Structural insight on the recognition of surface-bound by the I domain of complement 3

Goran Bajica, Laure Yatimea, Robert B. Simb, Thomas Vorup-Jensenc,1, and Gregers R. Andersena,1

Departments of aMolecular Biology and Genetics and cBiomedicine, Aarhus University, DK-8000 Aarhus, Denmark; and bDepartment of Pharmacology, University of Oxford, Oxford OX1 3QT, United Kingdom

Edited by Douglas T. Fearon, University of Cambridge School of Clinical Medicine, Cambridge, United Kingdom, and approved August 28, 2013 (received for review June 13, 2013)

Complement receptors (CRs), expressed notably on myeloid and degranulation, and changes in leukocyte cytokine production (2, lymphoid cells, play an essential function in the elimination of 5–7). CR3, and to a lesser degree CR4, are essential for the complement-opsonized pathogens and apoptotic/necrotic cells. In of complement-opsonized particles or complexes addition, these receptors are crucial for the cross-talk between the (6, 8, 9). Complement-opsonized immune complexes are cap- innate andadaptive branches ofthe . CR3 (also known tured in the lymph nodes by CR3-positive subcapsular sinus as Mac-1, integrin α β , or CD11b/CD18) is expressed on all macro- (SSMs) and conveyed directly to naïve B cells or M 2 γ phages and recognizes iC3b on complement-opsonized objects, en- through follicular dendritic cells (10) using CR1, CR2, and Fc abling their phagocytosis. We demonstrate that the C3d moiety of receptors for antigen capture (11, 12). Hence, antigen-presenting iC3b harbors the binding site for the CR3 αI domain, and our structure cells such as SSMs may act as antigen storage and provide B of the C3d:αI domain complex rationalizes the CR3 selectivity for iC3b. lymphocytes with antigens (10, 12). Based on extensive structural analysis, we suggest that the choice Here, we establish the C3d fragment as the minimal and high- affinity binding partner for the CR3 I domain. By contrast, the between a ligand glutamate or aspartate for coordination of a recep- binding site for the CR4 I domain was located in the C3c frag- tor metal ion-dependent adhesion site–bound metal ion is governed ment by electron microscopy (13). We present the crystal by the secondary structure of the ligand. Comparison of our structure structure of the CR3 I domain in complex with C3d. The classic to the CR2:C3d complex and the in vitro formation of a stable CR3:C3d: observation of CR3 binding to iC3b, but not to its precursor CR2 complex suggests a molecular mechanism for the hand-over of (14), is consistent with our structure. In addition, our structure CR3-bound immune complexes from macrophages to CR2-presenting and functional data suggest simultaneous binding of CR3 and cells in lymph nodes. another , CR2, to C3 fragments, which might provide the basis for trafficking of complement-opsonized innate immunity | phagocytosis | integrin receptor | structural biology immune complexes from macrophages to B cells and follicular dendritic cells in lymph nodes. ctivation of complement leads to proteolytic cleavage of the Results Acentral complement component, C3. Its major fragment, C3b, acts as an and becomes covalently bound to the CR3 and CR4 I Domains Recognize Distinct Binding Sites on iC3b. To activating surface via a reactive thioester located in the thioester quantitatively compare binding properties of the CR3 and CR4 I (TE) domain of nascent C3b (Fig. S1A). Proteolytic processing domain with regard to binding of C3 proteolytic fragments, C3b, by factor I within the CUB domain of C3b leads to the formation fi of iC3b and C3dg. Finally, C3d—which practically corresponds Signi cance to the TE domain present in C3, C3b, and iC3b (Fig. S1 B–G)— is formed by other plasma proteases. These activation products Fragments of complement component C3 tag surfaces such as are ligands for five complement receptors (1), with iC3b being those presented by microbial pathogens or dying host cells for the primary ligand of complement receptors (CRs) CR3 and recognition by cells from the . Comple- fi CR4 (also known as CD11c/CD18, p150,95, or integrin αXβ2), ment receptor (CR) 3 enables ef cient binding of complement- which is structurally similar to CR3. tagged surfaces by macrophages and dendritic cells, which Like other , CR3 is a heterodimeric complex of two eventually transport the CR3-bound material into lymph nodes. fi transmembrane , αM and β2. It is abundantly expressed The study identi es in atomic details the fragments of CR3 and on myeloid leukocytes, including , den- C3 required for such binding. The structural organization per- dritic cells, , and macrophages and also on lymphoid mits concomitant recognition by another complement receptor, natural killer (NK) cells (2). Most ligands, including iC3b (3), are namely CR2, expressed on cells of the , bound by the Von Willebrand factor A (VWA) domain in the suggesting a structural rationale for the exchange of antigens α-chain, also referred to as the αI domain owing to its insertion between leukocytes of the innate and adaptive immune sys- in the β-propeller domain. I domain residues coordinate a metal tems critical in the formation of humoral immune responses. ion essential for ligand recognition through a metal ion-dependent adhesion site (MIDAS). Integrins adopt at least three major Author contributions: G.B., L.Y., T.V.-J., and G.R.A. designed research; G.B. and T.V.-J. conformations in the cell membrane. The bent-closed conforma- performed research; R.B.S. contributed new reagents/analytic tools; G.B., L.Y., T.V.-J., tion is inactive in ligand binding, the extended-closed conforma- and G.R.A. analyzed data; and T.V.-J. and G.R.A. wrote the paper. tion has low ligand affinity, and the extended-open conformation The authors declare no conflict of interest. binds ligands with high affinity. The transition from the bent- This article is a PNAS Direct Submission. closed to the open-extended conformation is exerted by a cyto- Data deposition: Crystallography, atomic coordinates, and structure factors have been plasmic force on the leg of the β-subunit, a process usually referred deposited in the Data Bank, www.pdb.org (PDB ID code 4M76). to as the inside-out signaling (4). 1To whom correspondence may be addressed. E-mail: [email protected] Binding of ligands to CR3 leads to conformational changes in or [email protected]. its ectodomain transmitting an outside-in signal through the cell This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. membrane. This may lead to actin remodeling, phagocytosis, 1073/pnas.1311261110/-/DCSupplemental.

16426–16431 | PNAS | October 8, 2013 | vol. 110 | no. 41 www.pnas.org/cgi/doi/10.1073/pnas.1311261110 Downloaded by guest on September 27, 2021 + iC3b, C3c, C3dg, and C3d were immobilized in surface plasmon ion in the MIDAS site (Fig. 2B and Table S1). The ability of Ni2 resonance (SPR) flow cells. For both I domains good binding to stabilize MIDAS interactions with a ligand is well known from signals were observed with C3b, iC3b, and C3c as ligands (Fig. S2 the complement convertases (18). A–C and F–H). Nevertheless, even high concentrations (100 μM) Within the complex, C3d adopts the well-described compact of the CR3 or CR4 I domain did not lead to saturation. This is α-α6 barrel structure (Fig. 2A and Fig. S4C) with only minor consistent with X-ray crystallography (15) and inhibition experi- conformational differences compared with known structures ments showing that both the CR3 and CR4 I domain interact containing C3d (19–21). The α-helix connecting loop regions of weakly (Kd ∼300 μM) with acidic side chains acting as ligand C3d are presented in an alternating fashion at the circumference mimetics (16). The C3d and C3dg-coated surfaces produced of two opposite surfaces: a concave, mainly negatively charged robust SPR signals of ∼1,200 resonance units (RU) and showed surface and a positively charged convex surface. At the rim of the signs of saturation at high CR3 I domain concentrations (Fig. S2 concave surface Asp1247, situated in a loop region connecting D and E). By contrast, the CR4 I domain only poorly bound helices α10 and α11 (Fig. 2C and Fig. S4 C and D), provides the these fragments (Fig. S2 I and J). As detailed in other studies (16, final coordination bond to the divalent cation in the CR3 MI- 17), the binding kinetics of the CR3 and CR4 I domain ligand DAS (Fig. 2B). The quite polar interface between C3d and the I binding are not well-described with simple 1:1 Langmuir iso- domain is modest, with an area of ∼490 Å2, which is smaller but therms. The interactions were quantified by analysis of the sen- comparable to similar I domain:ligand complexes (Table S2). sorgrams with the EVILFIT algorithm, which calculates the Besides Asp1247 on C3d involved in the MIDAS ion interaction, minimal distribution in binding kinetics for the heterogeneous the nearby Asp1245 in C3d engages in a salt bridge with CR3 interactions with ligands (Fig. 1). In general, the modeled dis- Arg208 (Fig. 2C and Fig. S4D). C3d Lys1217 also seems to sta- tribution in kinetics efficiently described the experimental data bilize the interaction, because it is capable of forming salt bridges as reflected in the small rmsds. For the CR3 and CR4 I domain, with CR3 Glu178 and Glu179 and hydrogen bonds with main some of the interactions with C3b, iC3b, and C3c were of modest chain carbonyls of Leu205 and Leu206. Finally, C3d Arg1254 −4 −3 strength with Kd ∼10 to 10 M (Fig. 1 A and B and D–F), that also contributes to the interaction by forming hydrogen bonds is, quantitatively equivalent to binding of acidic side chains with main-chain carbonyls of Gly143 and Ile145 in the I domain reported earlier (16). However, the classic CR3 ligand iC3b via a water molecule. −7 −6 presented a population of interactions with Kd ∼10 to 10 M (Fig. 1B), not found with either C3b or C3c or for any fragments Conserved Features of MIDAS-Dependent Ligand Interactions. Be- probed with the CR4 I domain (Fig. S3). This high-affinity type sides iC3b, the CR3 I domain is responsible for interacting with of interaction dominated the binding of the CR3 I domain to a variety of ligands, including fibrinogen, ICAM-1, and RAGE C3dg and C3d (Fig. 1 and Fig. S3). (22), but structures of their I domain complexes are not avail- able. To identify general features among ligand side chains en- Crystal Structure of CR3 I Domain in Complex with C3d. Guided by gaging in MIDAS ion coordination that might promote the the quantitative investigations made above, we determined the identification of I domain interacting residues in other CR3 crystal structure of the CR3 I domain (subunit αM residues 127– ligands, we identified all unique structures containing a VWA 321, mature numbering) in complex with C3d (C3 residues 993– domain or a βI domain engaging in a MIDAS-dependent in- 1,288, prepro numbering) at 2.8 Å resolution (Table S1 and Fig. teraction with a ligand protein. Comparison of these structures S4A). The structure reveals a 1:1 complex between the CR3 I revealed obvious trends for the use of either aspartate or the domain and C3d (Fig. 2A). The integrin I domain folds into an longer glutamate side chains as MIDAS ligands. If the small α/β Rossmann fold and adopts its open conformation as shown aspartate side chain is used, as in the C3d complex, it is located by comparison with the open-conformation structure of the I in a loop region or at the termini of a peptide (Fig. S5 A–D). This domain (Fig. S4B). The I domain α7 helix is shifted toward its C is even true for complexes between antibody Fab fragments and I terminus, and this conformation is most likely favored by the domains (Fig. S5 E–G), where an aspartate is located periph- I316G mutation introduced for this purpose. The open confor- erally in the heavy chain CDR3 loop. Even smaller than the mation is likewise adopted in the MIDAS site, where the hexa- aspartate and located at the end of a flexible region is the ex- coordinated metal ion is coordinated by Ser142, Ser144, Thr209 posed C-terminal carboxylate group in the noncatalytic subunit and two water molecules (Fig. 2B), whereas the last coordination of complement convertases (C3b/cobra venom factor/C4b) co- BIOCHEMISTRY + position is occupied by an aspartate from C3d. Because Mg2 ordinating the MIDAS ion in the VWA domain (Fig. S5H) of the was not compatible with the electron density in the MIDAS site, catalytic subunit (factor B/C2). We found three unique examples + and because Ni2 was present in the crystallization buffer, we involving the longer glutamate side chain (Fig. S5 I–K). In com- + used anomalous diffraction data to confirm the presence of a Ni2 plexes between the LFA-1 I domain and ICAM-1/3/5, a glutamate

RU C3b iC3b C3d 1000 0 900 A B C 800 Fig. 1. SPR analysis of the C3 fragment binding 700 selectivity of the CR3 (A–C) and CR4 I (D–F) −1 600 500 domains. The CR3 or CR4 I domain, stabilized by 400 mutagenesis in the open, ligand-binding confor-

CR3 −2 300 200 mation (50, 51), were injected in concentrations 100 ranging from 250 nM to 100 μM over surfaces d −3 0 coupled with C3b (A and D), iC3b (B and E), or C3d k 500 0 10 D E F 450 (C and F). The data were analyzed with the EVILFIT

I domain 400 algorithm with settings assuming a priori all bind- 350 log −1 300 ing parameters to be equally likely (52, 53). The 250 volume of interactions, indicated with colored 200 CR4 −2 contours (in RU as shown by scale bars) was plotted 150 − 100 as a function of the dissociation constant (10 8 M ≤ − − − − 50 K ≤ 10 1 M) and rate (10 3 s 1 ≤ k ≤ 100 s 1). Red −3 0 D d −8 −7 −6 −5 −4 −3 −2 −1 −8 −7 −6 −5 −4 −3 −2 −1 −8 −7 −6 −5 −4 −3 −2 −1 arrows indicate a population of high-affinity inter-

actions for the CR3 I domain (KD ∼0.4 μM) shared log K between iC3b and C3d but not observed for inter- 10 D actions with C3b.

Bajic et al. PNAS | October 8, 2013 | vol. 110 | no. 41 | 16427 Downloaded by guest on September 27, 2021 α7 C-terminal CR3 I-domain A α5 B Asp 242

Ni2+ Ser 142 N-terminal C-terminal α1 Thr 209 α5

N-terminal C3d Ser 144

α3 Asp 1247 Q1013 Fig. 2. The structure of the C3d:I domain complex. Glu179 (A) The edge of C3d (brown) interacts with the MI- C + Leu205 DAS (marked by the Ni2 ion) of the CR3 I domain Leu205 2+ Glu179 (purple). (B) The Ni ion bound in the MIDAS. The 45° Ile145 Glu178 Ile145 electron density contoured at 6 σ obtained from Glu178 anomalous differences in diffraction data (Table S1) + is shown as a mesh around the Ni2 ion. (C) Details Gly143 Leu206 Gly143 Arg208 Asp1247 Arg1254 Lys1217 of the intermolecular interface with putative hy- Lys1217 Arg1254 drogen bonds and electrostatic interactions in- dicated by dashed lines. C3 labels are underlined Asp1245 in all panels. Fig. S4D shows a stereo view of the interface.

side chain at the end of a β-strandinanICAMIgdomainisa agreement with our structure. Furthermore, the participating MIDAS ion ligand. A resembling situation is found for the α2β1 I residues are strictly or highly conserved in mammalian C3 and domain:collagen complex. In both cases the glutamate is pro- CR3 αM sequences (Fig. S6A). In contrast, none of the C3 res- truding from a region containing regular secondary structure with idues involved in the I domain interaction is conserved in the little curvature and flexibility that would permit a closer approach structural C3 homolog C4 (Fig. S6B) (24), also being cleaved to to the MIDAS ion and the use of the shorter aspartate. In a third C4b and undergoing further degradation to iC4b and C4d. case, a crystal-packing interaction by a glutamate located imme- Likewise, αM residues with side chains interacting with C3d diately after an α-helix mimics a ligand of the CR3 I domain (15). (Arg208, Glu178, and Glu179) are not conserved in CR4. Im- Overall, there seems to be a steric selectivity in the MIDAS ion portantly, our structure also offers an explanation for why the coordination by ligand acidic groups: An aspartate side chain is iC3b proteolytic stage must be reached to allow CR3 interaction preferred in loops or flexible termini and the longer glutamate is (25). The CUB domain in C3b connects the TE domain to the preferred in regions with secondary structure, whereas both as- C3c moiety of C3b, and proteolytic degradation of the CUB partate and glutamate located next to secondary structure are domain probably causes its partial collapse in the resulting iC3b possible MIDAS ion coordinators. (Fig. S1E). Superposition of known structures of C3b suggests that the CUB domain prevents the I domain from interacting C3d Asp1247 Is Essential for CR3 I Domain Interaction. Because with the TE domain by steric hindrance (Fig. 4). Finally, our integrins may promiscuously bind acidic residues exposed on structure is in accordance with a favorable geometry where the protein surfaces (15, 16, 22) and to exclude crystal-packing CR3 binding site of iC3b or C3d(g) opsonizing the activator is artifacts, we mutated C3d residues in the intermolecular in- accessible for a phagocytotic CR3-presenting cell. The CR3 terface. We tested iC3b and WT or mutated C3d for their ability binding site is separated by 40 Å from the Gln1013 forming the to interact with CR3 I domain by isothermal titration calorimetry covalent bond to the complement activator surface (Fig. 4A). In (ITC) experiments (Fig. 3). All of the binding ligands interacted conclusion, the in vivo relevance of our structure is strongly with the CR3 I domain in a 1:1 stoichiometry, and iC3b gener- supported by (i) SPR and ITC experiments quantitating the in- ated from C3b with factor I bound the I domain with a KD of 600 teraction of iC3b, WT C3d, and mutated C3d with the I domain, nM. The I domain bound WT C3d with a KD of 450 nM. These (ii) the conservation of the residues in the molecular interface, numbers are in good agreement with the high-affinity site iden- (iii) our ability to rationalize the discrimination against CR3 tified by SPR (Fig. 1 B and C and Fig. S3B). Mutation of C3d binding to C3b, and (iv) the inferred lack of steric hindrance Asp1247 to alanine abolished CR3 I domain binding to C3d, imposed by the activator upon CR3 binding. suggesting a crucial role of this aspartate in the interaction with the MIDAS-coordinated cation. Replacing Asp1247 by a gluta- C3 Thioester Domain As a Molecular Hub. Our structure of the C3d:I mate also impaired the interaction, showing that simple con- domain complex together with the structures of C3 (26, 27), the servation of charge is not sufficient (Fig. 3). Most likely, when C3d complexes with (20, 28), and CR2 (21) demon- the longer glutamate side chain in this C3d mutant interacts with strate that a substantial fraction of the surface of the C3 TE the I domain MIDAS site the adjoining interactions in the in- domain is involved in intra- or intermolecular contacts in one of terface are difficult to form. The R1254A mutant (Fig. 3) dis- the multiple functional states of C3. Strikingly, surface residues played a lowered affinity of 2.2 μM, whereas the mutant C3d on the TE domain interacting with other domains in native C3 proteins K1217A and K1217A/ R1254A showed no detectable are almost perfectly separated from those surface patches for- binding. Together, these data indicate that C3d Asp1247 is es- ming contacts with fH, CR3, and CR2 in their C3d complexes. In sential for the interaction with CR3 I domain but is not suffi- C3, the convex surface of C3d forms contacts with the MG2, cient, because other residues of the opposite charge, namely CUB, and MG8 domains (total interface area of ∼2,650 Å2) that K1217 and R1254, are required to steer the interaction. bury more than 20% of the TE domain surface (Fig. 5A). On the circumference and the rim of the concave surface of C3d, surface Physiological Significance of the C3d:I Domain Structure. The in vitro patches interacting with fH (∼600 Å2), CR3 (∼500 Å2), and CR2 confirmation of our crystal structure by ITC and SPR experi- (∼1,100 Å2) are likewise almost perfectly nonoverlapping with ments is also supported by prior data, because the iC3b binding each other and with the above-mentioned C3 interdomain in- site on the CR3 I domain earlier suggested (23) is in excellent terface (Fig. 5 B and C). Hence, especially in the iC3b state, the

16428 | www.pnas.org/cgi/doi/10.1073/pnas.1311261110 Bajic et al. Downloaded by guest on September 27, 2021 80 90

KD = 0.6 μM KD = 0.45 μM KD = 2.2 μM

molar ratio iC3b:I domain molar ratio C3d WT:I domain molar ratio C3d R1254A:I domain molar ratio C3d D1247E:I domain

Fig. 3. ITC studies of the CR3 I domain interaction with iC3b, C3d WT, C3d R1254A, and C3d D1247E mutants. Raw titration isotherms and the integrated peak areas are shown. Dissociation constants calculated according to a simple independent binding site model are displayed.

TE domain may bind simultaneously and strongly to the two further corroborate the existence of the ternary CR2:C3d:CR3 complement receptors and collateral binding of factor H and complex, size-exclusion chromatography experiments were con- CR3 seems possible as well. ducted (Fig. S8). The chromatograms and SDS/PAGE analysis of fractions from these experiments showed that the complexes Simultaneous Binding of CR2 and CR3 to iC3b. CR2 binds with elute in the expected order CR3:C3d:CR2, CR3:C3d, and CR2: similar affinity to iC3b and C3d(g) and with C3d as the minimal C3d (Fig. 5E and Fig. S8). Importantly, the presence of CR2 in ligand (29). To verify whether the formation of the CR2:C3d: the early fractions from the CR3:C3d:CR2 experiment can only CR3 complex suggested above is possible, C3d and CR2 com- be explained by the fact that CR2 was engaged in the ternary plement control protein (CCP) domains 1-2 were subjected to complex together with C3d and CR3. In conclusion, our pull- pull-down experiments with immobilized CR3 I domain. Bound down and size-exclusion chromatography experiments confirmed proteins were eluted with EDTA to disrupt MIDAS–ligand the existence of a ternary complex in which the CR3 I domain interactions. In the presence of WT C3d, both CR2 and C3d and the CR2 CCP 1-2 fragment bind simultaneously to C3d. were eluted from the CR3 affinity column, demonstrating that the CR3 I domain and CR2 CCP 1-2 are able to simultaneously Discussion bind to C3d (Fig. 5D and Fig. S7). Using instead the C3d The engulfment by immune cells of complement-opsonized ob- D1247A mutant no protein was eluted, showing that CR2 is not jects is a fundamental property of the human immune defense interacting nonspecifically with the resin and that this C3d as- where CR3 plays an essential role (8). CR3’s specificity for iC3b partate is essential for interaction with the CR3 MIDAS. With as ligand has been established for 30 years. Now, our results the C3d mutant D1154A unable to bind CR2 (30), CR2 was define where and how CR3 interacts with iC3b and further virtually absent from the EDTA eluate (Fig. 5D and Fig. S7). To emphasize the astonishing central role of the C3d moiety in in-

termolecular contacts. The CR3 I domain is well established as BIOCHEMISTRY the primary binding site for iC3b (3, 31). The quantitative find- ings in our study clearly suggest that the binding of the CR3 I C345C domain to C3d is 100- to 500-fold stronger than its weaker AB binding of simple acidic groups. Hence, the structure of the C3d: Nt-α’ CR3 I domain complex shows a type of interaction distinct from CUB the earlier report showing the binding to a ligand mimetic, that βI-domain is, glutamate side chain contributed by a crystal lattice contact I-domain (15). Nevertheless, although the I domain is an important de- -propellerβ-propellerr terminant in the binding of iC3b and C3d(g), other regions in CR3 must contribute, because deletion of its I domain results in residual iC3b affinity (32). Both the αM β-propeller and the β2 I- like domain (Fig. 4A) have been suggested to be implicated in the interaction with iC3b (33–35). On the iC3b side, mutations in Ni2+ the Nt-α′ region weaken the iC3b–CR3 interaction (36). The TE degradation product C3d(g) is not normally considered as a CR3 ligand, and soluble C3d cannot outcompete erythrocyte-bound iC3b in a rosette assay (37, 38). However, coating of erythrocytes with C3d facilitates their phagocytosis by monocytes in a metal C3b ~40 Å Q1013 ion- and CR3-dependent manner, although C3d does this much fi Activator surface less ef ciently than iC3b (39). This is likely to directly mirror the interaction we observe in the crystal structure and have quanti- Fig. 4. The CR3 I domain discriminates against C3b. (A) The structure of C3b tated by SPR and ITC. In summary, there seem to be at least two superimposed with the C3d:CR3 I domain complex. The βI domain and the contact points between iC3b and CR3, a crucial one involving the α-chain β-propeller within CR3 are shown schematically. (B) Close-up of the C3d:I domain interaction established by us. If the iC3b Nt-α′ region framed in A. Notice the overlap between the CR3 I domain and the C3b region in the C3c moiety of iC3b and the αM β-propeller and the CUB domain in the hypothetical CR3:C3b complex. C3 labels are underlined. β2 I-like domain in CR3 indeed are important for the receptor–

Bajic et al. PNAS | October 8, 2013 | vol. 110 | no. 41 | 16429 Downloaded by guest on September 27, 2021 ligand interactions, contacts involving these are likely to be AC spatially well separated from the C3d:I domain interface (Fig. 4A). Our model of iC3b binding to CR3 is clearly distinct from that proposed for CR4 by prior EM studies (13). In side-by-side Q1013 SPR experiments of the I domain binding of C3 fragments, the iC3b and C3d(g) fragments presented high-affinity binding sites 2+ CR3 for the CR3 I domain, whereas C3b and C3c had no such Ni interactions. These findings corroborate earlier studies on the C3 I-domain intact receptor, supporting the relevance of studying the I do- 180o main selectivity towards C3 fragments. The CR4 I domain bound CR2 CCP 1-2 with almost indiscriminate kinetics to C3b, iC3b, and C3c, whereas the C3d(g) fragments were much poorer ligands. This is also in C3d close agreement with the work by Chen et al. (13) on the intact B CR3 ectodomain of CR4, which identified a major binding site for this receptor involving the MG3 and MG4 domains of C3c. The ori- entation of these domains, and hence the binding interface for CR4, are conserved between C3b and C3c (40), likely also making this the case for iC3b. This is quantitatively supported by the nearly identical kinetics for the CR4 I domain binding of these fragments CR2 observed in our study. fH Our finding that CR2 and the CR3 I domain can bind to the same molecule of C3d is intriguing, and because CR2 does not fH seem to recognize elements outside the C3d moiety of iC3b, our CCP 19-20 results imply that one iC3b or C3d(g) molecule on soluble im- D mune complexes can bridge a CR3-presenting cell with a CR2- presenting cell. The location of the ligand-binding region (CR2 Input Mw C3d WT C3d D1247A C3d D1154A proteins kDa CCP 1-2 and CR3 I domain) at the distant end of these large C3d CR2 50 FT W FeWT Elute FFWTFWT etulE TWFTWtulE receptors far from the cell membrane makes bridging between two cells a realistic scenario. One important 40 30 example is the suggested hand-over of complement-opsonized 25 immune complexes from CR3-bearing subcapsular sinus mac- 20 rophages to CR2 on naïve B cells or follicular dendritic cells in 15 lymph nodes (10, 12). Concerning the remaining complement receptors, the binding site for CR4 at the MG3–MG4 domains 10 mapped by EM (13) seems not to overlap with regions impli- cated in binding CR3. Both receptors are expressed together on many cell types including macrophages, monocytes, , Mw and NK cells (2), suggesting that simultaneous CR3 and CR4 E kDa C3d:CR2 C3d:CR3 CR3:C3d:CR2 binding to the same iC3b molecule is possible. The CRIg re- 50 ceptor expressed on a subset of tissue macrophages recognizes 40 C3b, iC3b, and C3c, and its binding site has been mapped to the C3d 30 MG domains 3, 4, 5, and 6 and the LNK regions of C3b (41). Be- CR3 25 cause the structure of iC3b is only known to low resolution and is CR220 controversial (13), the exact spatial relation between the iC3b 15 binding sites for CRIg and the CR3 I domain found by usis unknown but likely to be variable. However, these sites are nonoverlapping, 32 33 34 35 32 33 34 35 32 33 34 35 suggesting that both receptors could bind simultaneously to iC3b unless other parts of CR3 compete with CRIg. With respect to Fraction number , both CR1 and CR3 binding involve the α′ Fig. 5. The multiple interactions of the TE domain in C3 and its proteolytic same part of the Nt- region (36), suggesting mutually exclusive degradation products. (A) Surface areas (gray) in the C3 TE domain inter- binding of the two receptors to the same iC3b molecule. acting with the CUB, MG8, and the MG2 domains in native C3 are mapped Therapeutic intervention through blockade of CR3 has been onto C3d from its CR3 complex; the thioester Gln1013 is shown in red. (B)As suggested for treatment of cerebral stroke (42). Furthermore, in A after a 180° rotation. Surface areas of C3d interacting with CR2 (pink), migration to the brain is likely to involve CR3 (43). factor H (green), and CR3 (purple) are indicated. (C) Superposition of the However, owing to the diversity of CR3 ligands, their central role complex between C3d (brown surface) with the CR3 I domain (purple car- in immune clearance of pathogens and apoptotic cells by phago- toon) with that of the C3d:CR2 complex (21) and the C3d:factor H complex cytosis (8, 44), adhesion (45, 46), and transmigration (47) overall (20). (D) Silver-stained gel after denaturing SDS/PAGE analysis of fractions or even function-specific CR3 blockade is not without risks. As an from the CR3 I domain affinity column. Labels + and − indicate, respectively, example, systemic erythematosus is tightly linked with the presence or absence of the protein in the pull-down assay at that par- a single mutation in the αM chain of CR3 that mainly affects its ticular step. From left to right, molecular weight (Mw) marker together with role in phagocytosis (48). In the same manner, C3b and its pro- purified CR2 CCP 1-2 (insect cell-expressed) and C3d and CR3 I domain af- fi fl teolytic fragments also engage in interactions with a large number nity column pull-down using C3d WT showing the ow-through and the of other proteins, thereby severely complicating the development wash steps and the final EDTA elution. The same fractions are shown for the C3d D1247A and the D1154A control mutants unable to bind CR3 and CR2, of complement inhibitors. Our structure of the C3d:I domain respectively. All the volumes in wash and elution steps were the same. complex provides an important contribution to the development Likewise, the volumes loaded for SDS/PAGE analysis were identical. For of more selective complement inhibitors by identifying the sur- comparison, the pull-down was performed with the CR2 expressed in either face areas on the CR3 I domain and on the C3d moiety of im- bacteria or insect cells (Fig. S7). (E) Silver-stained gel after denaturing SDS/ portance for their mutual interaction. Owing to the proximity of PAGE analysis of fractions from analytical size-exclusion chromatography the C3d–CR2 interaction it may also contribute to improve the (Fig. S8). The same fractions resulting from the C3d:CR2, C3d:CR3 and CR3: CR2-based targeting strategies under development for site-spe- C3d:CR2 runs were analyzed. cific delivery of complement inhibitors (49).

16430 | www.pnas.org/cgi/doi/10.1073/pnas.1311261110 Bajic et al. Downloaded by guest on September 27, 2021 Methods the complex was determined by molecular replacement using Protein Data WT or mutated C3d and CR3 I domain were prepared as recombinant proteins Bank ID codes 1IDO and 1C3D as search models. Coordinates and structure in Escherichia coli. The CR2 CCP 1-2 fragment was prepared as recombinant factors have been deposited at the with ID code 4M76. De- protein in insect cells or E. coli. C3b was prepared by trypsin digestion of C3, tailed methods and the associated references can be found in SI Methods. and iC3b, C3d(g), and C3c were prepared from C3b by factor I digestion in the presence of fH. In the SPR experiments proteolytic C3 derivatives were coupled ACKNOWLEDGMENTS. We thank T. Springer for inspiration and advice, through primary amine groups to the CM4 sensor chip and the CR3 I or CR4 I Y. He and the beamline staff at the Swiss Light Source and European Syn- domain injected over the chip. ITC measurements were performed at 25 °C chrotron Radiation Facility for help with data collection, A.M. Bundsgaard by titration of CR3 I domain with iC3b or WT/mutated C3d. For the C3d pull- and B. W. Grumsen for technical assistance, and J. K. Jensen and L. T. Pallesen for help with isothermal titration calorimetry experiments. This work was down experiments, the CR3 I domain was coupled to CNBr-activated Sepharose, 2+ supported by the Lundbeck Foundation through the Lundbeck Foundation and C3d and CR2 were bound in the presence of Mg and eluted with EDTA. Nanomedicine Center, the MEMBRANES center, and the Novo-Nordisk Foun- The C3d:I domain complex was crystallized by vapor diffusion. Diffraction data dation through a Hallas-Møller Fellowship (to G.R.A.). T.V.-J. was supported from frozen crystals were collected with synchrotron radiation at the Swiss by the Carlsberg Foundation, the LEO Foundation, Helga og Peter Kornings Light Source or European Synchrotron Radiation Facility, and the structure of Fond, and Gluds Legat.

1. van Lookeren Campagne M, Wiesmann C, Brown EJ (2007) Macrophage complement receptor type 2 (CR2, CD21) does not abolish binding of iC3b or C3dg to CR2. JImmunol receptors and pathogen clearance. Cell Microbiol 9(9):2095–2102. 154(5):2303–2320. 2. Ross GD (2000) Regulation of the adhesion versus cytotoxic functions of the Mac-1/ 30. Clemenza L, Isenman DE (2000) Structure-guided identification of C3d residues es- CR3/alphaMbeta2-integrin glycoprotein. Crit Rev Immunol 20(3):197–222. sential for its binding to (CD21). J Immunol 165(7):3839–3848. 3. Diamond MS, Garcia-Aguilar J, Bickford JK, Corbi AL, Springer TA (1993) The I domain 31. Ueda T, Rieu P, Brayer J, Arnaout MA (1994) Identification of the complement iC3b is a major recognition site on the leukocyte integrin Mac-1 (CD11b/CD18) for four binding site in the beta 2 integrin CR3 (CD11b/CD18). Proc Natl Acad Sci USA 91(22): distinct adhesion ligands. J Cell Biol 120(4):1031–1043. 10680–10684. 4. Springer TA, Dustin ML (2012) Integrin inside-out signaling and the immunological 32. Yalamanchili P, Lu C, Oxvig C, Springer TA (2000) Folding and function of I domain- synapse. Curr Opin Cell Biol 24(1):107–115. deleted Mac-1 and lymphocyte function-associated antigen-1. J Biol Chem 275(29): 5. Lefort CT, et al. (2009) Outside-in signal transmission by conformational changes in 21877–21882. integrin Mac-1. J Immunol 183(10):6460–6468. 33. Li Y, Zhang L (2003) The fourth blade within the beta-propeller is involved specifically 6. Dupuy AG, Caron E (2008) Integrin-dependent phagocytosis: Spreading from micro- in C3bi recognition by beta 2. J Biol Chem 278(36):34395–34402. adhesion to new concepts. J Cell Sci 121(Pt 11):1773–1783. 34. MacPherson M, Lek HS, Prescott A, Fagerholm SC (2011) A systemic lupus erythematosus- 7. Shimaoka M, Takagi J, Springer TA (2002) Conformational regulation of integrin associated R77H substitution in the CD11b chain of the Mac-1 integrin compromises – structure and function. Annu Rev Biophys Biomol Struct 31:485–516. leukocyte adhesion and phagocytosis. J Biol Chem 286(19):17303 17310. fi 8. Underhill DM, Ozinsky A (2002) Phagocytosis of microbes: Complexity in action. Annu 35. Xiong Y-M, Haas TA, Zhang L (2002) Identi cation of functional segments within the – Rev Immunol 20:825–852. beta2I-domain of integrin alphaMbeta2. J Biol Chem 277(48):46639 46644. 9. Wright SD, Meyer BC (1986) Phorbol esters cause sequential activation and de- 36. Taniguchi-Sidle A, Isenman DE (1994) Interactions of human complement component activation of complement receptors on polymorphonuclear leukocytes. J Immunol C3 with factor B and with complement receptors type 1 (CR1, CD35) and type 3 (CR3, ’ 136(5):1759–1764. CD11b/CD18) involve an acidic sequence at the N-terminus of C3 alpha -chain. – 10. Phan TG, Grigorova I, Okada T, Cyster JG (2007) Subcapsular encounter and com- J Immunol 153(11):5285 5302. plement-dependent transport of immune complexes by lymph node B cells. Nat Im- 37. Schreiber RD, Pangburn MK, Bjornson AB, Brothers MA, Müller-Eberhard HJ (1982) munol 8(9):992–1000. The role of C3 fragments in endocytosis and extracellular cytotoxic reactions by – 11. Gray EE, Cyster JG (2012) Lymph node macrophages. J Innate Immun 4(5–6):424–436. polymorphonuclear leukocytes. Clin Immunol Immunopathol 23(2):335 357. fi fi 12. Gonzalez SF, et al. (2011) Trafficking of antigen in lymph nodes. Annu Rev 38. Ross GD, Lambris JD (1982) Identi cation of a C3bi-speci c membrane complement receptor that is expressed on lymphocytes, monocytes, neutrophils, and erythrocytes. Immunol 29:215–233. J Exp Med 155(1):96–110. 13. Chen X, Yu Y, Mi LZ, Walz T, Springer TA (2012) Molecular basis for complement 39. Gaither TA, Vargas I, Inada S, Frank MM (1987) The complement fragment C3d fa- recognition by integrin αXβ2. Proc Natl Acad Sci USA 109(12):4586–4591. cilitates phagocytosis by monocytes. Immunology 62(3):405–411. 14. Beller DI, Springer TA, Schreiber RD (1982) Anti-Mac-1 selectively inhibits the mouse 40. Janssen BJ, Christodoulidou A, McCarthy A, Lambris JD, Gros P (2006) Structure of C3b and human type three complement receptor. J Exp Med 156(4):1000–1009. reveals conformational changes that underlie complement activity. Nature 444(7116): 15. Lee JO, Rieu P, Arnaout MA, Liddington R (1995) Crystal structure of the A domain 213–216. from the alpha subunit of integrin CR3 (CD11b/CD18). Cell 80(4):631–638. 41. Wiesmann C, et al. (2006) Structure of C3b in complex with CRIg gives insights into 16. Vorup-Jensen T, et al. (2005) Exposure of acidic residues as a danger signal for rec- regulation of complement activation. Nature 444(7116):217–220. ognition of fibrinogen and other macromolecules by integrin alphaXbeta2. Proc Natl 42. Zhang L, et al. (2003) Effects of a selective CD11b/CD18 antagonist and recombinant Acad Sci USA 102(5):1614–1619. human tissue plasminogen activator treatment alone and in combination in a rat 17. Vorup-Jensen T (2012) Surface plasmon resonance biosensing in studies of the BIOCHEMISTRY embolic model of stroke. Stroke 34(7):1790–1795. binding between β₂ integrin I domains and their ligands. Methods Mol Biol 757:55–71. 43. Riou A, et al. (2013) MRI assessment of the intra-carotid route for macrophage de- 18. Fishelson Z, Müller-Eberhard HJ (1982) C3 convertase of human complement: En- livery after transient cerebral ischemia. NMR Biomed 26(2):115–123. hanced formation and stability of the enzyme generated with nickel instead of 44. Morelli AE, et al. (2003) Internalization of circulating apoptotic cells by splenic mar- magnesium. J Immunol 129(6):2603–2607. ginal zone dendritic cells: dependence on complement receptors and effect on cy- 19. Nagar B, Jones RG, Diefenbach RJ, Isenman DE, Rini JM (1998) X-ray crystal structure of tokine production. Blood 101(2):611–620. – C3d: A C3 fragmentandligand for complement receptor 2. Science 280(5367):1277 1281. 45. Diamond MS, et al. (1990) ICAM-1 (CD54): A counter-receptor for Mac-1 (CD11b/ 20. Morgan HP, et al. (2011) Structural basis for engagement by complement factor H of CD18). J Cell Biol 111(6 Pt 2):3129–3139. – C3b on a self surface. Nat Struct Mol Biol 18(4):463 470. 46. Chavakis T, et al. (2003) The pattern recognition receptor (RAGE) is a counterreceptor 21. van den Elsen JM, Isenman DE (2011) A crystal structure of the complex between for leukocyte integrins: A novel pathway for inflammatory cell recruitment. J Exp – human complement receptor 2 and its ligand C3d. Science 332(6029):608 611. Med 198(10):1507–1515. 22. Vorup-Jensen T (2012) On the roles of polyvalent binding in immune recognition: 47. Chavakis T, et al. (2004) The junctional adhesion molecule-C promotes neutrophil Perspectives in the nanoscience of immunology and the immune response to nano- transendothelial migration in vitro and in vivo. J Biol Chem 279(53):55602–55608. – medicines. Adv Drug Deliv Rev 64(15):1759 1781. 48. Fossati-Jimack L, et al. (2013) Phagocytosis is the main CR3-mediated function affected by 23. Ustinov VA, Plow EF (2005) Identity of the amino acid residues involved in C3bi the lupus-associated variant of CD11b in human myeloid cells. PLoS ONE 8(2):e57082. binding to the I-domain supports a mosaic model to explain the broad ligand rep- 49. Holers VM, Rohrer B, Tomlinson S (2013) CR2-mediated targeting of complement ertoire of integrin alpha M beta 2. Biochemistry 44(11):4357–4364. inhibitors: Bench-to-bedside using a novel strategy for site-specific complement 24. Kidmose RT, et al. (2012) Structural basis for activation of the by modulation. Adv Exp Med Biol 735:137–154. component C4 cleavage. Proc Natl Acad Sci USA 109(38):15425–15430. 50. Vorup-Jensen T, Ostermeier C, Shimaoka M, Hommel U, Springer TA (2003) Structure 25. Ross GD, Medof ME (1985) Membrane complement receptors specific for bound and allosteric regulation of the alpha X beta 2 integrin I domain. Proc Natl Acad Sci fragments of C3. Adv Immunol 37:217–267. USA 100(4):1873–1878. 26. Fredslund F, et al. (2006) The structure of bovine reveals 51. Xiong JP, Li R, Essafi M, Stehle T, Arnaout MA (2000) An isoleucine-based allosteric the basis for thioester function. J Mol Biol 361(1):115–127. switch controls affinity and shape shifting in integrin CD11b A-domain. J Biol Chem 27. Janssen BJ, et al. (2005) Structures of complement component C3 provide insights into 275(49):38762–38767. the function and evolution of immunity. Nature 437(7058):505–511. 52. Svitel J, Balbo A, Mariuzza RA, Gonzales NR, Schuck P (2003) Combined affinity and 28. Kajander T, et al. (2011) Dual interaction of factor H with C3d and glycosaminoglycans rate constant distributions of ligand populations from experimental surface binding in host-nonhost discrimination by complement. Proc Natl Acad Sci USA 108(7): kinetics and equilibria. Biophys J 84(6):4062–4077. 2897–2902. 53. Gorshkova II, Svitel J, Razjouyan F, Schuck P (2008) Bayesian analysis of heterogeneity 29. Diefenbach RJ, Isenman DE (1995) Mutation of residues in the C3dg region of human in the distribution of binding properties of immobilized surface sites. Langmuir complement component C3 corresponding to a proposed binding site for complement 24(20):11577–11586.

Bajic et al. PNAS | October 8, 2013 | vol. 110 | no. 41 | 16431 Downloaded by guest on September 27, 2021