Structure-guided identification of C3d residues essential for its binding to complement receptor 2 (CD21).

A vital role for complement in adaptive humoral immunity is now beyond dispute. The crucial interaction is that between B cell and follicular dendritic cell-resident complement receptor 2 (CR2, CD21) and its Ag-associated ligands iC3b and C3dg, where the latter have been deposited as a result of classical pathway activation. Despite the obvious importance of this interaction, the location of a CR2 binding site within C3d, a proteolytic limit fragment of C3dg retaining CR2 binding activity, has not been firmly established. The recently determined x-ray structure of human C3d suggested a candidate site that was remote from the site of covalent attachment to Ag and consisted of an acidic residue-lined depression, which accordingly displays a significant electronegative surface potential. These attributes were consistent with the known ionic strength dependence of the CR2-C3d interaction and with the fact that a significant electropositive surface was apparent in a modeled structure of the C3d-binding domains of CR2. Therefore, we have performed an alanine scan of all of the residues within and immediately adjacent to the acidic pocket in C3d. By testing the mutant iC3b molecules for their ability to bind CR2, we have identified two separate clusters of residues on opposite sides of the acidic pocket, specifically E37/E39 and E160/D163/I164/E166, as being important CR2-contacting residues in C3d. Within the second cluster even single mutations cause near total loss of CR2 binding activity. Consistent with the proposed oppositely charged nature of the interface, we have also found that removal of a positive charge immediately adjacent to the acidic pocket (mutant K162A) results in a 2-fold enhancement in CR2 binding activity.

Two clusters of acidic amino acids near the NH2 terminus of complement component C4 alpha'-chain are important for C2 binding.

Previous work has indicated a role for the NH2-terminal segment of the C3 alpha'-chain in the binding interactions of C3b with a number of its protein ligands. In particular, we have identified two clusters of acidic residues, namely, E736 and E737 and to a lesser extent D730 and E731, as being important in the binding of C3b to factor B and complement receptor 1 and the binding of iC3b to complement receptor 3. Whereas human C3 and C4 have an overall sequence identity of 29%, over a segment near the NH2 termini of their respective alpha'-chains the sequence identity is 56% (70% chemical similarity). Given the functional similarity between the C4b-C2 and C3b-B interactions in the respective formation of the classical and alternative pathway C3 convertases, as well as the sequence conservation of two acidic clusters, we hypothesized that residues 744EED and 749DEDD within the NH2-terminal segment of the C4 alpha'-chain would mediate in part the binding of C2 to C4b. We tested this hypothesis using three independent approaches. Site-directed mutagenesis experiments revealed that replacing subsets of the charged residues by their isosteric amides within either acidic cluster resulted in molecules having reduced C2 binding activity. Moreover, a synthetic peptide (C4 residues 740-756) encompassing the two acidic clusters was a specific inhibitor of the binding of C2 to red cell-associated C4b. Finally, Ab raised against the above peptide was able to block the interaction between red cell-associated C4b and fluid phase C2. Taken together, these results strongly suggest that the NH2-terminal acidic residue-rich segment of C4 alpha'-chain contributes importantly to the interaction of C4b with C2.

Identification of residues within the 727-767 segment of human complement component C3 important for its interaction with factor H and with complement receptor 1 (CR1, CD35).

Mapping approaches employing blocking antibodies and synthetic peptides have implicated the 727-767 segment at the NH2 terminus of C3b alpha'-chain as contributing to the interactions with factor B, factor H, and CR1. Our previous mutagenesis study on the NH2-terminal acidic cluster of this segment identified residues Glu-736 and Glu-737 as contributing to the binding of C3b to factor B and CR1 but not factor H. We have now extended the charged residue mutagenic scan to cover the remainder of the segment (738-767) and have assessed the ability of the C3b-like C3(H2O) form of the mutant molecules to interact with factor H, CR1, and membrane cofactor protein (MCP) using a cofactor-dependent factor I cleavage assay as a surrogate binding assay. We have found that the negatively charged side chains of Glu-744 and Glu-747 are important for the interaction between C3(H2O) and factor H, a result in general agreement with an earlier synthetic peptide study (Fishelson, Z. (1991) Mol. Immunol. 28, 545-552) which implicated residues within the 744-754 segment in H binding. The interactions of the mutants with soluble CR1 (sCR1) revealed two classes of residues. The first are residues required for sCR1 to be an I cofactor for the first two cleavages of alpha-chain. These are all acidic residues and include the Glu-736/Glu-737 pair, Glu-747, and the Glu-754/Asp-755 pairing. The second class affects only the ability of sCR1 to be a cofactor for the third factor I cleavage and include Glu-744 and the Lys-757/Glu-758 pairing. The dominance of acidic residues in the loss-of-function mutants is striking and suggests that H and CR1 contribute basic residues to the interface. Additionally, although there is partial overlap, the contacts required for CR1 binding appear to extend over a wider portion of the 727-767 segment than is the case for factor H. Finally, none of the mutations had any effect on the interaction between soluble MCP and C3(H2O), indicating that despite its functional homology to H and CR1, MCP differs in its mode of binding to C3b/C3(H2O).

The covalent binding reaction of complement component C3.

The covalent binding of C3 to target molecules on the surfaces of pathogens is crucial in most complement-mediated activities. When C3 is activated, the acyl group is transferred from the sulfhydryl of the internal thioester to the hydroxyl group of the acceptor molecule; consequently, C3 is bound to the acceptor surface by an ester bond. It has been determined that the binding reaction of the B isotype of human C4 uses a two-step mechanism. Upon activation, a His residue first attacks the internal thioester to form an acyl-imidazole bond. The freed thiolate anion of the Cys residue of the thioester then acts as a base to catalyze the transfer of the acyl group from the imidazole to the hydroxyl group of the acceptor molecule. In this article, we present results which indicate that this two-step reaction mechanism also occurs in C3.

X-ray crystal structure of C3d: a C3 fragment and ligand for complement receptor 2.

Activation and covalent attachment of complement component C3 to pathogens is the key step in complement-mediated host defense. Additionally, the antigen-bound C3d fragment interacts with complement receptor 2 (CR2; also known as CD21) on B cells and thereby contributes to the initiation of an acquired humoral response. The x-ray crystal structure of human C3d solved at 2.0 angstroms resolution reveals an alpha-alpha barrel with the residues responsible for thioester formation and covalent attachment at one end and an acidic pocket at the other. The structure supports a model whereby the transition of native C3 to its functionally active state involves the disruption of a complementary domain interface and provides insight into the basis for the interaction between C3d and CR2.

Native conformations of human complement components C3 and C4 show different dependencies on thioester formation.

The thioester bond in complement components C3 and C4 and the protease inhibitor alpha2-macroglobulin have traditionally been thought of as fulfilling the dual roles of mediating covalent attachment and maintaining the native conformational states of these molecules. We previously reported that several human C3 thioester-region mutants, including variants E1012Q and C1010A, in the latter of which thioester-bond formation is precluded, display an unexpected phenotype. Despite the lack of a thioester bond in these mutants, they appear to adopt a native-like conformation as suggested by the finding that they are cleavable by the classical pathway C3 convertase, C4b2a, whereas the C3b-like C3(H2O) species is not. Subsequently, a species referred to as C3(NH3)* was described which potentially could account for the observations with the above mutants. C3(NH3)* is a transient species formed on aminolysis of native C3 that can spontaneously re-form the thioester bond. Importantly, it has a mobility on cation-exchange HPLC that is distinct from both native C3 and C3(H2O), but like the native molecule, it is cleavable by an alternative-pathway C3 convertase. In this study we showed by using cation-exchange HPLC as an additional conformational probe that C3 C1010A and E1012Q mutant proteins did not resemble C3(NH3)*. Instead they displayed a chromatographic behaviour that was indistinguishable from that of native C3. To assess the general applicability of these observations, we engineered the equivalent mutations into human C4, specifically C4 C1010A and C4 E1012Q. As expected, thioester-bond formation did not occur in either of these C4 mutants, but in contrast with the results with C3 we found no evidence for the formation of a stable native-like conformation in either C4 mutant, as assessed using cleavability by C1s as the conformational probe. A possible interpretation of our data is that the adoption of the native conformational state during biosynthesis of C3 and C4 is an energetically permissible process, even if it is not locked in via thioester-bond formation. Whereas this conformational state is stable in mature C3, it is unstable in mature C4, perhaps reflecting the additional post-translational cleavage of C4 before its secretion.

Domain-switched mouse IgM/IgG2b hybrids indicate individual roles for C mu 2, C mu 3, and C mu 4 domains in the regulation of the interaction of IgM with complement C1q.

Although polymeric IgM and monomeric IgG are potent activators of the classical complement pathway, previous studies have indicated that monomeric IgM is inactive. To understand this and to examine the roles of the individual mu domains in complement activation, we created a set of IgM/IgG2b mouse chimeric Abs in which homologous domains of both Abs have been interchanged, either singly or together with adjacent domains. The monomer subunits (H2L2) of the resulting chimeras were analyzed for their capacities to bind C1q and to initiate complement-mediated lysis (CML) of haptenated erythrocytes. When C gamma 2 was flanked by C mu 4, the inherent C1q-binding activity of the C gamma 2 domain was lost. This demonstrates that C mu 4 can suppress the C1q-binding activity of the adjacent C gamma 2 domain, and suggests that C mu 4 may exert a similar effect on the C mu 3 domain in the IgM monomer subunit. When C mu 3 was located in an IgG2b background and potentially freed from the constraints imposed by the IgM background, the monomer was not able to bind C1q or initiate CML. This suggests that these activities are not expressed inherently in the C mu 3 domain. The transplantation of C mu 3 together with C mu 4 into the IgG background permitted polymer formation. This polymer was able to bind C1q, although neither the monomer nor the polymer forms were active in CML; conversely, all IgM polymers with a transplanted C gamma 2 domain were active in both C1q binding and CML, and demonstrated apparent Kd values similar to that of wild-type IgM.

Mouse complement component C4 is devoid of classical pathway C5 convertase subunit activity.

It has long been known that mouse C4 has unusually low hemolytic activity relative to the C4 of other mammalian species (e.g. human and guinea pig), the measurements being done in most cases using a C4-deficient guinea pig serum reagent in a one-step assay with EA. This low activity for mouse C4 previously had been attributed to "technical" difficulties such as lability of the protein during blood collection and partial species incompatibilities with guinea pig components. Recently, we presented evidence for the involvement of human C4 beta-chain residues 455-469, a putatively exposed hydrophilic segment, in contributing to a C5 binding site in the C4b subunit of the classical pathway C5 convertase, C4b3b2a. Given that there were five sequence differences between the human and mouse protein within this segment, we hypothesized that these substitutions may have compromised the C5 convertase subunit activity of mouse C4, thereby resulting in its low hemolytic activity. Using a multi-step hemolytic assay which was totally dependent upon C5 cleavage by the classical pathway, we found that mouse C4 was completely devoid of classical pathway C5 convertase subunit activity. We have been able to rule out the most obvious potential species incompatibilities (e.g. between C4mo and C5gp) as being responsible for this lack of activity. Moreover, we found that the low level of hemolytic activity of mouse C4 measured in the one-step assay can be ascribed totally to C5 cleavage, and subsequent terminal component assembly, by the alternative pathway C5 convertase, (C3b)2Bb. However, the assembly of the latter enzyme complex is dependent upon the presence of C3b molecules deposited initially via the classical pathway C3 convertase in which mouse C4b is a subunit. Finally, whereas conversion of human residues 458RP to the mouse-like sequence PL was sufficient to abrogate classical pathway C5 convertase subunit activity in human C4, the five substitutions which "humanized" the 452-466 segment of mouse C4 (corresponding to human residues 455-469) were on their own insufficient to impart this activity to mouse C4. This implies that, in addition to the 455-469 beta-chain segment of human C4, there are other regions of the molecule contributing to C5 binding which are also non-conserved between human and mouse C4.

Processing of human factor I in COS-1 cells co-transfected with factor I and paired basic amino acid cleaving enzyme (PACE) cDNA.

Factor I is an active serine proteinase in plasma that regulates both the classical and alternative complement pathways by cleaving C3b and C4b thereby preventing the assembly of C3 and C5 convertase enzymes. In this study, a full-length human factor I cDNA was cloned into the pMT2 expression vector and the pMT2-fI construct was expressed transiently in COS-1 cells and stably in CHO-K1 cells. The transfected COS-1 cells secreted large amounts of recombinant pro-factor I (85 kD). Co-transfection of COS-1 cells with pMT2-fI and the cDNA expression plasmid for PACE (paired basic amino acid cleaving enzyme), resulted predominantly in the secretion of a proteolytically processed form of recombinant factor I (heavy chain, 47 kD; light chain, 35 kD). Following co-transfection of pMT2-fI and pSVNeo.1 into CHO-K1 cells and selection in medium containing G418, a stably transfected clone was isolated that secreted pro-factor I (85 kd) and proteolytically processed factor I (heavy chain, 48 kD; light chain, 37 kD) in approximately equal amounts. The molecular sizes of the subunit chains of the expressed factor I were generally slightly smaller than those of human plasma factor I. The activity of recombinant factor I present in the culture supernatants of transfected COS-1 and CHO-K1 cells was assayed by its ability to cleave 125I-C3b in the presence of factor H and was found to be low when compared with factor I purified from human plasma. However, since the functional activity of purified factor I was reduced approximately 50% in the presence of conditioned medium from non-transfected cells, it is suggested that the cold C3b present in the factor I-deficient serum used to supplement the culture medium probably competed with the 125I-C3b tracer, thereby decreasing the sensitivity of the assay for the recombinant factor I proteins.

Evidence for the involvement of arginine 462 and the flanking sequence of human C4 beta-chain in mediating C5 binding to the C4b subcomponent of the classical complement pathway C5 convertase.

Replacement of human C4 beta-chain residue arginine 458 by tryptophan, a substitution that occurs naturally in the hemolytically inactive A6 allotype of C4, totally abrogates the molecule's ability to act as a C5 binding subunit of the classical pathway C5 convertase. Hydropathy plots predict R458 to be within a hydrophilic segment extending from residue 455 to 469 and having the sequence SIERPDSRPPRVGDT. To further assess the potential involvement of this segment in the C5 binding function of C4, we have engineered "ala-scan" mutants through this segment, concentrating predominantly on charged residues, and analyzed their functional profiles. C4B isotype mutant proteins S455A (0.7), E457A (1.1), R458A (0.3), D460A (0.2), R462A (0.0), R465A (0.6), and D468A (0.3) displayed the relative to wild-type hemolytic activities indicated in the parentheses. In all cases, the hemolytic defect was accounted for solely at the C5 convertase stage. The total absence of C5 binding activity in the R462A mutant suggests a requirement for the guanidinium group per se, because mutants with a charge-conservative lysine or a relatively isosteric methionine at this position were also completely inactive. In contrast, the inactivity of the C4A6-like R458W mutant is probably caused by the intolerance of tryptophan in a hydrophilic segment, as substitution of R458 by alanine or methionine yielded recombinant molecules that retained 30% and 60% of wild-type hemolytic activity, respectively. Taken together, the mutagenesis results strongly imply that residues in the 455-469 segment contribute to the C5 binding site in C4; however, the conformational context of the segment appears to be crucial, as a synthetic peptide corresponding to this segment displayed no ability to inhibit C5 binding to surface-bound C4b.

Mutation of residues in the C3dg region of human complement component C3 corresponding to a proposed binding site for complement receptor type 2 (CR2, CD21) does not abolish binding of iC3b or C3dg to CR2.

Most evidence points toward there being a shared binding site in complement receptor type 2 (CR2, CD21) for the complement ligand C3dg and the EBV surface envelope glycoprotein gp350/220. Indeed, synthetic peptide studies have suggested that the CR2-binding sites in human C3dg and EBV gp350/220 share a similar sequence motif. The proposed CR2-binding sequence in C3dg is EDPGKQLYNVEA (residues 1199-1210 of mature C3), whereas that in EBV gp350/220 is EDPGFFNVEI (residues identical to C3dg are underlined). To further examine the role of amino acids 1199-1210 in the binding of the C3 fragments iC3b and C3dg to CR2, the following alanine-substitution variants of human C3 were tested in two independent CR2-binding assays: ED1199,1200AA; KQ1203,1204AA; L1205A; Y1206A; NV1207,1208AA; E1209A; and ED-KQ-NV1199,1200-1203,1204-1207,1208AA-AA-AA. Also engineered and tested was a chimeric C3 molecule in which the 1199-1210 sequence (PVPGGYQLTLEA) from the non-CR2-binding trout C3 molecule was grafted onto a human C3 background. Recombinant C3 proteins were expressed transiently in COS-1 cells, deposited as C3b on C3 convertase-bearing sheep erythrocytes and finally converted to cell-bound iC3b or C3dg using factors H and I. Binding of EAC423bi and EAC423dg to CR2 on Raji cells or EAC423dg to soluble CR2 was assessed. In most cases, the substitutions had little effect on CR2-binding activity and even in the case of the most highly substituted variants, the decrease in CR2-binding activity was less than twofold. Thus, contrary to the results anticipated from synthetic peptide studies, the single and multiple substitutions to the C3 sequence tested failed to corroborate a role for the 1199-1210 sequence in the C3dg-CR2 interaction.

Interactions of human complement component C3 with factor B and with complement receptors type 1 (CR1, CD35) and type 3 (CR3, CD11b/CD18) involve an acidic sequence at the N-terminus of C3 alpha'-chain.

Complement component C3 is a multifunctional protein that interacts with many different ligands and receptors. Several experimental approaches involving blocking Abs, proteolytic fragmentation, and synthetic peptides have been used to predict which regions in C3 are required for its various functions. We have used site-directed mutagenesis to alter specific residues in the C3 segments 730-739 and 933-942 that have been proposed to be required for the binding of factor B by C3, and have examined, within the context of the intact C3b molecule, the effect of these substitutions on the C3b-factor B interaction. Because it has been suggested that factor H and complement receptors type 1, 2, and 3 may recognize sites in C3 that partially or completely overlap those of factor B, the relevant proteolytic fragment of each mutant C3 was tested for its ability to interact with these molecules. This study clearly demonstrates that a segment near the N-terminus of the alpha'-chain, which contains the negatively charged residues 730DE and 736EE, is involved in the interactions of C3 proteolytic fragments with factor B and complement receptors type 1 and 3, but not type 2. Factor H cofactor activity was also partially affected by these mutations. In contrast, mutation of the 937KED triplet in the C3 933-942 segment had little or no effect on any of these activities.

C1q binding properties of monomer and polymer forms of mouse IgM mu-chain variants. Pro544Gly and Pro434Ala.

The effect of replacing proline with alanine at position 434 in the C mu 3 domain (P434A) and with glycine at position 544 in the C mu 4 domain (P544G) of the mu-chain of mouse IgM has been studied. The P434A substitution results in the loss of measurable complement-mediated cytolytic activity (CML) and a decrease in the association rate constant at low ionic strength (mu = 0.06), that results in a diminished Ka for C1q binding to P434A IgM bound to haptenated cells (0.4 x 10(9) M-1). Binding of C1(qr2s2) could not be detected. In contrast, replacement of proline at 544 had no measurable effect on the cytolytic or C1q/C1 binding properties of the polymeric molecule, supporting the view that the C mu 3 domain is important in C1q binding and CML. The secreted monomeric subunit of P544G was not able to mediate CML. Also, whereas hapten-bound P544G polymer bound C1q with a functional affinity of 1.5 x 10(9) M-1 at low ionic strength (mu = 0.06), similar to that observed with wild-type polymer (1.7 x 10(9) M-1) and wild-type IgG monomer (4.7 x 10(9) M-1), no C1q binding was detected with the P544G IgM monomer. This could not be attributed to differences in glycosylation. Inasmuch as the P544G mutation per se had no effect on the C1q binding properties of the polymer, we conclude that unlike IgG, aggregation does not sufficiently enhance the avidity of IgM monomer to enable it to activate complement. Augmentation of the site must occur during polymerization or when the IgM binds to Ag.

Mapping of amino acid residues in the C mu 3 domain of mouse IgM important in macromolecular assembly and complement-dependent cytolysis.

By analyzing the effects of single site mutations of a TNP-binding mouse IgM we have identified amino acid residues clustered in two regions in the C mu 3 domain that are important in the complement-dependent cytolytic activity of polymeric IgM. Some of the mutations also impaired IgM polymerization. For one of these clusters, D432G, P434A, and P436S, which lies on the fy2 and fy3 strands and their connecting loop, polymerization was little affected and the effect on the cytolytic activity of the polymer fraction was taken to imply direct involvement of the residue in C1 binding. The other cluster, involving residues D356A K361A and D417G, is situated at the other end of the C mu 3 domain closer to the center of the Fc mu disc. The D356A K361A and D417G mutations significantly impaired polymer formation, suggesting that these residues are necessary for proper folding or packing of the C mu 3 domains and may affect cytolysis only indirectly. Some other mutations had little or no effect on polymerization or cytolytic activity (E423A, E527G), whereas some mutations impaired only IgM polymerization without affecting cytolytic activity (D344A, K361A, K443A P544G). In others the defect in polymerization was so profound that only the monomer formed (H430A/N/Q and K438G). Our results also suggest that the C1 binding site of IgM is not strictly homologous to the C1 binding site of IgG. Although mutation of E318 of IgG has been shown to reduce its cytolytic activity, mutation of the homologous residue in IgM, E423, was without effect as were mutations of other flanking-charged residues. Proline at 436 in IgM and 331 in IgG may, however, be a common element.

Identification of a secondary Fc gamma RI binding site within a genetically engineered human IgG antibody.

Although human IgG2 is not cytophilic, we have shown previously that an IgG2 antibody expressing the sequence PLLGG (underline = substitution) spanning CH2 domain residues 233-237 (Eu numbering) displayed IgG1-like Fc gamma RI binding activity. In contrast, IgG1 PLLGG exhibited 3-fold less affinity, whereas IgG2 ELLGG was 3-fold more active than native IgG1. These results suggested that additional site(s) conferred enhanced binding properties to the engineered, cytophilic IgG2 variant. These sites were shown to reside in the IgG2 CH2 domain, since the IgG1 CH2 module did not have enhanced activity in a panel of hybrid IgG1/IgG2 antibodies. To map these sites further, human IgG1 and IgG2 constant region gene segments were modified to allow reciprocal COOH-terminal half segment exchanges of CH2 exons. These were cloned into a pSV2neo expression vector bearing a rearranged MOPC 315 heavy chain variable region gene and transfected into a MOPC 315 heavy chain deletion mutant. The dinitrophenol affinity-purified IgGs were radiolabeled and assessed for Fc gamma RI binding activity in direct binding assays using U937 cells. The COOH terminus of the IgG2 CH2 domain was found to contain accessory site(s) since it enhanced the binding properties of both IgG1 PLLGG and native IgG1. In contrast, grafting of the COOH terminus of the IgG1 CH2 domain onto IgG2 PLLGG and IgG2 ELLGG diminished their cytophilic activity. The amino acid responsible for the enhancing properties of the COOH terminus of the IgG2 CH2 domain was shown to be threonine 339, since IgG1 PLLGG/Thr339 displayed increased Fc gamma RI binding affinity. Kinetics studies revealed that this is accomplished through an increase in the forward rate constant of the IgG-Fc gamma RI interaction.

Structural requirements for thioester bond formation in human complement component C3. Reassessment of the role of thioester bond integrity on the conformation of C3.

A unique thioester bond, formed between the side chains of neighboring C and Q residues, is present in complement components C3 and C4 and the protease inhibitor alpha 2-macroglobulin. This structure is essential for mediating covalent attachment to target acceptors and also for maintaining these proteins in their native conformation. An examination of the residues in the immediate vicinity of the C and Q reveals a very high degree of sequence similarity among the three proteins which crosses species barriers. The following is the sequence flanking the thioester residues in C3, the highly conserved amino acids being underlined and the the thioester-forming residues being indicated by italics: 1005V-T-P-S-G-C-G-E-Q-N-M-I-G-M-T-P-T1021. Through a site-directed mutagenesis and cDNA expression approach, we have examined the importance of the conserved amino acids in the formation, stability, and function of the thioester bond in C3. The behavior of the mutants fell into three categories. The potential loss in peptide backbone flexibility by the replacement of G1009 by A or S was permissive to thioester formation and function as was replacement of M1015 by the still fairly bulky residue F. In contrast, replacement of M1015 by A resulted in an alpha-chain which was highly unstable toward proteolytic degradation. The third category, which included mutant molecules P1007G, P1020G, E1012Q, and Q1013N, displayed an unusual phenotype in which both the autolytic fragmentation and the hemolytic activity characteristics of thioester-intact molecules were absent. However, like their wildtype counterpart, these molecules retained the ability to be cleaved by C3 convertase (C4b2a), a conformation-dependent property that is normally lost in the conversion of native C3 to thioester-hydrolyzed C3(H2O). Since an identical functional profile was obtained when the thioester was deliberately prevented from forming in the mutant C1010A, we conclude that if a stable thioester fails to form during biosynthesis, at least parts of the mature protein can adopt a more native-like conformation than is the case when the thioester is first formed and then hydrolyzed in the mature protein. In view of these new findings, the interpretation of the previously observed correlation between the loss of thioester integrity and the adoption of a C3b-like conformation must be reassessed.

A single arginine to tryptophan interchange at beta-chain residue 458 of human complement component C4 accounts for the defect in classical pathway C5 convertase activity of allotype C4A6. Implications for the location of a C5 binding site in C4.

In general, C4A allotypes of human C4 show one-fourth to one-third the hemolytic activity of C4B allotypes. An exception to this rule is C4A6 which is almost totally deficient in hemolytic activity. Previous studies have localized the defect in C4A6 to the C5 convertase stage. Of the two critical events required for C5 cleavage, namely formation of a covalent adduct between C3b and the C4b subunit of the C3 convertase (C4b2a), and binding of C5 to this C4b-C3b complex, it is a defect in the latter step that accounts for the aberrant activity of C4A6. DNA sequencing studies described in a companion paper have suggested that the sole C4A6-specific difference was a Trp for Arg replacement at beta-chain residue 458. To directly ascertain whether this single substitution was responsible for the hemolytic defect in C4A6, we have used site-directed mutagenesis to introduce this change into both C4A and C4B cDNA expression plasmids. We found that the R to W replacement totally abrogated hemolytic activity. However, irrespective of the amino acid at residue 458, the mutant proteins behaved like their wild-type counterparts with respect to covalent binding to C1-bearing targets, i.e., the C4B recombinants displayed higher binding to sheep and human red cells than did the C4A counterparts. Furthermore, the mutants were able to form covalent C4b-C3b adducts. There was, however, substantially less C5 cleavage produced by cell-bound C4boxy23b complexes made with R458W mutant C4B than with wild-type C4B. These results are consistent with the sole defect in the mutants being at the C5 binding stage and strongly suggest that Arg 458 of the C4 beta-chain contributes to the C5 binding site of the molecule.

Covalent binding of C3b to C4b within the classical complement pathway C5 convertase. Determination of amino acid residues involved in ester linkage formation.

C5 convertase of the classical complement pathway is a protein complex consisting of C4b, C2a, and C3b. Within this complex C3b binds to C4b via an ester linkage. We now present evidence that the covalent C3b-binding site on human C4b is Ser at position 1217 of C4. We also show that formation of the covalently linked C4b.C3b complex occurs in the mouse complement system and that the C3b-binding site on mouse C4b is Ser at position 1213 which is homologous to Ser-1217 of human C4. Therefore, covalent binding of C3b to a single specific site on C4b within the classical pathway C5 convertase is likely a common phenomenon in the mammalian complement system. Specific noncovalent association of metastable C3b with C4b would occur first, leading to reaction of the thioester with a specific hydroxy group. This is supported by two lines of experimental evidence, one which shows that a mutant C4 that does not make a covalent linkage with C3b is still capable of forming C5 convertase and a second in which the C4b.C3b complex has been demonstrated by cross-linking erythrocytes bearing this C5 convertase.

Mutagenesis of the Arg-Gly-Asp triplet in human complement component C3 does not abolish binding of iC3b to the leukocyte integrin complement receptor type III (CR3, CD11b/CD18).

The leukocyte integrin complement receptor type III (CR3, CD11b/CD18) binds the C3 cleavage product iC3b. Many other integrins bind their ligands via an Arg-Gly-Asp (RGD) triplet. Both the RGD-containing C3 peptide 1390TRYRGDQDATMS1401 (pro-C3 numbering) and the RGD-like fibrinogen peptide GGAKQAGDV, which binds to the platelet integrin glycoprotein IIb-IIIa, were shown to inhibit the iC3b-CR3 interaction, suggesting that this binding is also RGD-mediated (Wright, S.D., Weitz, J.I., Huang, A. J., Levin, S.M., Silverstein, S.C., and Loike, J.D. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 7734-7738). However, unlike other integrin-ligand interactions, that of CR3 and iC3b is unaffected by the hexapeptide GRGDSP, and substitutions in the RGD triplet of C3 from other species appear to be tolerated. It was, therefore, proposed (Grossberger, D., Marcuz, A., du Pasquier, L., and Lambris, J.D. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 1323-1327) that the highly conserved DATMS portion of the inhibitory C3 peptide may have been responsible for its binding. To address these inconsistencies and directly assess the role of the 1390-1401 segment within the complete iC3b molecule in mediating binding to CR3, a human C3 cDNA was altered by site-directed mutagenesis and the expressed recombinant proteins were examined in a CR3-specific assay. Replacement of RGD by AAA did not abolish rosetting of the corresponding iC3b-coated erythrocytes to human CR3-bearing leukocytes. In addition, mutant iC3b molecules in which the positively charged R1391 (corresponding to K in the fibrinogen peptide) and the highly conserved 1397DATMS sequence were replaced by Q and NAAMA respectively, were still bound by CR3. We conclude that the iC3b-CR3 interaction is not mediated by the RGD triplet or its neighboring residues.

Identification of the Fc gamma receptor class I binding site in human IgG through the use of recombinant IgG1/IgG2 hybrid and point-mutated antibodies.

To characterize the region on human IgG1 responsible for its high-affinity interaction with the human Fc gamma receptor class I (Fc gamma RI), we have analyzed the binding properties of a series of genetically engineered chimeric antidinitrophenyl antibodies with identical murine antibody combining sites and hybrid IgG1/IgG2 human constant (C) regions. In addition, we have investigated a panel of reciprocally point-mutated IgG1 and IgG2 chimeric antibodies to identify the amino acid residues that confer cytophilic properties to human IgG1. Our data unambiguously indicate that cytophilic activity of IgG1 is an intrinsic property of its heavy-chain C region 2 (CH2) domain. We report that the entire sequence spanning residues 234-237 (LLGG) is required to restore full binding activity to IgG2 and IgG4 and that individual amino acid substitutions failed to render IgG2 active. Nevertheless, the reciprocal single point mutations in IgG1 either significantly lowered its activity or abolished it completely. Finally, we observed that an IgG2 antibody containing the entire ELLGGP sequence (residues 233-238) was more active than wild-type IgG1. This finding suggests that in addition to the primary contact site identified in the N terminus of the gamma 1 CH2 domain, secondary sites involving residues from the C-terminal half of the domain may also contribute to the IgG1-Fc gamma RI interaction.