Post on 26-May-2020
Structure of the C5‐eculizumab complex
1
Structural basis for eculizumab-mediated inhibition of the complement terminal pathway 1
2
Running title: Structure of the C5-eculizumab complex 3
Janus Asbjørn Schatz-Jakobsen†, Yuchun Zhang‡, Krista Johnson‡, Alyssa Neill‡, Douglas 4
Sheridan‡ & Gregers Rom Andersen†* 5
†Department of Molecular Biology and Genetics, Aarhus University, Gustav Wieds Vej 10C, DK-6
8000 Aarhus, Denmark 7
‡Alexion Pharmaceuticals Inc., 100 College Street, New Haven, CT, 06510 USA 8
*Correspondence should be addressed to 9
Gregers R. Andersen, Department of Molecular Biology and Genetics, Aarhus University, Gustav 10
Wieds Vej 10C, DK-8000 Aarhus, Denmark, email gra@mbg.au.dk, phone +45 51446530, fax +45 11
8619 6500 12
Or 13
Douglas Sheridan, Alexion Pharmaceuticals Inc., 100 College Street, New Haven, CT, 06510 USA, 14
email SheridanD@alxn.com, phone +01 2032718297, fax +01 2032718190 15
Structure of the C5‐eculizumab complex
2
Abstract 1
Eculizumab is a humanized monoclonal antibody approved for treatment of patients with 2
paroxysmal nocturnal hemoglobinuria (PNH) and atypical hemolytic uraemic syndrome. 3
Eculizumab binds complement component C5 and prevents its cleavage by C5 convertases, 4
inhibiting release of both the proinflammatory metabolite C5a and formation of the membrane 5
attack complex via C5b. Here we present the crystal structure of the complex between C5 and a Fab 6
fragment with the same sequence as eculizumab at a resolution of 4.2 Å. Five complementarity 7
determining regions (CDRs) contact the C5 MG7 domain, which contains the entire epitope. A 8
complete mutational scan of the sixty-six CDR residues identified twenty-eight residues as 9
important for the C5-eculizumab interaction, and the structure of the complex offered an 10
explanation for the reduced C5-binding observed for these mutant antibodies. Furthermore the 11
structural observations of the interaction are supported by the reduced ability of a subset of these 12
mutated antibodies to inhibit MAC formation as tested in a hemolysis assay. Our results suggest 13
that eculizumab functions by sterically preventing C5 from binding to convertases and explain the 14
exquisite selectivity of eculizumab for human C5 and how polymorphisms in C5 cause eculizumab-15
resistance in a small number of PNH patients. 16
Structure of the C5‐eculizumab complex
3
Complement is an ancient part of innate immunity that functions to detect and clear pathogen 1
invasion and altered self-cells. The complement system is activated upon detection of pathogen-2
associated molecular patterns (PAMPs) or danger-associated molecular patterns (DAMPs), which in 3
turn will lead to the formation of the C5 convertases, C4b2a3b (1) and C3bBb3b (2, 3), able to 4
cleave C5 and initiate the terminal pathway (TP) (Fig. 1A-B). Upon C5 cleavage by the C5 5
convertases, two fragments are formed. The smaller fragment, C5a, functions as a potent pro-6
inflammatory anaphylatoxin by signaling through C5aR1 triggering increased vascular permeability 7
on endothelial cells, chemotaxis and oxidative burst on phagocytes, release of pro-inflammatory 8
molecules, and activation of the adaptive immune system (reviewed by Klos et al (4)). The larger 9
fragment C5b can associate with C6, C7, C8 and 12-18 copies of C9 to form the membrane attack 10
complex (MAC) also known as C5b-9. Upon assembly, the MAC inserts into membranes of 11
pathogens without a cell wall resulting in a lytic pore (5, 6). Host cells are normally protected from 12
MAC lysis by the membrane associated GPI-anchored regulators CD55 and CD59, and by the 13
soluble regulators vitronectin and clusterin (7-11). 14
The humanized antibody eculizumab is the first and only therapeutic inhibitor of terminal 15
complement. Eculizumab was approved by the FDA and EMA for the treatment of paroxysmal 16
nocturnal hemoglobinuria (PNH) and atypical hemolytic uremic syndrome (aHUS) in 2007 and 17
2011, respectively. PNH is a rare form of hemolytic anemia caused by somatic mutations of the 18
PIG-A gene in hematopoietic stem cell lineages (12, 13). This causes a loss of GPI anchors for 19
CD55 and CD59 (12) in the mature erythrocytes bearing the mutation, making them susceptible to 20
lysis by as little as a single MAC. aHUS is a rare disease of complement dysregulation leading to 21
thrombotic microangiopathy (TMA) and severe kidney damage. Risk factors for aHUS include 22
mutations in factor H (CFH), factor I (CFI), factor B (CFB), C3 and membrane cofactor protein 23
(MCP) (reviewed in Noris and Remuzzi (14)), as well as infectious environmental stimuli and 24
Structure of the C5‐eculizumab complex
4
autoantibodies to complement proteins. The pathological conditions of aHUS are caused by the 1
insertion of MAC into the renal capillary endothelial cells and recruitment of proinflammatory cells 2
by C5a. Eculizumab binds to human C5 with high affinity and prevents its cleavage by complement 3
convertases into C5a and C5b (15) thereby providing a mechanism for the complete inhibition of 4
terminal complement activity in both PNH and aHUS. 5
Eculizumab was developed from the murine antibody 5G1.1, which was mapped to bind to the N-6
terminal region of the -chain of human C5. Moreover it has been shown to bind C5-derived 7
peptide fragments containing the KSSKC motif (residues 879 – 883) within the C5 MG7 domain 8
(16), distal to the scissile bond (R751-L752) that is cleaved by the convertase. The recent 9
identification of Arg885His/Cys polymorphisms in a small population of eculizumab-resistant 10
patients with PNH (17) provided further evidence that the antibody binds in this region. Eculizumab 11
is also highly specific for human C5 with little to no antagonist activity toward C5 from any other 12
primate species tested (Sheridan, unpublished data). 13
In order to comprehend in structural detail how eculizumab prevents convertase cleavage of its 14
target, we prepared a Fab molecule with the same sequence as eculizumab in the variable region 15
and determined the structure of the complex between this Fab and complement C5. Despite the 16
limited resolution of 4.2 Å our structure rationalizes existing functional data and genetic data 17
regarding the C5-eculizumab interaction. In addition, we present a complete mutational scan of the 18
eculizumab complementarity determining regions and a structure-based interpretation of the 19
observed functional defects induced by individual point mutations. 20
Structure of the C5‐eculizumab complex
5
MaterialsandMethods1
Protein purification 2
C5 was purified as described (18) with modifications. Following elution from the DEAE column, 3
Na2SO4 was added to C3/C5 fractions to a concentration of 400 mM and loaded on a 75 mL Phenyl 4
Sepharose FF column (GE Healthcare) equilibrated in Buffer A (20 mM HEPES pH 7.5, 400 mM 5
Na2SO4). The column was washed with 250 mL buffer A followed by 150 mL 50% buffer B (20 6
mM HEPES pH 7.5, 150 mM NaCl) and eluted with a 150 mL linear gradient from 50-100 % 7
buffer B. C5 fractions were adjusted with solid Na2SO4 to 400 mM and loaded on a 9 mL SOURCE 8
15PHE column (GE Healthcare) equilibrated in buffer A. The column was washed in buffer A and 9
eluted with a 120 mL gradient from 0-100% buffer C (20 mM HEPES pH 7.5, 50 mM NaCl). 10
Relevant fractions were pooled, concentrated and subjected to size exclusion chromatography 11
(SEC) on a 16/600 Superdex 200 column (GE Healthcare) in buffer D (20 mM HEPES pH 7.5, 75 12
mM NaCl). SEC purified C5 was finally loaded on a 1 mL MonoQ (GE Healthcare) in buffer D and 13
eluted with a 20 mL linear gradient from 75-500 mM NaCl. 14
A dual cytomegalovirus (CMV) expression plasmid encoding a light chain and a Fab fragment of 15
the heavy chain with the same primary sequence as eculizumab (BNJ416) was expressed transiently 16
in Expi293F cells according to the manufacturer’s recommendations (Invitrogen, Grand Island, 17
NY). The Fab fragment was purified from the cell culture supernatant using a kappa select resin 18
(GE Healthcare) equilibrated with PBS pH 7.4 and eluted with 100 mM glycine pH 2.7. Free light 19
chain was removed from the kappa select pool using cation exchange with SP sepharose (GE 20
Healthcare). The cation exchange column was equilibrated in Buffer E (50 mM sodium acetate pH 21
4.8), and eluted with a linear salt gradient of 0 – 30 % Buffer F (50 mM sodium acetate pH 4.8, 1 M 22
NaCl). Following purification, the Fab was determined to be greater than 99% pure using both size 23
exclusion chromatography (SEC-HPLC) and reverse phase chromatography (RP-HPLC). SEC-24
Structure of the C5‐eculizumab complex
6
HPLC was performed using 10 µg of purified sample on a TSKgel GW3000 column (Tosoh) using 1
an isocratic mobile phase consisting of 150 mM NaCl, 10 mM NaPO4, pH 7.0. Absorbance was 2
monitored at 214 nm. RP-HPLC was performed using 10 µg of purified sample on a Vydac C4 3
column (Grace) with a mobile phase of 64% Buffer G (5% acetonitrile, 0.1% trifluoroacetic acid 4
[TFA] in water) and 36% Buffer H (80% acetonitrile, 0.1% TFA in water) and monitoring 5
absorbance at 280 nm. 6
To isolate the complex, C5 was mixed with the BNJ416 Fab in a 1:1.2 molar ratio. The complex 7
was purified on a 10/300 GL Superdex 200 column (GE Healthcare) equilibrated in buffer D. 8
Complex containing fractions were pooled and concentrated to 8.5 mg/mL. 9
Structure determination 10
Initial crystallization conditions obtained with commercial screens from Hampton Research and 11
Molecular Dimensions were optimized with MIMER (19). The best crystals were obtained by 12
mixing the complex at 5-7 mg/mL 1:1 with a reservoir solution consisting of 0.1 M imidazole pH 13
6.2, 4% v/v Tacsimate and 7.8% - 8% w/v PEG 3,350 in sitting drops, which were equilibrated for 14
2-7 days at 4 oC prior to streak seeding. Crystals were cryoprotected in 0.1 M imidazole pH 6.2, 4% 15
v/v Tacsimate, 21.5% w/v PEG 3,350 and 8.5% v/v glycerol and frozen in liquid nitrogen. Data 16
were collected at ESRF ID29 at 100 K with a wavelength of λ= 0.97625 Å and processed with XDS 17
(20). The structure was solved by molecular replacement (MR) with phenix.phaser (21). The model 18
was improved by rebuilding using the programs ‘O’ (22) and Coot (23) and refined with 19
phenix.refine (24) using non-crystallographic restraints in combination with rigid body, simulated 20
annealing, individual ADPs and TLS refinement. Molprobity (25) was used for validation. Figures 21
were prepared with PyMol (www.pymol.org). Coordinates and structure factors are available at the 22
RCSB protein data bank www.rcsb.org with accession code 5I5K. 23
Structure of the C5‐eculizumab complex
7
Binding and Functional Assays 1
To screen for contact residues in eculizumab important for binding to C5, each amino acid in all 6 2
complementarity determining regions (CDRs) was subjected independently to substitution with 3
histidine and screened for altered binding kinetics to C5. Briefly, full length IgGs with the same 4
primary sequence as eculizumab containing single amino acid substitutions were expressed 5
transiently in Expi293F cells by co-transfecting 1:1 ratios of CMV expression plasmids encoding 6
the antibody light and heavy chains. Antibody expression levels were quantified by biolayer 7
interferometry on protein A sensors (ForteBio cat # 18-5010) mounted on an Octet QK (ForteBio) 8
and normalized to 2.4 µg/mL in 1× kinetic buffer (0.01% BSA and 0.002% (v/v) TWEEN® 20 in 9
PBS buffer, pH 7.4). The antibodies were then immobilized on protein A sensors and exposed to C5 10
at 26 nM to determine binding kinetics on an Octet Red with association in 1× kinetic buffer, pH 11
7.4 and dissociation in 1× kinetic buffer, pH 7.4 or pH 6.0. 12
Binding kinetics on a small subset of representative antibodies with single amino acid substitutions 13
were assayed using surface plasmon resonance (SPR) on a BIAcore 3000 instrument (GE 14
Healthcare). The experiments were performed using an anti-Fc human capture method at pH 15
7.4. Anti-Fc-Human (KPL #01-10-20) diluted to 0.1 mg/mL in 10 mM sodium acetate pH 5.0, was 16
immobilized on two flow cells of a CM5 chip for 8 minutes by amine coupling. The parental 17
antibody with the same primary sequence as eculizumab (EHL000) and single histidine 18
substitutions variants were diluted to 0.5 µg/mL in running buffer (HBS-EP, pH 7.4) and injected 19
onto one flow cell to achieve a capture level of 50 RUs on the immobilized anti-Fc human 20
surface. The second flow cell was used as a reference surface. C5 in a range of 8-10 concentrations 21
per antibody sample were injected over both flow cells to determine kinetics. The surface was 22
regenerated each cycle with 20 mM HCl, 0.01 % P20 (200 µL injection at 100 µL/min). The data 23
Structure of the C5‐eculizumab complex
8
was processed with a 1:1 Langmuir model using BIAevaluation 4.1 software with ‘double 1
referencing’- subtraction of both the reference flow cell and a 0 nM C5 (blank) cycle. 2
3
The same subset of mutated antibodies analysed by SPR were assayed for their potential to inhibit 4
terminal complement activity in a classical pathway chicken red blood cell (cRBC) hemolysis assay 5
in normal human serum. The EHL000 antibody was used as a positive control. Antibody-sensitized 6
cRBCs were prepared for each assay from 400 µL of chicken whole blood in Alsever’s buffer 7
(Lampire Biologicals) and washed 4 times with 1 mL of GVB++ (gelatin veronal buffered saline 8
plus calcium and magnesium, [Complement Technology]) at 4 °C and re-suspended in GVB++ at 9
5x107cells/mL. To sensitize chicken erythrocytes, a polyclonal anti-chicken RBC antibody 10
(Rockland Immunochemicals) was added to the cells (150 µg/mL) and incubated for 15 min on ice. 11
After washing with GVB++ once, the cells were re-suspended in GVB++ to a final volume of 3.6 12
mL. 13
Antibodies were serially diluted in neat normal human serum (NHS) at a ratio of 1:2 to final 14
concentrations ranging from 300 µg/mL to 2.343 µg/mL then 20 µL of each antibody/serum sample 15
was diluted in 80 µL of GVBS and incubated at room temperature for 20 min. Sensitized cRBCs 16
were added to the antibody/serum mixture at 30 µL per well (2.5x106 cells) and incubated at 37 °C 17
for 30 min. The plates were centrifuged at 1800 ×g for 3 min and 75 µL of the supernatant was 18
transferred to a new flat-bottom 96-well plate. The absorbance was measured at 415 nm. Samples 19
containing serum without anti-C5 antibodies with or without 10 mM EDTA were used as no lysis or 20
complete lysis controls, respectively. Sample conditions were run in triplicate and each sample was 21
tested twice in the same assay. 22
Structure of the C5‐eculizumab complex
9
Results1
Structure of the Fab-C5 complex 2
In order to obtain structural insight into how eculizumab binds C5 and prevents its activation by 3
complement convertases we generated a Fab fragment with the same sequence as found in 4
eculizumab and crystallized the Fab:C5 complex. After screening more than 200 crystals, we 5
obtained X-ray diffraction data extending to a maximum resolution of 4.2 Å (Table I). We found 6
two copies of the complex in the asymmetric unit that turned out to have virtually identical 7
structures (Supplementary Table 1). Here we describe the complex containing chain B (C5) and 8
chains H and L (Fab) (Fig. 1C). The electron density of the MG1 and MG3 and especially the MG4 9
and MG5 domains and the C-terminal C345c domain of C5 are better in this complex compared to 10
the second complex but comparable at the interface between C5 and the Fab (Supplementary figure 11
1). Due to the limited resolution of the data, fine details concerning intermolecular interactions 12
cannot be derived, but the structure unambiguously reveals the regions of C5 and the Fab engaged 13
in contacts. We consider the main chain tracing at the intermolecular interface to be reliable, 14
whereas the direction of the side chains may differ in a high resolution structure (Fig. 1D and 15
Supplementary Fig. 2A). In addition we cannot account for water mediated interactions. Thus 16
intermolecular interactions described in the following should be considered putative. 17
The complex is highly non-globular with the Fab fragment extending in an almost orthogonal 18
manner from the MG7 domain as compared to the major axis of C5 (Fig. 1C). The conformation of 19
C5 is quite similar to that of C5 in complex with cobra venom factor (CVF), with the C-terminal 20
C345c domain stably associated with the C5 MG7 and MG8 domains. Comparison of the Fab-21
bound C5 with CVF-bound (26) and unbound C5 (27) indicate rotations of the CUB and C5d 22
domains relative to the MG1-MG6 domains of 4o for Fab-bound C5 relative to CVF-bound and 11o 23
relative to unbound C5. Compared to our previous C5 containing structures, we observe a change in 24
Structure of the C5‐eculizumab complex
10
the scissile bond region (26, 28), the linker between the MG6 and MG7 domains and we have also 1
corrected a small register error in the first strand of the CUB domain. The same features are 2
observed in the C5:RaCI:OmCI complexes (Rhipicephalus appendiculatus C5 Inhibitor; RaCI, 3
Ornithodoros moubata complement inhibitor; OmCI) (29). Hence, these changes are unlikely to be 4
caused by Fab binding. At the Fab:C5 interface (Fig. 1C), the variable domains of both heavy and 5
light chain exclusively interact with the MG7 domain of C5. Although caution must be taken due to 6
the low resolution, analysis with PISA (30) suggests that roughly 1800 Å2 is buried in the interface 7
with the heavy chain and light chain accounting for 65 % and 35 % of the binding surface area, 8
respectively. 9
The C5 residues recognized by the Fab are all part of the antiparallel four-stranded β-sheet in the 10
MG7 domain. Residues 851-858 and 882-888 are found in two antiparallel β-strands held together 11
by a disulfide bridge while a third region (residue 915-920) connects two β-strands in a hairpin loop 12
(Fig. 2A and B). Except for light chain CDR2, all CDR regions are engaged in interaction with C5 13
with the most prominent interactions being formed by H-CDR3 (Fig. 2C-F). Residues 101-107 of 14
H-CDR3 together with residues 30-32 of L-CDR1 and Leu92 in L-CDR3 form a binding pocket for 15
C5 Trp917 and Phe918 (Fig. 2D). This explains the exquisite specificity of eculizumab for human 16
C5, since a Trp917Ser substitution is found in most other species (Fig. 2B and Supplementary Fig. 17
3). H-CDR3 is also engaged in forming the binding pocket for C5 Arg885 (Fig. 2E). Our structure 18
suggests that the resistance to eculizumab in PNH patients with Arg885His/Cys polymorphisms 19
(17), can be explained by a disruption of the C5-eculizumab interface due to the histidine/cysteine 20
being too small to fill the arginine binding pocket. This most likely causes the extended 21
conformation of H-CDR3 to change significantly with a concomitant loss of affinity. Moreover, as 22
Arg885 is spatially close to the disulfide bridge C856-C883 (Fig. 2A), the Arg885Cys 23
Structure of the C5‐eculizumab complex
11
polymorphism could result in an altered disulfide bridge pattern, potentially perturbing the integrity 1
of MG7. 2
Validation of the Fab-C5 structure 3
In an orthogonal evaluation of the contribution of CDR residues to the C5 binding, we created 66 4
single point mutants in which each CDR residue had been mutated to histidine allowing us to 5
selectively study the effect of a positive (pH 6.0) or a neutral side chain (pH 7.4) mutation on the 6
dissociation of the C5:antibody complex. A total of 28 of the 66 point mutants showed a significant 7
effect on association, dissociation or both (Fig. 3A-C). All of the important mutations increased 8
dissociation at pH 6.0, while 60 % of them also had an effect on the dissociation at pH 7.4 9
(dissociation of Asp110 from H-CDR3 could not be assessed as its association with C5 was highly 10
compromised). Precise determination of the binding kinetics was determined for a subset of the 11
most important mutant antibodies using SPR, and the results are in agreement with the findings 12
from the BLI assays (Fig. 3D and Table II). EHL000bound to C5 with an affinity of 17.6 pM. This 13
affinity decreased up to 7700-fold for the tested mutant antibodies ranging from 0.443 nM for 14
Thr94His of L-CDR3 to 136 nM for Trp107His of H-CDR3 (Table II). Both the association and 15
dissociation rate constants determined by SPR were in good agreement with the more qualitative 16
BLI data and we conclude that increases in the dissociation rates are the primary contributor to the 17
loss of affinityrather than a decrease in association rates (Fig. 3A, B and D and Table II). It is worth 18
noting that the mutants seem to fall into two classes as observed from the SPR curves. The first 19
class behaves somewhat similar to the WT antibody and includes the Glu59His mutant of H-CDR2 20
and the Thr94His mutant of L-CDR3. These antibodies have slower on-rates and markedly slower 21
off-rates as compared to the second group, including the Trp33His mutant of H-CDR1, the 22
Phe101His and Trp107His mutants of H-CDR3 and the Ala32His mutant of L-CDR1. For the same 23
subset of antibody mutants tested using SPR, the impact of loss of binding affinity on the functional 24
Structure of the C5‐eculizumab complex
12
inhibition of terminal complement was assessed in a chicken red blood cell hemolysis assay at pH 1
7.4 (Fig. 3E). These results clearly show that the effects on inhibitory activity are in agreement with 2
the effect each mutation has on binding to C5, where mutations that significantly affect binding 3
(heavy chain Trp107His, Phe101His, Trp33His and light chain Ala32His) severely limit the ability 4
to inhibit hemolysis. Mutations conferring smaller effects on binding kinetics (heavy chain 5
Glu59His and light chain Thr94His) also have a less pronounced impact on function in the 6
hemolysis assay. 7
In order to rationalize the effect of these mutations we inspected the surroundings of the affected 8
residues in our structure (Fig. 3C and Supplementary Table II). Trp33 (H-CDR1), Tyr99 (H-CDR3) 9
and Trp107 (H-CDR3) are prominent examples of affected residues in direct contact with C5. We 10
predict that mutation of these residues to histidine causes a disruption of the binding pocket 11
mentioned above accommodating Arg885 of C5 (Fig 2E). Furthermore, a positive charge of these 12
residues at pH 6.0 might result in electrostatic repulsion of the arginine. Histidine substitution of the 13
non-interacting Pro95 (L-CDR3) might cause a conformational change in the main chain, 14
perturbing the C5-interacting residues Thr94 and Leu96 (L-CDR3) and the neighboring Glu59 (H-15
CDR2), and a positive charge at the proline position might disrupt the potential electrostatic 16
interaction between Glu59 (H-CDR2) and C5 Lys887 (Fig. 2F). The same effect might occur upon 17
Thr94His (L-CDR3) substitution. A second residue not in direct contact with C5 is Glu50 (H-18
CDR2), however Glu50 appears to electrostatically stabilize Arg885 of C5 suggesting that a 19
positively charged histidine substitution might cause repulsion of the C5 arginine (Fig. 2E). 20
Substitution of Asp110 from H-CDR3 severely impacted the antibodies ability to associate with C5, 21
however this residue is at least 10 Å from the nearest C5 atom. We suggest that the effect of 22
mutating Asp110 is indirect and caused by destabilization of Arg98 at the N-terminal end of H-23
CDR3, which according to our structure is packed between Tyr27 and Tyr32 from H-CDR1 24
Structure of the C5‐eculizumab complex
13
(Supplementary Fig. 2B). This is proposed to affect the conformation of H-CDR3 and possibly also 1
H-CDR1 causing an effect upon association with C5. Overall the effects of the mutations are in 2
excellent agreement with the Fab-C5 interface observed in our crystal structure. Even for residues 3
not involved in direct contacts with C5, our structure still provides plausible explanations for the 4
observed effects (Supplementary Table II). 5
Structure of the C5‐eculizumab complex
14
Discussion1
Eculizumab is known to function by preventing cleavage of C5 by the complement C5 convertases, 2
however in our structure no residue in the Fab comes closer than 30 Å to the scissile bond region 3
(Fig. 1C). This suggests that eculizumab functions by sterically hindering the convertase from 4
associating with C5 but not by directly preventing access to the scissile bond. The primary non-5
catalytic subunit of the C5 convertases (C3b in the alternative pathway and C4b in the classical 6
pathway) are believed to interact with their substrates through a two point interaction similar to that 7
observed in the C5:CVF complex (26) where CVF serves as a model for the non-catalytic subunit of 8
the C5 convertase (Fig. 1B). The MG4 and MG5 domains in C5 may interact with the same 9
domains in C3b/C4b while the C5 MG7 domain is likely to contact the MG6 and MG7 domains of 10
C3b/C4b (Fig. 4A). A comparison of the interaction sites for CVF and the Fab on the C5 MG7 11
domain reveals significant overlap of the two binding sites (Fig. 4B-C). This further strengthens the 12
idea that eculizumab prevents the interaction of C5 with the convertase by directly competing for 13
binding to the MG7 domain and possibly indirectly by averting the MG4 and MG5 recognition by 14
the convertase. It has very recently been suggested that the role of the additional/nearby C3b in the 15
C5 convertase as compared to the C3 convertase is to prime the conformation of C5 such that it can 16
bind a nearby C3 convertase rather than inducing a particular conformation in the C5 convertase 17
with high affinity for the substrate (29). Both models are equally compatible with our suggestion 18
that steric hindrance is the predominant mechanism through which eculizumab inhibits C5 19
cleavage. 20
This mechanism of inhibition based on steric hindrance is shared with the bacterial C5 inhibitor 21
SSL7. Although SSL7 binds to the MG1, MG5 and MG6 domains in C5 quite far from the putative 22
convertase interacting residues in the MG4, MG5 and MG7 domains, SSL7 elegantly hijacks IgA 23
into a ternary IgA-SSL7-C5 complex and in this way introduce steric hindrance for convertase 24
Structure of the C5‐eculizumab complex
15
binding (28). The tick protein OmCI also prevents C5 cleavage by the convertases and based on 1
small angle X-ray scattering data, we previously suggested OmCI to bind near the C345c domain of 2
C5, far from the scissile bond (27). This is now confirmed by the crystal structures of the C5-3
OmCI-RaCI complexes (29). Interestingly, these structures also reveal that three different RaCI 4
inhibitors all bind in the same pocket between the C5d, MG2 and MG1 domains, again very far 5
from the scissile bond region. They might function by locking C5 into a conformation that prevents 6
the conformational change needed to allow the C5 scissile bond region to unfold and insert into the 7
catalytic site of the catalytic subunit of the convertase. OmCI appears to work differently as 8
compared to the RaCI inhibitors as it does not prevent inhibit C5 cleavage by CVFBb, but this may 9
relate to the positional stabilization of the C5 C345c domain upon OmCI binding (27, 29). Perhaps 10
flexibility of this domain is required for cleavage by the endogenous C5 convertases but not by 11
CVFBb. 12
In conclusion, our structure provides an explanation of both the exquisite selectivity of eculizumab 13
for human C5 and the mechanism of eculizumab resistance observed in a small subset of PNH 14
patients with polymorphisms encoding mutations at Arg885. With further refinement, this model 15
may provide structure-based insights to guide engineering efforts to make even more efficient 16
versions of eculizumab by pinpointing the key residues that interact with C5. 17
Acknowledgements: We would like to thank the beamline staffs at ESRF for support during data 18
collection and Tristan Croll for advice on model building. 19
Structure of the C5‐eculizumab complex
16
Legends to figures 1
FIGURE 1. Complement activation and the Fab:C5 structure. (A) The flow of the complement 2
system leading to activation of the terminal pathway, which is inhibited by eculizumab. Upon 3
recognition of pathogen-associated molecular patterns or danger-associated molecular patterns the 4
complement system is activated. This leads to the formation of the C3 and C5 convertases 5
ultimately resulting in the release of the anaphylatoxin C5a and the assembly of the membrane 6
attack complex through C5b. (B) Schematic representation of C5 convertases. Left part: the 7
endogenous heterotrimeric surface-anchored C5 convertases consisting of the non-catalytic subunit 8
C3b/C4b (dark green) bound to the catalytic subunit Bb/C2a (brown) and a second regulatory C3b 9
molecule (light green). Right part: A fluid phase C5 convertase can be formed with cobra venom 10
factor (CVF, blue) as the substrate binding subunit bound to the catalytic subunit Bb (brown) (C) 11
Cartoon representation of the Fab:C5 complex. The MG7 domain is shown in salmon color, the 12
anaphylatoxin in red and the heavy and light chain of the Fab in orange and green, respectively. (D) 13
Upper part: Omit non-averaged 2mFo-DFc electron density map contoured at 1σ around the H-14
CDR3 loop and the surrounding C5 MG7 domain and Fab light chain. H-CDR3 (Ala97-Val111) 15
was omitted prior to map calculation. Lower part: Omit non-averaged 2mFo-DFc electron density 16
map contoured at 1σ around the L-CDR3 loop and the surrounding C5 MG7 domain and Fab heavy 17
chain. L-CDR3 (Gln89-Thr97) was omitted prior to map calculation. 18
FIGURE 2. The C5 MG7 domain and its interaction with the Fab-fragment of eculizumab. (A) 19
Cartoon representation of the C5 MG7 domain with the residues comprising the epitope shown in 20
sticks. (B) Alignment of mammalian C5 MG7 domain sequences with numbering according to the 21
human sequence and secondary structure elements depicted at the top. Residues marked in black 22
differ from the human sequence. Light blue squares indicate C5 residues within 3.8 Å of the Fab 23
according to the structure. (C) Primary structure of eculizumab heavy and light chains variable 24
Structure of the C5‐eculizumab complex
17
domains. CDR regions are highlighted in grey and secondary structure is depicted on top. Residues 1
marked with a green star are within 3.8 Å of C5 according to the structure. Blue triangles depict 2
residues presented in Fig. 3A more than 3.8 Å away from C5. (D-F) Details of the Fab:C5 interface 3
shown from the perspective of different C5 residues (grey C-atoms) with heavy chain C-atoms 4
colored orange and light chain C-atoms colored green. 5
FIGURE 3. Validation of the Fab:C5 structure through functional and binding studies. (A). All 6
residues in eculizumab CDR regions were independently replaced with histidine and the binding of 7
C5 to each antibody bearing a single point mutation was investigated with Bio-Layer 8
Interferometry. Antibodies were immobilized on a protein A tip and association of C5 was 9
measured at pH 7.4 while complex dissociation was measured at both pH 7.4 and pH 6.0. 10
Association kinetics (average of 2 sample runs) are reported as % of the mean Rmax value for the 11
control antibody EHL000 (average of 16 runs). Relative dissociation kinetics are reported as % 12
dissociation at 600 seconds (average of 2 sample runs). White shading indicate that the single point 13
mutant antibody has association and dissociation kinetics within the ranges observed for the 14
parental antibody EHL000. The three shades of gray indicate effects on the association/dissociation 15
rates relative to the parental antibody divided roughly into thirds, with darker shading indicating 16
more severe effects. Dissociation kinetics for Asp110His could not be determined as the association 17
signal was so weak (ND: dotted fill). Data for single amino acid substitutions without significant 18
effects are not shown. (B) Representative BLI curves of a subset of histidine mutant antibodies 19
binding to C5. The association was performed at pH 7.4 whereas dissociation was performed at pH 20
6.0. Residues boxed in orange and green belong to the heavy and light chain, respectively. (C) 21
Ribbon representation of the CDR regions. Cα positions of residues with significant altered 22
dissociation rates upon mutation to histidine at pH 7.4 (panel A) are shown in spheres. The larger 23
the sphere the more significant is the effect. (D) Representative SPR curves for a subset of 24
Structure of the C5‐eculizumab complex
18
immobilized histidine mutant antibodies binding to C5. Residues are colored and as in panel B. (E) 1
Inhibition of complement classical pathway (CCP) hemolytic activity on sensitized chicken 2
erythrocytes by anti-C5 mAbs in NHS. Data are shown as mean ± S.D. of triplicate run of one 3
assay. Residues are boxed as in Fig. 3D. 4
5
FIGURE 4. Overlap of the CVF and eculizumab interfaces of human C5 MG7 domain. (A) Surface 6
representation of C5 with the MG7 domain colored in salmon. Residues within 3.8 Å of CVF in the C5:CVF 7
complex are shown in blue. (B) Zoom on the MG7 domain of C5 framed by the square in panel A. (C) As 8
panel B but with C5 residues within 3.8 Å of Fab heavy and light chain colored orange and green, 9
respectively. Residues present in both the CVF and Fab interfaces are marked. 10
Structure of the C5‐eculizumab complex
19
References 1
1. Takata, Y., T. Kinoshita, H. Kozono, J. Takeda, E. Tanaka, K. Hong, and K. Inoue. 1987. Covalent 2 association of C3b with C4b within C5 convertase of the classical complement pathway. The Journal 3 of experimental medicine 165: 1494‐1507. 4
2. Medicus, R. G., O. Gotze, and H. J. Muller‐Eberhard. 1976. Alternative pathway of complement: 5 recruitment of precursor properdin by the labile C3/C5 convertase and the potentiation of the 6 pathway. The Journal of experimental medicine 144: 1076‐1093. 7
3. Kinoshita, T., Y. Takata, H. Kozono, J. Takeda, K. S. Hong, and K. Inoue. 1988. C5 convertase of the 8 alternative complement pathway: covalent linkage between two C3b molecules within the 9 trimolecular complex enzyme. J Immunol 141: 3895‐3901. 10
4. Klos, A., E. Wende, K. J. Wareham, and P. N. Monk. 2013. International Union of Basic and Clinical 11 Pharmacology. [corrected]. LXXXVII. Complement peptide C5a, C4a, and C3a receptors. Pharmacol 12 Rev 65: 500‐543. 13
5. Gotze, O., and H. J. Muller‐Eberhard. 1970. Lysis of erythrocytes by complement in the absence of 14 antibody. The Journal of experimental medicine 132: 898‐915. 15
6. Lachmann, P. J., and R. A. Thompson. 1970. Reactive lysis: the complement‐mediated lysis of 16 unsensitized cells. II. The characterization of activated reactor as C56 and the participation of C8 17 and C9. The Journal of experimental medicine 131: 643‐657. 18
7. Moskovich, O., and Z. Fishelson. 2007. Live cell imaging of outward and inward vesiculation induced 19 by the complement c5b‐9 complex. The Journal of biological chemistry 282: 29977‐29986. 20
8. Podack, E. R., W. P. Kolb, and H. J. Muller‐Eberhard. 1978. The C5b‐6 complex: formation, isolation, 21 and inhibition of its activity by lipoprotein and the S‐protein of human serum. J Immunol 120: 1841‐22 1848. 23
9. Tschopp, J., and L. E. French. 1994. Clusterin: modulation of complement function. Clinical and 24 experimental immunology 97 Suppl 2: 11‐14. 25
10. Heinen, S., A. Hartmann, N. Lauer, U. Wiehl, H. M. Dahse, S. Schirmer, K. Gropp, T. Enghardt, R. 26 Wallich, S. Halbich, M. Mihlan, U. Schlotzer‐Schrehardt, P. F. Zipfel, and C. Skerka. 2009. Factor H‐27 related protein 1 (CFHR‐1) inhibits complement C5 convertase activity and terminal complex 28 formation. Blood 114: 2439‐2447. 29
11. Rollins, S. A., and P. J. Sims. 1990. The complement‐inhibitory activity of CD59 resides in its capacity 30 to block incorporation of C9 into membrane C5b‐9. J Immunol 144: 3478‐3483. 31
12. Bessler, M., P. J. Mason, P. Hillmen, T. Miyata, N. Yamada, J. Takeda, L. Luzzatto, and T. Kinoshita. 32 1994. Paroxysmal nocturnal haemoglobinuria (PNH) is caused by somatic mutations in the PIG‐A 33 gene. The EMBO journal 13: 110‐117. 34
13. Takeda, J., T. Miyata, K. Kawagoe, Y. Iida, Y. Endo, T. Fujita, M. Takahashi, T. Kitani, and T. Kinoshita. 35 1993. Deficiency of the GPI anchor caused by a somatic mutation of the PIG‐A gene in paroxysmal 36 nocturnal hemoglobinuria. Cell 73: 703‐711. 37
14. Noris, M., and G. Remuzzi. 2009. Atypical Hemolytic–Uremic Syndrome. New England Journal of 38 Medicine 361: 1676‐1687. 39
15. Thomas, T. C., S. A. Rollins, R. P. Rother, M. A. Giannoni, S. L. Hartman, E. A. Elliott, S. H. Nye, L. A. 40 Matis, S. P. Squinto, and M. J. Evans. 1996. Inhibition of complement activity by humanized anti‐C5 41 antibody and single‐chain Fv. Molecular Immunology 33: 1389‐1401. 42
16. Evans, M. J., L. A. Matis, E. E. Mueller, S. H. Nye, S. Rollins, R. P. Rother, J. P. Springborn, S. P. 43 Squinto, T. C. Thomas, and J. A. Wilkins. 2002. C5‐specific antibodies for the treatment of 44 inflammatory diseases. US Patents. 45
17. Nishimura, J., M. Yamamoto, S. Hayashi, K. Ohyashiki, K. Ando, A. L. Brodsky, H. Noji, K. Kitamura, T. 46 Eto, T. Takahashi, M. Masuko, T. Matsumoto, Y. Wano, T. Shichishima, H. Shibayama, M. Hase, L. Li, 47 K. Johnson, A. Lazarowski, P. Tamburini, J. Inazawa, T. Kinoshita, and Y. Kanakura. 2014. Genetic 48
Structure of the C5‐eculizumab complex
20
variants in C5 and poor response to eculizumab. The New England journal of medicine 370: 632‐1 639. 2
18. Sottrup‐Jensen, L., and G. R. Andersen. 2014. Purification of human complement protein C5. 3 Methods in molecular biology 1100: 93‐102. 4
19. Brodersen, D. E., G. R. Andersen, and C. B. Andersen. 2013. Mimer: an automated spreadsheet‐5 based crystallization screening system. Acta crystallographica. Section F, Structural biology and 6 crystallization communications 69: 815‐820. 7
20. Kabsch, W. 2010. Xds. Acta crystallographica. Section D, Biological crystallography 66: 125‐132. 8 21. McCoy, A. J., R. W. Grosse‐Kunstleve, L. C. Storoni, and R. J. Read. 2005. Likelihood‐enhanced fast 9
translation functions. Acta crystallographica. Section D, Biological crystallography 61: 458‐464. 10 22. Jones, T. A., J. Y. Zou, S. W. Cowan, and M. Kjeldgaard. 1991. Improved methods for building 11
protein models in electron density maps and the location of errors in these models. Acta 12 crystallographica. Section A, Foundations of crystallography 47 ( Pt 2): 110‐119. 13
23. Emsley, P., B. Lohkamp, W. G. Scott, and K. Cowtan. 2010. Features and development of Coot. Acta 14 crystallographica. Section D, Biological crystallography 66: 486‐501. 15
24. Adams, P. D., P. V. Afonine, G. Bunkoczi, V. B. Chen, I. W. Davis, N. Echols, J. J. Headd, L. W. Hung, 16 G. J. Kapral, R. W. Grosse‐Kunstleve, A. J. McCoy, N. W. Moriarty, R. Oeffner, R. J. Read, D. C. 17 Richardson, J. S. Richardson, T. C. Terwilliger, and P. H. Zwart. 2010. PHENIX: a comprehensive 18 Python‐based system for macromolecular structure solution. Acta crystallographica. Section D, 19 Biological crystallography 66: 213‐221. 20
25. Chen, V. B., W. B. Arendall, III, J. J. Headd, D. A. Keedy, R. M. Immormino, G. J. Kapral, L. W. Murray, 21 J. S. Richardson, and D. C. Richardson. 2010. MolProbity: all‐atom structure validation for 22 macromolecular crystallography. Acta Crystallographica Section D 66: 12‐21. 23
26. Laursen, N. S., K. R. Andersen, I. Braren, E. Spillner, L. Sottrup‐Jensen, and G. R. Andersen. 2011. 24 Substrate recognition by complement convertases revealed in the C5‐cobra venom factor complex. 25 The EMBO journal 30: 606‐616. 26
27. Fredslund, F., N. S. Laursen, P. Roversi, L. Jenner, C. L. Oliveira, J. S. Pedersen, M. A. Nunn, S. M. Lea, 27 R. Discipio, L. Sottrup‐Jensen, and G. R. Andersen. 2008. Structure of and influence of a tick 28 complement inhibitor on human complement component 5. Nat Immunol 9: 753‐760. 29
28. Laursen, N. S., N. Gordon, S. Hermans, N. Lorenz, N. Jackson, B. Wines, E. Spillner, J. B. Christensen, 30 M. Jensen, F. Fredslund, M. Bjerre, L. Sottrup‐Jensen, J. D. Fraser, and G. R. Andersen. 2010. 31 Structural basis for inhibition of complement C5 by the SSL7 protein from Staphylococcus aureus. 32 Proceedings of the National Academy of Sciences of the United States of America 107: 3681‐3686. 33
29. Jore, M. M., S. Johnson, D. Sheppard, N. M. Barber, Y. I. Li, M. A. Nunn, H. Elmlund, and S. M. Lea. 34 2016. Structural basis for therapeutic inhibition of complement C5. Nat Struct Mol Biol. 35
30. Krissinel, E., and K. Henrick. 2007. Inference of Macromolecular Assemblies from Crystalline State. 36 Journal of Molecular Biology 372: 774‐797. 37
38
39
A
CEcu H-chain
Ecu L-chain
C5a
MG8 MG7
C5 C5
180o
C5d
MG4MG5
MG1
C345c
MG6
H-CDR3
C5 MG7
>30Å
H-CDR2
C5 MG7
L-CDR3
L-CDR2
Bb/C2a:C3b/C4b:C3bBb:CVF
PAMPsDAMPs
C3 convertase
C5 convertase
C5a
MAC
Eculizumab
H-H-H-H-H-H-H-HH-H-H-H-H-H-HHH-H-HH-H-H-H-H-HH-H-HH-HHHHHH-HHHHHHHHHHH CDCCCCCCCCC R2
CCCCCCCCCCCCCCCCCCCCCC5 55 5 55 55 55 555555555555555CCCCCCCCCCCCCCCCCCCCCCCCC MG7
L-CDR3RRR33333333LLLLL
B D
MG7
Q V Q L V Q S G A E V K K P G A S V K V S C K A S G Y I F S N Y W I Q W V R Q A P G
Q G L E WM G E I L P G S G S T E Y T E N F K D R V T M T R D T S T S T V Y M E L S
S L R S E D T A V Y Y C A R Y F F G S S P N W Y F D V W G Q G T L V
1 10 20 30 40
50 60 70 80
90 100 110
D I Q M T Q S P S S L S A S V G D R V T I T C G A S E N I Y G A L N W Y Q Q K P G K
A P K L L I Y G A T N L A D G V P S R F S G S G S G T D F T L T I S S L Q P E D F A
T Y Y C Q N V L N T P L T F G Q G T K V E I K R
1 10 20 30 40
50 60 70 80
90 100
AK882
K920
K853V884
R885
K887
S851
F918
W917
V91
L96
T94 K887
E59
W107
W33W107
Y99
F101
R885E50
W33
L52
Nt
Y30
F918
W917
W107
V91A32
L92
C
CDR2
CDR3
D
CDR3
CDR2
CDR1 CDR1
Heavy chain Light chain
C5 MG7 domainB
E F
HumanChimpanzeeSumatran_OrangutanRatMouse
HumanChimpanzeeSumatran_OrangutanRatMouse
HumanChimpanzeeSumatran_OrangutanRatMouse
D V F L E M N I P Y S V V R G E Q I Q L K G T V Y N Y R T S G M Q F C V K M
S A V E G I C T S . E S P V I D H Q G T K S S K C V R Q K V E G S S S H L V
T F T V L P L E I G L H N I N F S L E T W F G K E I L V K T L R V V
D V F L E M N I P Y S V V R G E Q I Q L K G T V Y N Y R T S G M Q F C V K M
S A V E G I C T S . E S P V I D H Q G T K S S K C V R Q K V E G S S S H L V
T F T V L P L G I G L H N I N F S L E T S F G K E I L V K T L R V V
D V F V E M N I P Y S V V R G E Q I Q L K G T V Y N Y R T S G M Q F C V K M
S A V E G I C T S . E S P A I D H Q G T K S S K C V H R K V E G S S S H L V
T F T V L P L E I G L H N I S F S L E T S F G K E I L V K T L R V V
D V F L E M N I P Y S V V R G E Q I Q L K G T V Y N Y R T S G T M F C V K M
S A V E G I C T P . G S S A A S P Q T S R S S R C V R Q R I E G S S S H L V
T F S L L P L E I G L H S I N F S L E T S F G K E I L V K T L R V V
E V F L E M N I P Y S V V R G E Q I Q L K G T V Y N Y M T S G T K F C V K M
S A V E G I C T S . G S S A A S L H T S R P S R C V F Q R I E G S S S H L V
T F T L L P L E I G L H S I N F S L E T S F G K D I L V K T L R V V
830 840 850
860 870 880 890
900 910 920 930
F101
W107
A32
W33
E50
Q35
T94 E59
A
C
CDR A.A. % Rmax Assoc. (7.4)
% Dissoc (pH 7.4)
% Dissoc. (pH 6.0)
L-CDR1
E27 100 6 13 G31 100 8 39 A32 37 100 100 L33 70 52 91 N34 82 23 63
L-CDR2 G50 71 34 83
L-CDR3
V91 84 13 28 L92 90 25 77 N93 98 5 17 T94 98 32 83 P95 105 16 18 L96 131 15 31
H-CDR1
Y27 113 9 10 Y32 114 5 22 W33 28 100 100 I34 112 6 16 Q35 63 87 98
H-CDR2
E50 78 74 36 L52 113 9 16 P53 92 10 15 S57 126 5 10 E59 126 47 81
H-CDR3
Y99 90 38 94 F101 21 100 100 P105 89 43 46 W107 15 100 100 F109 101 26 43 D110 28 ND ND
W107HF101HW33HA32HE59HT94HEHL000%
lysi
s
IgG (μg/mL)1 10 100 1000
-20
0
20
40
60
80
100
120
Time (s)
nm
200 300 400 500 600 700 800 900 1000 1100 12000
5
10
15
20
25
30
35
Time (s)
Resp
. Diff
.
D E
B
400 600 800 1000 1200 1400 1600 18000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
anti-C5 IgG binding to hC5 with dissociation at pH 6.0
anti-C5 IgG binding to hC5 at pH 7.4
EHL000W107HF101HW33HE59H
A32HT94H
MG7
MG4 MG5
C883
R885
E915
W917
K920
A B C
Supplementary material: Structure of the C5‐eculizumab complex
1 1
Supplementary material: Structure of the C5‐eculizumab complex
2
FIGURE S1. Electron density from both Fab:C5 complexes in the asymmetric unit. Non-averaged 1
2mFo-DFc electron density map contoured at 1σ is shown around different domains of the Fab:C5 2
complex. Heavy and light chain are shown in orange and green, respectively, the MG7 domain of 3
C5 is shown in salmon and the rest of the C5 domains are shown in grey. Similar domains from the 4
two complexes are viewed in the same orientation. 5
Supplementary material: Structure of the C5‐eculizumab complex
3
1 FIGURE S2. Stereo representation of the Fab-C5 interface. A) Non-averaged 2mFo-DFc electron 2
density map contoured at 1σ around the para- and epitope of the Fab:C5 complex. Heavy and light 3
chain CDR loops are shown in orange and green, respectively and the MG7 domain of C5 is shown 4
in salmon. B) Cartoon representation of the Fab:C5 interface shown in stereo view, with important 5
residues shown in sticks. The N-terminal Cα atom of the heavy chain is shown in spheres. The color 6
scheme is similar to panel A. 7
Supplementary material: Structure of the C5‐eculizumab complex
4
1 FIGURE S3. Complete sequence alignment of the MG7 domain from complement C5. The 2
alignment is made as in Fig. 2B but with additional sequences. The sequences have the following 3
entries: Human (BAD92268.1), Western Lowland Gorilla (XP_004048612.1), Chimpanzee 4
(XP_520228.2), Bonobo (XP_003833147.1), White cheeked Gibbon (XP_003264119.1), Rhesus 5
Macaque (XP_001095750.2), Cynomolgus Macaque (EHH57109.1), Olive Baboon 6
(XP_003911347.1), Sumatran Orangutan (XP_002820208.1), Squirrel Monkey (XP_003925224.1), 7
Marmoset (XP_002743307.1), Dog (XP_532046.3), Guinea Pig (XP_003479183.1), Rat 8
(XP_345343.4), Mouse (NP_034536.1), Naked Mole Rat (EHB01835.1), Cat (XP_003995868.1), 9
Sheep (XP_004004015.1), Cow (NP_001160088.1), Yak (ELR60902.1), Wild Boar 10
(NP_001001646.1), African Elephant (XP_003407821.1) and Giant Panda (XP_002916001.1). 11
Supplementary material: Structure of the C5‐eculizumab complex
5
Chain CDR Res. Explanation
Glu27 A positive charge might disrupt the potential interaction of C5 Ser851 with L-CDR3 Asn93
Gly31 Could disturb main chain conformation losing affinity for C5 Trp917 CDR1 Ala32 Potentially disturbs the tight non-polar interaction with C5 Trp917 Leu33 The positive charge of His+ could disturb the structure of L-CDR3 locally Asn34 Putative loss of the potential hydrogen bond to H-CDR3 Trp107 main chain
L CDR2 Gly50 Might disturb main chain conformation potentially affecting both L-CDR1 and H-CDR3
Val91 Could disturb conformation of the crucial H-CDR3 Trp107 and positive charge possibly disturbs the binding pocket of C5 Arg885
Leu92 Most likely disturbs the tight non-polar interaction with C5 Trp917
Asn93 Positive charge might lead to electrostatic interaction with Glu27 in L-CDR1 thereby perturbing the nearby Leu92 participating in the binding pocket for C5 Trp917
CDR3 Thr94 Could disturb the potential electrostatic interaction between C5 Lys887 and H-CDR2 Glu59
Pro95 Potentially disturbs main chain conformation of L-CDR3 locally
Leu96 A positive charge might disturb residues of H-CDR3 forming the binding pocket of C5 Arg885
Tyr27 Positive charge could disrupt packing with H-CDR3 Arg98 perturbing the packing of H-CDR1 against H-CDR3
Tyr32 The effect is most likely similar to Tyr27. Moreover a positive charge might result in electrostatic repulsion of C5 Lys882
CDR1 Trp33 Would disrupt the binding pocket of C5 Arg885 and a positive charge might result in repulsion of the C5 arginine
Ile34 Positive charge could disrupt the interaction with other H-chain non-polar residues
Gln35 Might disrupt the conformation of the neighboring H-CDR2 Glu50 and H-CDR3 Tyr99 both potentially interacting with C5 Arg885
Glu50 Could perturb the stabilization of the positive charge from the C5 Arg885. A positive charge might stabilize the π-electrons of H-CDR1 Trp33 or induce changes in the surrounding electrostatic interactions
CDR2 Leu52 A positive charge might disrupt packing with other H-chain non-polar residues H Pro53 The effect is most likely similar to Leu52
Ser57 A positive charge could result in repulsion of C5 Lys887
Glu59 Most likely disrupts the potential interaction with C5 Lys887 and a positive charge would strengthen this and result in repulsion of the C5 lysine
Tyr99 Could disrupt the binding pocket of C5 Arg885 and a positive charge might result in repulsion of the C5 arginine
Phe101 Potentially disrupts the stacking with H-CDR3 Trp107
CDR3 Pro105 Might disturb the main chain conformation and thereby the structure of the apex of the H-CDR3 loop
Trp107 Would disrupt the binding pocket of C5 Arg885 and a positive charge might result in repulsion of the C5 arginine
Phe109 Could disrupt the stacking with H-CDR3 Tyr99
Asp110 Loss of negative charge most likely disturbs the conformation of H-CDR3 and H-CDR1, by disrupting the potential salt bridge with Arg98 of H-CDR3 which stacks against aromatic residues of H-CDR1
1
Supplementary Table I. Eculizumab histidine mutations affecting C5 binding. The overall location 2
of these residues is presented in Fig. 3C. 3