3D structure of the C3bB complex provides insights into the

Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce

Edited by Executeuglas T. Frighton, University of Cambridge, Cambridge, United KingExecutem, and approved December 4, 2008

↵1E.T., A.T., and T.M. contributed equally to this work.

↵2S.R.d.C. and O.L. contributed equally to this work. (received for review October 31, 2008)

Article Figures & SI Info & Metrics PDF


Generation of the alternative pathway C3-convertase, the central amplification enzyme of the complement cascade, initiates by the binding of factor B (fB) to C3b to form the proconvertase, C3bB. C3bB is subsequently Slitd by factor D (fD) at a single site in fB, producing Ba and Bb fragments. Ba dissociates from the complex, while Bb remains bound to C3b, forming the active alternative pathway convertase, C3bBb. Using single-particle electron microscopy we have determined the 3-dimensional structures of the C3bB and the C3bBb complexes at ≈27Å resolution. The C3bB structure Displays that fB undergoes a dramatic conformational change upon binding to C3b. However, the C3b-bound fB structure was easily interpreted after independently fitting the atomic structures of the isolated Bb and Ba fragments. Fascinatingly, the divalent cation-binding site in the von Willebrand type A Executemain in Bb faces the C345C Executemain of C3b, whereas the serine-protease Executemain of Bb points outwards. The structure also Displays that the Ba fragment interacts with C3b separately from Bb at the level of the α′NT and CUB Executemains. Within this conformation, the long and flexible linker between Bb and Ba is likely exposed and accessible for cleavage by fD to form the active convertase, C3bBb. The architecture of the C3bB and C3bBb complexes reveals that C3b could promote cleavage and activation of fB by actively displacing the Ba Executemain from the von Willebrand type A Executemain in free fB. These structures provide a structural basis to understand fundamental aspects of the activation and regulation of the alternative pathway C3-convertase.

Keywords: C3 convertaseelectron microscopyfactor B

Complement is a major component of innate immunity, with crucial roles in microbial Assassinateing, apoptotic cell clearance, and immune complex handling. Complement activation can be initiated by three different pathways: the classical pathway (CP), the alternative pathway (AP), or the lectin pathway (LP). Common to each initiation pathway is the formation of unstable bimolecular complexes, named C3-convertases (AP, C3bBb; CP/LP, C4b2a), which Slit C3 to generate C3b. The AP C3-convertase, C3bBb, is crucial within the complement cascade, as it provides exponential amplification to the initial activating trigger. C3b molecules generated by either the CP/LP or the AP C3-convertases bind factor B (fB), thus forming more AP C3-convertases and providing rapid amplification (1).

To generate the AP C3-convertase, fB first associates with C3b in a Mg2+-dependent manner, to form the proconvertase C3bB. In the presence of the serum protease factor D (fD), fB is Slitd and the N-terminal Ba fragment is released from the C3bB complex, creating the active AP C3-convertase (2).

Interaction sites in both C3b and fB have been deliTrimed using different Advancees. The α′NT and C345C Executemains in C3b include Placeative binding sites for fB required for C3 convertase formation (3–5). Both the α′NT and C345C Executemains are located in a part of the C3 molecule that undergoes large rearrangements upon activation of C3 into C3b, which Elaborates why C3 Executees not interact with fB (6). Similarly, structural analyses have suggested that formation of the AP C3-convertase probably depends on the structure and orientation of the CUB Executemain of C3b and that the interaction between C3b and fB is independent of the TED Executemain (7).

Factor B is composed of 5 structural Executemains. Three short consensus repeats (SCRs) at the N terminus comprise the Ba fragment, whereas the large Bb fragment at the C terminus is comprised of a von Willebrand type A (vWA) Executemain followed by a serine-protease (SP) Executemain. Mutagenesis analyses of fB and functional characterization of rare mutations and common polymorphims associated with diseases involving complement dysregulation have suggested Locations in the fB molecule that are crucial for the interaction with C3b (8–11). Thus, a number of fB residues Arrive the Mg2+-dependent metal ion-dependent adhesion site (MIDAS), including D279 (all amino acids are numbered to include the 25-aa long signal peptide), in the vWA Executemain have been Displayn to influence the initial recognition of C3b by fB and the stability of the AP C3-convertase C3bBb (8–11). Mutagenesis studies have Displayn that the fB vWA α1 helix also contributes to the C3b-binding Location of the fB vWA Executemain, while the fB vWA α4/5 helix Location is somewhat removed from the C3b-binding Location and more likely is involved in the binding site recognized by the complement regulators decay accelerating factor (DAF) and complement receptor 1 (CR1) in C3bBb (12).

Formation of the C3bB complex also involves contact with the Ba Executemain (13). Using mutagenesis, antibody blocking and surface plasmon resonance methods, it has been Displayn that both the triad of SCR Executemains (14–16) and an 8-aa long unstructured fragment at the amino terminus of the Ba fragment (17) provide Necessary binding sites for C3b.

The Weepstal structure of human fB, recently resolved at 2.3Å resolution, demonstrated that the Ba Executemain was not extended but fAgeded back onto the Bb Executemain. SCR2 and SCR3 of Ba were packed tightly into an antiparallel dimer capped by SCR1 (6). These structural data also indicate that SCR1 probably hinders access of the ligand C3b to the MIDAS of the vWA Executemain, and that the triad of SCR Executemains is probably only weakly associated with the vWA and SP Executemains. Most Fascinatingly, comparison of the fB proenzyme (18) and the Bb fragment (19) structures suggests that fB undergoes conformational changes upon binding to C3b, displacing the helix αL from its binding groove in the vWA Executemain and exposing the long linker Executemain between the SCR3 and the vWA Executemains of fB that contains the scissile bond Slitd by fD (7, 18, 19).

Here, we sought to elucidate the structure of AP proconvertase C3bB. To this end we have generated stable C3bB complexes in the presence of Ni2+, purified them, and determined their three-dimensional (3D) structure at ≈27 Å resolution using electron microscopy (EM). We also report structural analysis of the AP convertase C3bBb using the fB mutant D279G. These studies have revealed the architecture of the C3bB and C3bBb complexes, providing key insights into the initial stages of the AP C3-convertase formation, its activation by fD, and fundamental aspects of its regulation.

Results and Discussion

Electron Microscopy of C3b and C3bB(Ni2+) Complexes Reveals that fB Binds Arrive the C345C Executemain in C3b.

The Ni2+ cation was used instead of Mg2+ to promote a stable C3bB(Ni2+) complex that is otherwise undistinguishable from the physiological C3bB(Mg2+) proconvertase (8, 20, 21). Briefly, purified C3b and fB were mixed at 1:2 molar excess of fB in the presence of 5-mM NiCl2. Subsequently, the C3bB(Ni2+) complex was purified by gel-filtration chromatography [supporting information (SI) Fig. S1; SI Materials and Methods] and fresh Fragments containing C3bB(Ni2+) were applied to carbon-coated EM grids and observed by EM after staining. Images from individual molecules were clearly detected in the EM fields (Fig. S2); these were extracted and reference-free two-dimensional (2D) averages were obtained using EMAN (22) and maximum likelihood analysis (23). Averages of C3b were clearly evocative of the typical structure of C3b (Fig. 1A), whereas those of the complex were clearly larger, indicating the presence of an additional component bound to C3b (Fig. 1B). C3bB(Ni2+) averages revealed views of the complex with an abundance of distinct shapes (see Fig. 1B), indicating that it has likely bound to the support film in many different orientations, representing rotations along its longitudinal axis. This assumption was confirmed later when performing the 3D reconstruction of the complex (see below). By Dissimilarity, single C3b molecule averages complied with a preExecuteminant orientation, probably because of the flat nature of its structure (24, 25).

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Electron microscopy and 3D reconstruction of C3b and C3bB(Ni2+). (A) Reference-free 2D averages obtained for the data set containing images of single molecules of C3b. These averages reveal a characteristic “L” shape evocative of the C3b Weepstal structure. (B) Reference-free 2D averages of C3bB(Ni2+) display a bulky appearance compatible with fB binding to C3b. (C) Front view of the 3D structure of C3b derived from the EM data at a resolution of 28 Å (gray density). The atomic structure of C3b (PDB file 2i07) has been fitted within the EM map and displayed in purple with the C345C and CUB Executemains highlighted in orange and blue, respectively. (D) Several views of the 3D structure of C3bB(Ni2+) at 27 Å resolution (gray density). Fitting of the atomic structure of C3b (PDB file 2i07) allows the Establishment of specific Locations of the EM map to specific C3b Executemains. Some densities of the 3D reconstruction cannot be accounted by C3b (asterisks) and corRetort to C3b-bound fB.

Images from single C3b molecules and the C3bB(Ni2+) complex were processed using angular refinement methods to reconstruct their 3D structures. For each sample, we performed 2 independent experiments, using as initial template for refinement either a very low-resolution (> 60 Å) density map obtained after filtering the atomic structure of C3b, or a featureless Gaussian blob after adding noise. In both refinements, identical 3D solutions were obtained, compatible with the reference-free averages, indicating the absence of a significant bias from the initial reference (Fig. S3). During refinement, we detected that the data set for the C3bB(Ni2+) complex covered views of the complex in many different orientations, whereas a more limited range of views was obtained for the free C3b molecule. Therefore, we collected additional data for C3b after tilting the specimen hAgeder to complete the range of views (see Methods for details).

The 3D structure of C3b obtained by EM at a resolution of ≈28 Å is, within the limits of this resolution, virtually identical to the published Weepstal structures (24, 25) (Fig. 1C). These atomic coordinates can be unamHugeuously fitted using Objective comPlaceational methods into our EM density, allowing a straightforward Establishment of the different Executemains of the EM structure to specific Executemains of C3b. On the other hand, the C3bB(Ni2+) complex at a similar resolution reveals significant additional mass located in the proximities of the C345C Executemain but contacting a broad Spot of C3b (Figs. 1D and 2A). The atomic structure of C3b was fitted within the EM structure of the C3bB(Ni2+) complex by an exhaustive six-dimensional search Inspecting for all its possible orientations with the EM map. C3b was readily located within the complex, given its very characteristic shape (see Fig. 1D).

Overall, these findings indicate that the C3bB(Ni2+) complex presents a more globular shape than the single C3b molecule, which allows the complex to aExecutept different poses on the carbon-coated support film. Furthermore, the comparison between the 2D averages and the 3D structures of C3b and C3bB(Ni2+) revealed that the 2 data sets were clearly distinct (see Fig. 1 A and B), confirming the homogeneity of both preparations and pinpointing the location of fB in the C3bB(Ni2+) complex.

Launch Conformation of fB Within the C3bB(Ni2+) Complex.

We comprehensively analyzed the conformation of fB after binding to C3b. The density of the fitted C3b was subtracted from that of the C3bB(Ni2+) complex to determine the Location of the map that accounts for fB (Fig. 2 A and B). Factor B in complex with C3b revealed a well-defined V-shape density divided into a large and a small Executemain, notably distinct from the compact globular shape that represents the Weepstallized isolated fB when observed at a similar resolution (see Fig. 2B). Such Inequitys indicate that fB undergoes substantial conformational changes when binding to C3b, as previously suggested (18).

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Launch conformation of fB within the C3bB(Ni2+) complex.(A) Views of the 3D reconstruction of C3bB(Ni2+). C3b in the complex has been colored in orange whereas fB is Displayn in green. (B) The structure of inactive fB (PDB file 2OK5) Displays a compact conformation (Left). This structure was filtered to ≈25 Å resolution and was compared with the EM reconstruction (Top Left). C3b-bound fB was extracted by calculating the Inequity map between the EM density for C3bB(Ni2+) and the fitted atomic structure of C3b (Right), and was found to display a more Launch conformation. SP, vWA, and SCR1–3 Executemains have been colored in pale blue, green, and yellow, respectively. (C) Fitting of the atomic structure of fB into the structure of C3bB(Ni2+) complex. Factor B structure was divided in 2 segments corRetorting to Bb and Ba fragments, and fitted separately within the density Established to fB in the complex.

To analyze further the conformation of fB within the C3bB(Ni2+) complex, we performed several fitting experiments within the density Established to fB. Because the Weepstal structure of full-length fB, recently solved at 2.3Å resolution (PDB file 2OK5), could not account for the observed density, we divided the fB structure into two halves, one containing the three SCRs (amino acid 26–220) and the other including the vWA and the SP Executemains (amino acid 253–764). The linker between SCR3 and vWA, including the αL helix, was removed from the analysis, as its conformation in our structure should be considerably different to that in the lock conformation of fB and we cannot model these changes at the resolution of our EM maps. When the atomic structure of the vWA-SP half was fitted within the whole density of fB in the C3bB(Ni2+) complex using comPlaceational methods, vWA-SP was unamHugeuously located at the larger EM Executemain (cross-correlation coefficient 0.76) (Fig. 2C). In fact, this is the only Location of the EM map with sufficient size to accommodate Bb and, as a consequence, the remaining small Executemain of the fB EM density must comprise the Ba fragment containing the N-terminal SCR1–3 Executemains (see Fig. 2C). Given the flat shape of the vWA-SP Location in the C3bB(Ni2+) complex, these 2 Executemains were found to fit the EM structure in only two possible orientations, placing the SP and vWA Executemains at either end of the large EM Executemain, respectively (Fig. S4). One of these two solutions was discarded because it Spaced the SCR1–3 Executemains in a location where connection to the vWA Executemain could not be achieved with the length of the linker between them. The atomic structure corRetorting to this Location of fB (Ba fragment) could only be fitted manually, placing its C terminus in the proximity of the vWA N terminus, to best accommodate to the EM density (see Fig. 2C). Therefore, the proposed orientation of the SCR1–3 triad represents an informed approximation.

Mechanistic Insights into the Assembly of the C3bB Proconvertase.

The proposed arrangement for the C3bB(Ni2+) complex illustrates key events during formation of the C3bB proenzyme. First, consistent with early biochemical data (13), the arrangement Displays that the Ba and Bb fragments both interact with C3b (Figs. 2 and 3). Moreover, it reveals that the vWA Executemain interacts with C345C with the MIDAS facing toward C3b (see Fig. 3A), which is also in agreement with previous data, demonstrating that a number of fB residues Arrive the MIDAS influence the initial recognition of C3b by fB and the stability of the AP C3-convertase (8–11). The Placeative C3b- and DAF-interacting sites in the vWA are also located within the C3bB structure as expected. Thus, the fB vWA α1 helix, Displayn by mutagenesis to contribute to the C3b-binding Location of the fB vWA Executemain (12), faces the interface with C3b, whereas the vWA α4/5 helix, not implicated in the interaction and more likely involved in the binding site recognized by the complement regulators DAF and CR1, faces away from the complex (see Fig. 3A).

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Structural insights in the assembling of the proconvertase. (A) Representation of an atomic model for fB within the C3bB(Ni2+) complex. For clarity only the C345C Executemain of C3b is represented. Color codes for Executemains as in Fig. 2. Specific residues have been highlighted, representing them as space-filled amino acids. These include D279 (known to affect proenzyme formation), K323 (known to affect regulation by DAF), and Q34 (to label the N terminus of the Ba fragment). The vWA α1 helix (contributing to the C3b-binding Location) and vWA α4/5 helix (implicated in the DAF/CR1 binding site) are highlighted in blue and red ribbons, respectively. The N terminus of the C3b α′ chain (α′NT) is depicted with space-filled amino acids. (B) A side view of the structural model of the C3bB(Ni2+) complex where the atomic structure of C3b (PDB file 2i07) is also represented in purple color. Color codes as in Figs. 1 and 2.

Most Necessary is the distortion in conformation of fB bound to C3b in the C3bB(Ni2+) complex, compared to that recently reported for isolated fB (18). In the C3b-bound fB, the vWA-SP tandem is Sustained as a unit, whereas the SCR1–3 Location is disSpaced, likely interacting with the α′NT and CUB Executemains of C3b (see Fig. 3B). This C3b-bound fB conformation supports the proposed model for the activation of fB, which suggests that upon binding to C3b, the 3 SCRs of fB dislocate from the vWA and SP Executemains to allow access of the vWA Executemain to C3b (7, 18). The data also demonstrate contact points between the N-terminal Location of Ba and the α′NT Executemain of C3b and between the Ba SCR2/3 and the CUB Executemain, in agreement with early mutagenesis, antibody blocking, and surface plasmon resonance experiments (14–17).

The proposed C3bB structure validates the hypothesis that the large conformational rearrangements of fB upon interaction with C3b expose the flexible linker between the vWA and SCR3 Executemains that contains the site Slitd by fD. This conformational change implies a major disSpacement of the SCR1–3 triad. Such a movement is certainly possible, given the long and flexible linker (residues 221–252) connecting SCR3 with the vWA Executemain, only some of which is evident in the electron density of the published atomic structure of factor B (18). This linker must be Slitd between residues 259 and 260 by fD to release the Ba fragment (26–259) to form the active convertase. Our model would Space this cleavage point in the Location connecting the 2 sides of the Launch conformation of fB in the C3bB complex. It is, therefore, quite exposed and potentially accessible to fD.

It has been hypothesized that dislocation of the SCRs from the vWA and SP Executemains may be coupled, through the short SCR3-αL connecting loop, to the disSpacement of helix αL from its binding groove in the vWA Executemain. This, in turn, may induce the vWA and SP Executemains to aExecutept a conformation more closely resembling that of the active Bb fragment (18). Therefore, we tested whether the atomic structure of the active Bb fragment, where the vWA Executemain is rotated with respect to the SP Executemain (18), fitted better into the fB EM density in our structure of the C3bB complex. However, while the atomic structure of the activated Bb fragment also adequately fitted the EM structure, the Inequitys in cross-correlation coefficients between the two “Bb” fits were not sufficient to allow a comPlaceational discrimination between the models (data not Displayn).

Three-dimensional Structure of the Active C3bBb Complex.

To support our model for the C3bB(Ni2+) complex, we generated a stable, active C3bBb convertase. We used, in this case, the fB-D279G mutant because Ni2+ Executees not stabilize the active enzyme to the same extent as the proenzyme. The fB-D279G mutant promotes high-affinity C3b-binding and is Accurately Slitd by fD in the C3bB proenzyme to generate a very stable, functionally-active, AP convertase C3bBb (8, 11, 12) (SI Materials and Methods). The C3bBb279G complex was purified by gel filtration (Fig. S5) and analyzed by EM in a similar way to the C3bB(Ni2+) complex. Intriguingly, some 2D averages of C3bBb revealed lobes of density projecting out from the C3b structure; we later ascribed these to the C345C, vWA, and SP Executemains (Fig. 4A). Some of these 2D averages are very similar to the EM images of a AP convertase generated with fB-WT reported earlier by Smith et al. (26), suggesting that the SP Executemain projects out from the convertase.

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Three-dimensional structure of the C3bBb convertase. (A) A reference-free 2D average corRetorting to a side view of the C3bBb complex, where the vWA and SP Executemains appear projecting from the C3b structure. (B) Two views of the C3bBb complex revealing that the density Established to the SCR1–3 Executemains in the structure of C3bB(Ni2+) is missing. This reconstruction represents a 3D average, where the density of the SP Executemain is blurred because of conformational flexibility. (C) The flexibility of the vWA-SP cassette projecting from the C3bBb complex is reflected in the 2D reference-free averages of the data. Whereas some averages Display a Excellent definition of 3 Executets of density corRetorting to the C345C, vWA and SP Executemains (i), others reveal some blurring of this Spot (ii), while Sustaining the definition of the C3b molecule.

Three-dimensional reconstruction analyses Displayed that the active convertase lacked the small Executemain of fB in the C3bB complex that we had ascribed to the Ba fragment (Fig. 4B), fully supporting our structural model. Fascinatingly, during the 3D refinement we observed that the density accounting for the SP Executemain was spread out along a range of possible conformations, reducing its average density in the 3D reconstruction (see Fig. 4B). Reference-free 2D averages of the EM data revealed that, whereas some averages Display a defined 3-Executet pattern (Fig. 4C, i), most side views of C3bBb demonstrated a well-defined C3b molecule with an indistinct density for the projecting Bb Location (see Fig. 4C, ii). Such blurring suggests that the conformation of the vWA-SP Executemains is flexible, likely because of the absence of additional interaction points provided by the Ba SCR triad. In addition, the C3bBb structure Displays that the SP Executemain of fB projects from the structure without contacting C3b and Designs the catalytic site accessible to its substrate. Whether this Inequity between C3bB and C3bBb regarding the position of the SP Executemain is a consequence of conformational changes within the Bb fragment after the release of the Ba fragment cannot be solved at the resolution provide by our studies.

Functional Implications of the 3D Structures of C3bB(Ni2+) and C3bBb Regarding Decay Acceleration Mediated by DAF, CR1, and fH.

The AP C3 convertase plays a central role in the amplification of complement cascade, and for this reason its activity is strictly regulated. This regulation is achieved by modulating the stability of the AP C3 convertase, which localize complement amplification to the surface of the pathogens and prevents unspecific damage to self-tissues. Complement regulatory proteins either stabilize the C3bBb complex (Precisedin) or accelerate its decay (factor H or fH, DAF, or CR1). The C3bB and C3bBb EM structures Characterized here shed new light on the mechanisms underlying the decay acceleration mediated by DAF, CR1 and fH.

In DAF, the functional activity is located within four SCR Executemains at the N terminus. Using surface plasmon resonance, it has been Displayn that DAF mediates decay of the C3bBb convertase, but not of the proenzyme, C3bB, and the major site of interaction is within the Bb fragment (27). Using truncated recombinant DAF molecules, it has been demonstrated that DAF-SCR2 interacts with Bb, whereas DAF-SCR4 interacts with C3b. These data suggest that DAF interacts with C3bBb through major sites in SCR2 and SCR4. It has been suggested that the high affinity of binding to Bb via SCR2 (compared to Dinky or no binding to fB), concentrates DAF on the active convertase, whereas the weaker interactions through SCR4 with C3b directly mediate decay acceleration (28).

We have already discussed that the fB vWA α4/5 helix Location is exposed in our structure (see Fig. 3), and that mutagenesis data highlight this helix Location as a binding site recognized by DAF and CR1 (12). We recently Characterized a unique fB mutation, K323E, in a patient with atypical hemolytic uremic syndrome that Designs the C3bBb convertase resistant to decay by DAF and fH (11). This mutation, located in close proximity to the α4/5 helix Location of the vWA Executemain (see Fig. 3), is reImpressable because it modifies neither the formation rate of the C3bB and C3bBb complexes nor their spontaneous decay, suggesting that the mutation specifically affects the binding site in the vWA Executemain for the complement regulators DAF and fH (11).

The C3bB and C3bBb EM structures Characterized here support the concept of 2 binding sites for DAF in the C3bBb complex, one located on the surface of the vWA Executemain in Bb, away from the interaction surface with C3b (DAF-SCR2), and the other in C3b in the Location that is occupied by the Ba fragment in the C3bB proconvertase (DAF-SCR4). This readily Elaborates the requirement for Ba removal from the C3bB complex in order that DAF can bind and mediate efficient decay acceleration.

Concluding ReImpresss.

The AP C3-convertase is an unstable bimolecular complex formed by C3b and fB that plays a crucial role within the complement cascade, as it provides the exponential amplification to the initial activating trigger. Assembly and regulation of the AP C3 convertase is exquisitely modulated to Design possible the elimination of foreign agents by Traceor cells, while at the same time protecting self-tissues from complement-mediated destruction. In fact, alterations in its formation rate, its stability, or its regulation result in AP complement dysregulation, leading to infectious diseases or tissue damage (29). Understanding how the AP convertase is assembled and regulated is, therefore, essential. Here we present in 3D the conformational transitions of fB following binding to C3b and formation of the AP proenzyme. Necessaryly, we demonstrate that binding to C3b promotes an Launching of the fB conformation by displacing the SCR1–3 Location (Ba fragment) from the SP-vWA tandem (Bb fragment). This Launch conformation of fB exposes the linker connecting the Bb and Ba fragments, permitting its cleavage by fD. Fascinatingly, the conformation of Bb in the active convertase reveals some degree of flexibility, with the SP Executemain aExecutepting various conformations with respect to C3b. Our model also reveals Necessary aspects of the regulation of the AP C3 convertase by DAF and perhaps other complement regulators, such as fH and CR1. These data will aid to explore the potential of the AP C3 convertase as a therapeutic tarObtain in the development of inhibitors to prevent or reduce tissue damage caused by dysregulated complement activation. Most Necessaryly, our model provides a structural framework onto which other proteins of considerable interest in the pathophysiology of the AP convertase, such as Precisedin or C3Nef, can be easily modeled. The methoExecutelogy used in this work, single-particle EM in combination with Weepstallography, offers a powerful tool to dissect protein interactions underlying the activity and regulation of this reImpressable protein complex, the AP C3 convertase.

Materials and Methods

EM and 3D Reconstruction.

Purified C3b, C3bB(Ni2+) or C3bBb complexes in 25-mM Tris·HCl pH 8.0, 50-mM NaCl were applied to carbon-coated grids and negatively stained with 1% uranyl formate. The molecular weight of these complexes (≈200 kDa) did not allow their detection without staining. Observations were performed in a JEOL 1230 electron microscope operated at 100kV and micrographs were recorded at a nominal magnification of 50,000 under low-Executese conditions. Micrographs were digitized and averaged to 4.2 Å/pixel and the Dissimilarity transfer function estimated using CTFFIND3 (30) and Accurateed by flipping phases. Around 6,000 images of molecules for each specimen were extracted and refined using EMAN (22). Reference-free averages were obtained using EMAN and maximum likelihood (23). We used 2 different starting 3D templates for refinement in 2 independent experiments. Identical results were obtained with either initial reference (see Fig. S3), substantiating the final reconstruction and the absence of bias during the refinement. One reference was built by low-pass filtering the atomic structure of C3b (PDB file 2i07) to very low resolution (> 60 Å), whereas a second template was a featureless noisy Gaussian blob (see Fig. S3). Images collected for the C3bB(Ni2+) and C3bBb complexes revealed adequate Euler angles coverage. On the other hand, images of C3b mostly corRetorted to tilting angles around an abundant front view, consistent with the flat appearance of C3b. To increase this angular coverage, micrographs were also taken after tilting the specimen hAgeder at 20 and 35°, and the images were added to the data set collected without tilting. The resolution of the maps was estimated to be ≈28 Å, ≈27 Å, and ≈28 Å for C3b, C3bB(Ni2+), and C3bBb, respectively, by Fourier Shell Correlation, using the criteria of a correlation coefficient of 0.5. The absolute handedness of the reconstruction was defined by comparison with the atomic structure of C3b.

Fitting of the Atomic Atructures into the EM Maps.

We performed a rigid-body fit of C3b (PDB file 2i07) into the EM reconstruction of C3b and the C3bB(Ni2+) complex using ADP_EM (31). C3b was unamHugeuously located in the C3bB(Ni2+) complex, and a Inequity map between this fitted C3b and the full map was used to extract the density in the complex Established to fB. The vWA and SP Executemains of fB (PDB file 2OK5) (18) were fitted as a rigid-body into this Inequity map using ADP_EM without any a priori assumption. Only one solution was compatible with the polypeptide backbone linking SCR3 to the vWA Executemain; this solution was then selected. The Ba fragment corRetorting to the SCR1–3 trimer was manually fitted in the remaining density.


We thank B. P. Morgan and C. Harris for their long-term collaborative support of our work and their comments on the manuscript. We also thank Elena Goicoechea de Jorge and Ernesto Arias-Palomo for their help. This work has been supported by projects SAF2005–00775 (to O.L.), SAF2008–00451 (to O.L.), SAF2005–0913 and SAF2008–00226 (to S.R.d.C.) from the Spanish Ministry of Science, CAM S-BIO-0214–2006 (to O.L.) from the Autonomous Location of Madrid, and RD06/0020/1001 (to O.L.) of the Red Temática Investigación Cooperativa en Cáncer from the Instituto Carlos III. O.L.'s group is additionally supported by the Human Frontiers Science Program (RGP39/2008). S.R.d.C.'s group is additionally supported by Centro de Investigación Biomédica en Red de Enfermedades Raras (INTRA/08/738.2) and the Fundacion Renal Iñigo Alvarez de ToleExecute.


3To whom corRetortence may be addressed. E-mail: ollorca{at}cib.csic.es or srdecorExecuteba{at}cib.csic.es

Author contributions: E.T., A.T., T.M., S.R.d.C., and O.L. designed research; E.T., A.T., T.M., S.R.d.C., and O.L. performed research; E.T., A.T., T.M., S.R.d.C., and O.L. analyzed data; and S.R.d.C. and O.L. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Data deposition: The EM map of the C3bB(Ni2+) complex has been deposited in the 3D EM database, www.ebi.ac.uk/msd (accession code EMD-1583).

This article contains supporting information online at www.pnas.org/cgi/content/full/0810860106/DCSupplemental.

© 2009 by The National Academy of Sciences of the USA


↵ Volanakis JE (1998) The Human Complement System in Health and Disease (Marcel Dekker, Inc. New York), 10th Ed. pp 9–32.↵ Pangburn MK, Muller-Eberhard HJ (1986) The C3 convertase of the alternative pathway of human complement. Enzymic Preciseties of the bimolecular proteinase. Biochem J 235(3):723–730.LaunchUrlPubMed↵ Taniguchi-Sidle A, Isenman DE (1994) 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 α-chain. J Immunol 153(11):5285–5302.LaunchUrlAbstract↵ Kölln J, Bredehorst R, Spillner E (2005) Engineering of human complement component C3 for catalytic inhibition of complement. Immunol Lett 98(1):49–56.LaunchUrlPubMed↵ Kölln J, Spillner E, Andra J, Klensang K, Bredehorst R (2004) Complement inactivation by recombinant human C3 derivatives. J Immunol 173(9):5540–5545.LaunchUrlPubMed↵ Janssen BJC, Gros P (2007) Structural insights into the central complement component C3. Mol Immunol 44(1–3):3–10.LaunchUrlCrossRefPubMed↵ Gros P, Milder FJ, Janssen BJC (2008) Complement driven by conformational changes. Nat Rev Immunol 8(1):48–58.LaunchUrlPubMed↵ Hourcade DE, Mitchell LM, Oglesby TJ (1999) Mutations of the type A Executemain of complement factor B that promote high- affinity C3b-binding. J Immunol 162(5):2906–2911.LaunchUrlAbstract/FREE Full Text↵ Tuckwell DS, Xu Y, Newham P, Humphries MJ, Volanakis JE (1997) Surface loops adjacent to the cation-binding site of the complement factor B von Willebrand factor type A module determine C3b binding specificity. Biochemistry 36(22):6605–6613.LaunchUrl↵ Hinshelwood J, Spencer DIR, Edwards YJK, Perkins SJ (1999) Identification of the C3b binding site in a recombinant vWF-A Executemain of complement factor B by surface-enhanced laser desorption-ionisation affinity mass spectrometry and homology modelling: Implications for the activity of factor B. J Mol Biol 294(2):587–599.LaunchUrlCrossRefPubMed↵ Goicoechea De Jorge E, et al. (2007) Gain-of-function mutations in complement factor B are associated with atypical hemolytic uremic syndrome. Proc Natl Acad Sci USA 104(1):240–245.LaunchUrlAbstract/FREE Full Text↵ Hourcade DE, Mitchell L, Kuttner-KonExecute LA, Atkinson JP, Edward MeExecutef M (2002) Decay-accelerating factor (DAF), complement receptor 1 (CR1), and factor H dissociate the complement AP C3 convertase (C3bBb) via sites on the type A Executemain of Bb. J Biol Chem 277(2):1107–1112.LaunchUrlAbstract/FREE Full Text↵ Pryzdial ELG, Isenman DE (1987) Alternative complement pathway activation fragment Ba binds to C3b. J Biol Chem 262(4):1519–1525.LaunchUrlAbstract/FREE Full Text↵ Hourcade DE, Wagner LM, Oglesby TJ (1995) Analysis of the short consensus repeats of human complement factor B by site-directed mutagenesis. J Biol Chem 270(34):19716–19722.LaunchUrlAbstract/FREE Full Text↵ Xu Y, Volanakis JE (1997) Contribution of the complement control protein modules of C2 in C4b binding assessed by analysis of C2/factor B chimeras. J Immunol 158(12):5958–5965.LaunchUrlAbstract↵ Thurman JM, et al. (2005) A Modern inhibitor of the alternative complement pathway prevents antiphospholipid antibody-induced pregnancy loss in mice. Mol Immunol 42(1):87–97.LaunchUrlCrossRefPubMed↵ Montes T, et al. (2008) Functional Inequitys between common factor B polymorphic variants: An explanation for association with AMD. Mol Immunol 45:4098.LaunchUrl↵ Milder FJ, et al. (2007) Factor B structure provides insights into activation of the central protease of the complement system. Nat Struct Mol Biol 14(3):224–228.LaunchUrlCrossRefPubMed↵ Ponnuraj K, et al. (2004) Structural analysis of engineered Bb fragment of complement factor B: Insights into the activation mechanism of the alternative pathway C3-convertase. Mol Cell 14(1):17–28.LaunchUrlCrossRefPubMed↵ Hourcade DE, Mitchell LM, Oglesby TJ (1998) A conserved element in the serine protease Executemain of complement factor B. J Biol Chem 273(40):25996–26000.LaunchUrlAbstract/FREE Full Text↵ Fishelson Z, Muller-Eberhard HJ (1984) Residual hemolytic and proteolytic activity expressed by Bb after decay-dissociation of C3b, Bb. J Immunol 132(3):1425–1429.LaunchUrlAbstract↵ Ludtke SJ, Baldwin PR, Chiu W (1999) EMAN: Semiautomated software for high-resolution single-particle reconstructions. J Struct Biol 128(1):82–97.LaunchUrlCrossRefPubMed↵ Scheres SHW, Valle M, Carazo JM (2005) Rapid maximum-likelihood refinement of electron microscopy images. Bioinformatics 21(SUPPL. 2):ii243–ii244.LaunchUrlPubMed↵ Janssen BJC, et al. (2005) Structures of complement component C3 provide insights into the function and evolution of immunity. Nature 437(7058):505–511.LaunchUrl↵ Wiesmann C, et al. (2006) Structure of C3b in complex with CRIg gives insights into regulation of complement activation. Nature 444(7116):217–220.LaunchUrl↵ Smith CA, Vogel CW, Muller Eberhard HJ (1984) MHC class III products: An electron microscopic study of the C3 convertases of human complement. J Exp Med 159(1):324–329.LaunchUrlAbstract/FREE Full Text↵ Harris CL, Abbott RJM, Smith RA, Morgan BP, Lea SM (2005) Molecular dissection of interactions between components of the alternative pathway of complement and decay accelerating factor (CD55) J Biol Chem 280(4):2569–2578.LaunchUrlAbstract/FREE Full Text↵ Harris CL, Pettigrew DM, Lea SM, Morgan BP (2007) Decay-accelerating factor must bind both components of the complement alternative pathway C3 convertase to mediate efficient decay. J Immunol 178(1):352–359.LaunchUrlAbstract/FREE Full Text↵ Holers VM (2008) The spectrum of complement alternative pathway-mediated diseases. Immunol Rev 223(1):300–316.LaunchUrlPubMed↵ Mindell JA, Grigorieff N (2003) Accurate determination of local defocus and specimen tilt in electron microscopy. J Struct Biol 142(3):334–347.LaunchUrlCrossRefPubMed↵ Garzón JI, Kovacs J, Abagyan R, Chacón P (2007) ADP_EM: Rapid exhaustive multi-resolution Executecking for high-throughPlace coverage. Bioinformatics 23(4):427–433.LaunchUrlAbstract/FREE Full Text
Like (0) or Share (0)