits the classical pathway of complement after C1 deposition

Bordetella pertussis produces a 73-kDa protein, BrkA (Bordetella resistance to killing), which inhibits the bactericidal activity of complement. In this study we characterized the step in the complement cascade where BrkA acts, using three strains: a wild-type strain, a strain containing an insertional disruption of brkA, and a strain containing two copies of the brkA locus. Following incubation with 10% human serum, killing was greatest for the BrkA mutant, followed by that for the wild-type strain, while the strain with two copies of brkA was the most resistant. Complement activation was monitored by enzyme-linked immunosorbent assay (ELISA) or Western blotting. ELISAs for SC5b-9, the soluble membrane attack complex, showed that production of SC5b-9 was greatest with the brkA mutant, less with the wild type, and least with the strain containing two copies of brkA. Deposition of complement proteins on the bacteria was monitored by Western blotting. A decrease in deposition on the bacteria of C4, C3, and C9 corresponded with decreased complement sensitivity. Deposition of C1, however, was not affected by the presence of BrkA. These studies show that BrkA inhibits the classical pathway of complement activation and prevents accumulation of deposited C4.     

Spontaneous classical pathway activation and deficiency of membrane regulators render human neurons susceptible to complement lysis

This study investigated the capacity of neurons and astrocytes to spontaneously activate the complement system and control activation by expressing complement regulators. Human fetal neurons spontaneously activated complement through the classical pathway in normal and immunoglobulin-deficient serum and C1q binding was noted on neurons but not on astrocytes. A strong staining for C4, C3b, iC3b neoepitope and C9 neoepitope was also found on neurons. More than 40% of human fetal neurons were lysed when exposed to normal human serum in the presence of a CD59-blocking antibody, whereas astrocytes were unaffected. Significant reduction in neuronal cell lysis was observed after the addition of soluble complement receptor 1 at 10 microg/ml. Fetal neurons were stained for CD59 and CD46 and were negative for CD55 and CD35. In contrast, fetal astrocytes were strongly stained for CD59, CD46, CD55, and were negative for CD35. This study demonstrates that human fetal neurons activate spontaneously the classical pathway of complement in an antibody-independent manner to assemble the cytolytic membrane attack complex on their membranes, whereas astrocytes are unaffected. One reason for the susceptibility of neurons to complement-mediated damage in vivo may reside in their poor capacity to control complement activation.     

An amphioxus gC1q protein binds human IgG and initiates the classical pathway: Implications for a C1q‐mediated complement system in the basal chordate

The origin of the classical complement pathway remains open during chordate evolution. A C1q‐like member, BjC1q, was identified in the basal chordate amphioxus. It is predominantly expressed in the hepatic caecum, hindgut, and notochord, and is significantly upregulated following challenge with bacteria or lipoteichoic acid and LPS. Recombinant BjC1q and its globular head domain specifically interact with lipoteichoic acid and LPS, but BjC1q displays little lectin activity. Moreover, rBjC1q can assemble to form the high molecular weight oligomers necessary for binding to proteases C1r/C1s and for complement activation, and binds human C1r/C1s/mannan‐binding lectin‐associated serine protease‐2 as well as amphioxus serine proteases involved in the cleavage of C4/C2, and C3 activation. Importantly, rBjC1q binds with human IgG as well as an amphioxus Ig domain containing protein, resulting in the activation of the classical complement pathway. This is the first report showing that a C1q‐like protein in invertebrates is able to initiate classical pathway, raising the possibility that amphioxus possesses a C1q‐mediated complement system. It also suggests a new scenario for the emergence of the classical complement pathway, in contrast to the proposal that the lectin pathway evolved into the classical pathway.     

On the site of C4 deposition upon complement activation via the mannan-binding lectin pathway or the classical pathway

The mannan-binding lectin (MBL) pathway and the classical pathway of complement activation are initiated by the binding of the recognition structure of the initiator complexes, MBL and C1q, respectively, to their ligands, i.e. carbohydrate structures or immune complexes. Proenzymes associated with MBL or C1q are then activated and generate C3 convertase through the activation of C4 and C2. The cleavage product of C4, C4b, attaches covalently to nearby hydroxyl or amino groups. The current picture is that C2 must then attach to C4b before being cleaved by the same associated proteases into the enzymatically active fragment, C2b. This suggests a stringent requirement for the deposition of C4b very close to the initiator complex, or indeed onto the initiator complex. We examined the possibility of C4b being bound to the initiator complex by a solid-phase assay, allowing for the selective elution of the initiator complexes, followed by quantification of the C4b being eluted and the C4b remaining on the solid phase. Also, we estimated the generation of complexes between the released initiator complex and C4b. More than 99% of deposited C4b was bound directly to the solid phase rather than to the initiator complex. Our approach cannot answer the question of the whereabouts of the C2 when it is cleaved.     

Lack of binding of C3 to IgG antibodies during the activation of the classical complement pathway on the red cell

We have studied the interaction between C3 and natural human and rabbit anti-MTX IgM and hyperimmune IgG antibody bound to red cells to which MTX was covalently coupled. IgM Ab molecules bound to the cell surface were measured by their interaction with rabbit anti-human mu-chain IgG Ab: the bound anti-mu-chain or anti-MTX IgG was quantitated with 125I-labelled PA. C3 uptake by EMTX AC142 complexes from purified preparation of C3 was detected by the interaction of bound C3 with rabbit anti-C3 IgG Ab; the bound IgG was then measured with radiolabelled PA. In some experiments the uptake of C3 by the EMTX AC142 complexes was measured by using 125I-labelled C3. To find out if C3 was bound to Ab molecules, anti-MTX Abs were eluted from the cells by excess fluid-phase MTX. After dialysis the eluted anti-MTX IgM or IgG Ab was than reattached to fresh EMTX. All cells were then analyzed for Ab and C3 content by a radioimmunoassay. It was found that when anti-MTX IgM or IgG was eluted from EMTX AC423 complexes no C3 was removed from the cells and that eluted IgM or IgG did not carry with them either C3 antigen or 125I-labelled C3. It was concluded that, during the activation of the classical C-pathway at the red cell surface, no C3 was bound to IgM (rabbit or human) or IgG (rabbit) antibody molecules.     

Binding of human C4 to C-reactive protein-pneumococcal C-polysaccharide complexes during activation of the classical complement pathway

Sequential interaction of CRP-PnC aggregates, made at slight CRP excess, with purified human C1, C4 and C2oxy resulted in formation of an effective C3-convertase, indicating the binding of C1, C4 and C2 on the aggregates. Immunoprecipitation experiments demonstrated that, following cleavage of 125I-C4 by CRP-PnC-C1 complexes, approximately 3% of the 125I-C4 was bound to CRP while a lower percentage was bound to PnC, CRP-C4 complexes could also be demonstrated by substituting 125I-CRP for 125I-C4. The nature of the CRP-C4 bond was examined by electrophoretic analysis. Complexes of 125I-C4-CRP prepared as earlier were incubated at 100 degrees C for 2 min in buffer containing 2% SDS and 5% beta-mercaptoethanol and subjected to electrophoresis in SDS-containing polyacrylamide gradient slab gels. Autoradiography of the dried gels revealed the presence of high mol. wt bands containing the alpha'-chain of C4b. CRP could also be demonstrated in these high mol. wt bands which apparently represented covalent complexes between the alpha'-chain of C4b and CRP monomers. Since CRP contains no detectable carbohydrate, it seems likely that an amide bond is formed between the two proteins.     

Consumption of C4b-binding protein (C4BP) during in vivo activation of the classical complement pathway.

C4BP has a central role in regulating the classical complement (C') pathway, but it is still uncertain whether or not it is consumed during in vivo complement activation. Attempts to demonstrate changes in C4BP plasma levels in systemic lupus erythematosus and essential mixed cryoglobulinaemia have failed, probably due to up-regulation of this protein during the inflammatory reaction. We have studied one patient with severe post-transfusion complement-mediated anaphylaxis (CMA), and 67 patients with hereditary C1 inhibitor deficiency (hereditary angioedema (HAE)). The first of these two conditions is characterized by the absence of systemic inflammatory reaction and the second by acute and chronic activation of the C' classical pathway. C4BP, C4BP-C4b complex, and soluble terminal C' complex (sC5b-9) were measured in the patients' plasmas by ELISA techniques and C3a and C4a by radioimmunoassays. In CMA, 15 min after the transfusion, there was a massive C' activation, with increases in C4a, C3a, sC5b-9, C4BP-C4b complexes and decreases in C4, C3 and C4BP. All parameters reverted to preinfusion values within 24 h. Depletion of C4 was correlated with that of C4BP. In patients with HAE, the median value of C4BP (83% range 54-165) was significantly lower (P < 0.0001) than in normal controls (99% range 70-159), with no difference between patients in remission or during acute attacks. C4BP-C4b complexes could not be detected in HAE patients. The results of this study indicate that C4BP is consumed in vivo during acute, and possibly during chronic activation of the C' classical pathway, and that this protein, after interaction with C4b, not longer circulates in plasma.     

Escherichia coli do not express Fc-receptors for human immunoglobulin G (IgG)

Gram-positive bacterial pathogens express immunoglobulin (Ig) binding proteins that perturb Fc-dependent functions such as the interaction with complement or phagocytic Fc-receptors. The possession of such molecules by gram-negative bacteria, including Escherichia coli (E. coli), has also been documented. In many such studies, the detection of Ig binding has relied on the use of polyclonal antibodies as detecting reagents. These are not ideal since such preparations may be contaminated with E. coli specific IgG, allowing for the potential misinterpretation of specific F(ab')(2) binding as non-specific Fc mediated binding. Here we use mono-specific recombinant antibodies to develop a novel assay for Ig binding non reliant on traditional polyclonal antibodies, allowing us to demonstrate the unequivocal absence of Fc binding proteins from E. coli.     

Evidence of IgG-mediated enhancement of the antibody response in vivo without complement activation via the classical pathway

The complement (C) dependency of IgG-mediated enhancement of the antibody response was investigated by immunizing mice with trinitrophenyl-coupled keyhole limpet hemocyanin (TNP-KLH) and either a C-activating TNP-specific monoclonal IgG2a antibody (Hy-1.2) or a mutant, non-C-activating variant of Hy-1.2 (M12). Hy-1.2 as well as M12 efficiently enhanced the anti-KLH response, although Hy-1.2 was more active. In addition, also a naturally non-C-activating TNP-specific IgG1 antibody enhanced the response to TNP-coupled bovine serum albumin. Moreover, C-activating IgG could enhance the antibody response in mice depleted of C3 by treatment with cobra venom factor. These findings suggest that the classical pathway of C activation is not required for IgG-mediated enhancement.     

Staphylococcal superantigen-like protein 3 binds to the Toll-like receptor 2 extracellular domain and inhibits cytokine production induced by Staphylococcus aureus, cell wall component, or lipopeptides in murine macrophages

Staphylococcal superantigen-like proteins (SSLs) are a family of exoproteins sharing structural similarity with superantigens, but no superantigenic activity. Corresponding host target proteins or receptors against a portion of SSLs in the family have been identified. In this study, we show that SSL3 specifically binds to Toll-like receptor 2 (TLR2) and inhibits the stimulation of macrophages by TLR2 ligands. An approximately 100-kDa protein was recovered by using recombinant His-tagged SSL3-conjugated Sepharose from the lysate of porcine spleen, and the protein was identified as porcine TLR2 by peptide mass fingerprinting analysis. The SSL3-conjugated Sepharose recovered human and mouse TLR2 but not TLR4 from human neutrophils and mouse macrophage RAW 264.7 cells, as well as a recombinant TLR2 extracellular domain chimera protein. The production levels of interleukin 12 (IL-12) from mouse macrophages treated with heat-killed Staphylococcus aureus and of tumor necrosis factor alpha (TNF-α) from RAW 264.7 cells induced by peptidoglycan or lipopeptide TLR2 ligands were strongly suppressed in the presence of SSL3. The mutation of consensus sialic acid-containing glycan-binding residues in SSL3 did not abrogate the binding ability to TLR2 or inhibitory activity on TLR2, indicating that the interaction of SSL3 with TLR2 was independent of the sialic acid-containing glycan-binding residues. These findings demonstrate that SSL3 is able to bind the extracellular domain of TLR2 and interfere with TLR2 function. The present study provides a novel mechanism of SSL3 in immune evasion of S. aureus via interfering with its recognition by innate immune cells.     

Killing of the S and Re forms of Salmonella minnesota via the classical pathway of complement activation in guinea-pig and human sera.

The S (wildtype) and Re form (heptose-deficient, core-defective mutant) of Salmonella minnesota were killed by treatment with normal guinea-pig serum (GPS). Using C4-deficient GPS and serum containing 0.02 M ethyleneglycol-bis-(beta-aminoethylether)-tetraacetic acid and 0.02 M MgCl2 (EGTA-Mg2+) a reduced killing rate was observed. In normal GPS diluted 1:10 containing 0.02 M EGTA-Mg2+ or in C4-deficient GPS diluted 1:10 no killing occurred, whereas the same serum dilution without EGTA-Mg2+ showed a strong bactericidal effect indicating a dependency upon C4 and Ca2+ ions. Furthermore, in contrast to normal human serum (NHS) no killing occurred in a selective complete C1q-deficient human serum. The bactericidal effect, however, could be restored by addition of highly purified C1q; this is a further indication for a dependency upon the classical pathway of C activation. The C-dependent bactericidal activity was totally abolished when phosphate buffer was used, partially reduced in the presence of veronal-buffered saline (VBS), and not affected by tris-(hydroxymethyl)-aminomethane(Tris) or thioglycollate-buffered system EGTA-Mg2+ alone slightly reduced the growth rate of the bacteria whereas disodium ethylene diaminetetraacetate (EDTA) had a bacteriostatic effect on the S-form. The inhibition of the growth of the Re-form by EDTA was amplified by the addition of serum. Pre-incubation of bacteria with serum for absorption of antibodies did not increase the killing rate of such pre-treated bacteria excluding an antibody-mediated bactericidal reaction. Furthermore, pre-treatment of the bacteria with GPS at 0 degrees reduced the serum sensitivity of both types of bacteria.     

Opsonization of four Bacteroides species: role of the classical complement pathway and immunoglobulin

Previous investigators have suggested that opsonization of two Bacteroides species is mediated exclusively by the alternative complement pathway and requires immunoglobulins. In this study, the nature of the opsonic factors in nonimmune human serum for four species of Bacteroides was investigated by measuring uptake of [(3)H]thymidine-labeled bacteria by human polymorphonuclear leukocytes. Normal human serum, C2-deficient serum, immunoglobulin-deficient serum, and serum chelated with ethylene glycol-bis(beta-aminoethyl ether)-N,N-tetraacetic acid (EGTA), MgEGTA, and ethylenediaminetetraacetic acid (EDTA) were used as opsonic sources. Heat inactivation of each of these sera significantly reduced its opsonic activity for all four Bacteroides species, suggesting that serum complement was essential for effective opsonization. All strains were opsonized in the absence of the classical complement pathway; however, kinetics studies revealed that opsonization proceeded at a significantly faster rate when the classical complement pathway was intact. Although two strains were opsonized in immunoglobulin-deficient sera, opsonization was less efficient and appeared to occur via the alternative complement pathway. Unexpectedly, all strains were well opsonized by the classical complement pathway in 10% serum which had been effectively chelated with EGTA or EDTA. The explanation for this finding is unknown; however, it is possible that cell wall cations of Bacteroides species may participate in the activation of complement in chelated serum, resulting in effective opsonization. It was also found that Bacteroides, when incubated with an Escherichia coli strain in normal serum, could compete for opsonins and thereby reduce phagocytosis of E. coli. It is possible that competition for opsonins among bacterial species contributes to the synergistic role these organisms share in mixed floral infections.