The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McGrath, F. D. G.
Right arrow Articles by Roos, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McGrath, F. D. G.
Right arrow Articles by Roos, A.
The Journal of Immunology, 2006, 176: 2950-2957.
Copyright © 2006 by The American Association of Immunologists

Evidence That Complement Protein C1q Interacts with C-Reactive Protein through Its Globular Head Region1

Fabian D. G. McGrath*, Mieke C. Brouwer*, Gérard J. Arlaud{dagger}, Mohamed R. Daha{ddagger}, C. Erik Hack2,* and Anja Roos3,{ddagger}

* Department of Immunopathology, Sanquin Research, Amsterdam, The Netherlands; {dagger} Institut de Biologie Structurale Jean-Pierre Ebel, Laboratoire d’Enzymologie Moléculaire, Grenoble, France; and {ddagger} Department of Nephrology, Leiden University Medical Center, Leiden, The Netherlands


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
C1q acts as the recognition unit of the first complement component, C1, and binds to immunoglobulins IgG and IgM, as well as to non-Ig ligands, such as C-reactive protein (CRP). IgG and IgM are recognized via the globular head regions of C1q (C1qGR), whereas CRP has been postulated to interact with the collagen-like region (C1qCLR). In the present study, we used a series of nine mAbs to C1q, five directed against C1qGR and four against C1qCLR, to inhibit the interaction of C1q with CRP. The F(ab')2 of each of the five mAbs directed against C1qGR inhibited binding of C1q to polymerized IgG. These five mAbs also successfully inhibited the interaction of C1q with CRP. Moreover, these five mAbs inhibited C1 activation by CRP as well as by polymerized IgG in vitro. In contrast, none of the four mAbs against C1qCLR inhibited C1q interaction with CRP or IgG, or could reduce activation of complement by CRP or polymerized IgG. These results provide the first evidence that the interaction of C1q with CRP or IgG involves sites located in the C1qGR, whereas sites in the CLR do not seem to be involved in the physiological interaction of C1q with CRP.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The classical pathway of complement becomes activated under inflammatory conditions such as immune complex diseases or ischemia/reperfusion injury (1, 2). The mechanism of activation in the latter condition involves natural Abs (3, 4) as well as the acute phase protein C-reactive protein (CRP)4 (5, 6, 7). These molecules bind to newly exposed epitopes or ligands present on ischemic and/or apoptotic cells (8, 9, 10). Excessive activation of the classical pathway induces additional tissue injury in ischemia/reperfusion injury (11). Hence, inhibitors that attenuate classical pathway activation by Igs or CRP provide novel therapeutic options in ischemia/reperfusion injury and possibly other conditions.

Activation of the classical pathway is triggered by binding of C1, the first component of the pathway, to an activator. C1 consists of a complex of one C1q, two C1r, and two C1s molecules in which C1q acts as the recognition subunit while C1r and C1s associate to form a calcium-dependent tetramer that functions as the catalytic subunit (1, 12). The C1q molecule has a molecular mass of 460 kDa, and is composed of three different polypeptide chains, A, B, and C, containing 223-, 226-, and 217-aa residues, respectively. Six copies of each chain are found per C1q molecule (13, 14). Overall, the molecule looks like a bouquet of six tulips, each tulip consisting of an A, B, and C chain. The N-terminal region of each chain consists of a sequence that is involved in the formation of A-B and C-C interchain disulphide bonds. This sequence is followed by a collagen-like sequence of ~81 aa (15), in which the A, B, and C chains form a heterotrimeric collagenous triple helix. This part is referred to as the C1q collagen-like region (C1qCLR), and makes up the stalk of each tulip. At their C terminus each C1q chain contains a globular head region consisting of ~135 residues. These domains of A-, B-, and C-chain together form the globular head regions (C1qGR) of the tulip, six of these being present per C1q molecule.

C1q can bind various activators such as immunoglobulins IgG (16) and IgM (17), as well as non-Ig proteins like CRP (7), SAP (18), PTX3 (19), and beta-amyloid fibrils (Abeta) (20). A common binding site on C1q for all these activators may serve as target for potential inhibitors. However, binding sites on C1q for the Fc domain of aggregated IgG or IgM have been mapped to the C1qGR (21, 22), whereas those for some other ligands were shown to be located within its CLR, as was first reported for the interaction with CRP (23). In particular, the sequences involving the amino acid residues 14–26 and 76–92 within the CLR of the A chain are thought be important for the interaction with CRP (23). Whether, in addition to these, common binding sites for ligands such as CRP and IgM or IgG exist on C1q, is not known. In a search for such common binding sites, we evaluated the effect of several mAbs against human C1q on the binding of C1q to IgG and CRP, and on the activation of complement by these activators. Our results indicate that both binding sites, involved in the interaction of C1q with either activator, are located on the globular domain.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Proteins

C1q was purified from recalcified frozen plasma according to Tenner et al. (24). The final preparation was dialyzed against PBS (pH 7.4). C1q concentration in the preparation was derived from its absorbance at 280 nm using E1 cm1% = 6.82. Purity of the preparation was assessed from 12% SDS-PAGE analysis (Invitrogen Life Technologies) using silver staining (Fig. 1). C1q was radiolabeled as described (125I-labeled C1q (125I-C1q)) (25).


Figure 1
View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 1. SDS-PAGE analysis of C1q and C1q fragments. Human C1q, C1qCLR, and C1qGH (2 µg/lane) were analyzed by SDS-PAGE (12%, reducing conditions), followed by silver staining. C1q shows a double band containing the A and B chains (31 and 30 kDa, respectively), and the C chain at 26 kDa. C1qCLR runs in a single band of ~16 kDa, and C1qGR shows three bands corresponding to the A, B, and C moieties (respectively at 19, 17, and 15 kDa).

 
C1qCLR were prepared by pepsin digestion of purified C1q (15). Briefly, C1q was incubated with pepsin (Sigma-Aldrich) at 30:1, w/w, for 3.5 h at 37°C in 0.2 M sodium acetate, 100 mM NaCl (pH 4.4), after which the reaction was stopped by addition of 1 M Tris (pH 9.0). Pepsin and residual intact C1q were removed by sucrose gradient ultracentrifugation. The 6.3S-peak fractions were pooled and dialyzed against PBS. Purity of the preparation was determined by analysis on 12% SDS-PAGE using silver staining (Fig. 1). The concentration of C1qCLR was determined from absorbance measurements at 280 nm using OD1 cm1% = 2.1. Finally, the preparation was stored in aliquots at –30°C.

C1qGR (1 mg/ml) was prepared by digestion of C1q purified from human serum with Achromobacter iophagus collagenase (Roche Applied Science) (enzyme/protein ratio = 1/5, w/w) for 24 h at 37°C in 250 mM NaCl, 5 mM CaCl2, 50 mM Tris-HCl (pH 7.4), in the presence of 10 µM E64 and 10 µM leupeptin (Roche Applied Science). The resulting globular domain was treated with Clostridium perfringens type X neuraminidase (Sigma-Aldrich) (0.2 U/mg) for 5 h at 25°C and then further incubated in the presence of 50 µM iodoacetamide for 1 h at 25°C. The material was dialyzed against 25 mM NaCl, 50 mM MES (pH 6.5), and loaded onto a 6-ml Resource S column (Amersham Biosciences) equilibrated in the same buffer. Purification was achieved by applying a 600-ml linear gradient from 25 to 80 mM NaCl in the same buffer (26). Purity was confirmed by SDS-PAGE analysis (Fig. 1).

rCRP was purchased from BiosPacific and native CRP was purified from human plasma obtained from Sigma-Aldrich. The purity of CRP preparations was confirmed using 12% SDS-PAGE and Coomassie staining (Fig. 2A) as well as by gel filtration. Gel filtration was performed using a 25-ml Superdex 200 column (Amersham) using veronal-buffered saline (VBS) containing 2 mM EDTA as a running buffer. For both preparations, a single peak was observed running between IgG and albumin (Fig. 2B), confirming the pentameric structure of both recombinant and native CRP.


Figure 2
View larger version (43K):
[in this window]
[in a new window]
  		FIGURE 2. SDS-PAGE and gel filtration analysis of CRP. A, rCRP and CRP from plasma were analyzed by SDS-PAGE (10 µg/lane, 10% gel, reducing and nonreducing conditions as indicated), followed by Coomassie blue staining. Monomeric CRP shows under both conditions a monomeric band at ~25 kDa. B, CRP (~100 µg) was run on a 25-ml Superdex 200 column. The complete protein profile is shown, demonstrating a single peak running between IgG and albumin (expected molecular mass is 125 kDa).

 
Aggregated human IgG (AHG) was prepared by heating purified human IgG (Sanquin Research) at 60 mg/ml in PBS for 20 min at 63°C. The preparation was then diluted to 10 mg/ml, and stored at –30°C.

Monoclonal Abs

Mouse mAbs to human C1q were raised using standard hybridoma technology (27). The following anti-C1q mAbs were used: anti-C1q-2, -42, -52, -85, -89, -124, -130, -171 (also known as mAb 2204; Ref.19) and anti-C1q-193. F(ab')2 were prepared by incubation of each mAb with pepsin (Sigma-Aldrich) with a mAb-pepsin ratio of 40:1 w/w, in 0.1 M sodium citrate (pH 3.5), at 37°C for 20 h. The reaction was stopped by adding 1 M Tris to adjust the pH to 8.5. Intact Fc domains were removed by adsorption onto protein A-Sepharose (Amersham Biosciences). The preparations were then dialyzed against PBS and stored at –30°C. The F(ab')2 preparations were examined on 12% SDS-PAGE with Coomassie staining. The fragments showed a band at 100 kDa, as expected, without detectable intact mAb (Fig. 3).


Figure 3
View larger version (102K):
[in this window]
[in a new window]
  		FIGURE 3. SDS-PAGE analysis of F(ab')2. F(ab')2 of anti-C1q mAbs (clones as indicated) were analyzed by SDS-PAGE (12% gel, nonreducing conditions), showing a band at 100 kDa, as well as some additional bands presumably representing incomplete F(ab')2.

 
ELISA to map binding sites of anti-C1q mAbs on the domains of C1q

Microtiter plates (Nunc Maxisorp; Nalge Nunc International) were coated with anti-C1q mAbs overnight at room temperature (5 µg/ml in 0.1 M carbonate buffer (pH 9.6), 100 µl/well). The plates were washed twice in PBS and blocked with 2%, w/v, BSA (BSA; Merck) in PBS, 150 µl/well for 30 min at room temperature. After incubation, the plates were washed five times in PBS containing 0.02%, v/v, Tween 20. This washing procedure was repeated after each of the subsequent incubations. C1q/C1qGR/C1qCLR preparations were diluted in hypertonic assay buffer (PBS containing 0.02%, w/v, Tween 20, 1%, w/v, gelatin, 0.5 M NaCl, 0.3%, w/v, BSA) and 100 µl of these dilutions were added to the appropriate wells. The plates were incubated again for 1 h, followed by incubations with biotinylated rabbit polyclonal anti-C1q Ab (Sanquin), and HRP-conjugated streptavidin (Amersham Biosciences), both diluted in hypertonic assay buffer. Finally, 100 µl of 3,3' 5,5'-tetra-methyl-benzidine, 100 µg/ml in 0.11 M sodium acetate (pH 5.5) containing 0.003%, v/v, H2O2, was added to each well and color development was observed. The reaction was stopped by addition of 100 µl of 2 M H2SO4 to each well. The OD of each well was then read at 450/540 nm and results were recorded.

Radioimmunoassays (RIA) to study the interaction of C1q with IgG or CRP

Human IgG was coupled to cyanogen bromide-activated-Sepharose (Amersham) according to manufacturer’s instructions. The IgG-Sepharose beads were then washed twice and resuspended in RIA buffer (VBS containing 5 mM CaCl2, 0.3% BSA, 0.1% Tween 20) to a concentration of 10 mg of Sepharose/ml. Fifty microliters of anti-C1q F(ab')2 (25 µg/ml in PBS) was preincubated with 50 µl of 125I-C1q (~10,000 cpm or 1 µg) for 1 h at room temperature. Then, 0.5 ml of IgG-Sepharose suspension was added and the tubes were incubated overnight by head-over-head rotation. Following incubation the Sepharose beads were washed four times in PBS, the supernatant was removed, and the radioactivity of the spun down Sepharose was read on a LKN Wallac Multigamma II (Wallac Oy). Results were expressed as the percent of input. All tests were done in duplicate.

The effect of anti-C1q mAbs on the interaction of C1q with CRP was tested in a similar way. Immobilized p-aminophenyl-phosphorylcholine coupled to Sepharose (PC-Sepharose) was purchased from Pierce. The binding capacity of this Sepharose was 2.5–5.5 mg of CRP/ml gel. Before use, 0.5 ml of PC-Sepharose was washed twice in 25 ml of RIA buffer (VBS containing 5 mM CaCl2, 0.3% BSA, and 0.1% Tween 20) and resuspended in 1 ml of RIA buffer. Subsequently, 1 ml of rCRP (1 mg/ml) was added, and the tube was mixed for 1 h on a roller to allow CRP to bind, yielding CRP-PC-Sepharose. Unbound CRP was then removed by washing four times in RIA buffer. The CRP-PC-Sepharose was diluted to 25 ml in RIA buffer to give the working suspension. The effects of anti-C1q mAbs on the binding of 125I-C1q to the CRP-PC-Sepharose were tested similarly as described above for IgG-Sepharose.

RIA to assess mutual effect of CRP and IgG on binding of C1q

Similar procedures as described above to investigate the ability of anti-C1q mAbs to inhibit binding of C1q to IgG or CRP were used to assess the inhibition of C1q binding to immobilized CRP by IgG, and vice versa. rCRP directly coupled to Sepharose was used in these experiments, as the use of PC-Sepharose incubated with rCRP precluded the use of rCRP as a control competitor, because free PC sites on the PC-Sepharose bound the added rCRP. In initial experiments, the rCRP- and IgG-coupled Sepharoses were titrated (in 50 µl of RIA buffer) to yield 25–50% binding of 125I-C1q input, to enhance the sensitivity of the competition assays. Dilutions of competitors were prepared in RIA buffer to give stock concentrations of 17.4 x 10–6 M rCRP, 4 x 10–4 M IgG, 3.3 x 10–5 M AHG, and 3 x 10–4 M human serum albumin (HSA).

In the RIA, 50 µl of each competitor or control, at varying dilutions, was preincubated with 50 µl of 125I-C1q (~10,000 cpm) for 1 h at room temperature. After this, 50 µl of the appropriate Sepharose was added and the tubes were incubated on a shaker for 4 h. The beads were washed four times in PBS, the supernatant removed and the radioactivity of the spun down Sepharose was read. All tests were done in duplicate. Results were assessed as percentage of maximum 125I-C1q bound, in the absence of competing ligands.

Effects of anti-C1q mAbs on complement activation by AHG or CRP

Microtiter plates (Nunc Maxisorp) were coated with CRP (5 µg/ml) or IgG (5 µg/ml) overnight at room temperature, in 0.1 M carbonate buffer (pH 9.6), 100 µl/well. The plates were blocked with PBS containing 2% BSA, followed by addition of normal human serum diluted in VBS containing 0.5 mM MgCl2, 2 mM CaCl2, 0.05% Tween 20, and 0.1% gelatin (pH 7.5). For inhibition experiments, diluted normal human serum (1/1000) was preincubated with F(ab')2 anti-C1q (10 µg/ml) for 30 min at room temperature, followed by addition of the mixture to the plate and incubation for 1 h at 37°C. Complement activation was detected using digoxigenin-conjugated mAb C4–4a (anti-human C4d), followed by HRP-conjugated sheep anti-digoxigenin Abs (Fab from Boehringer Mannheim), both diluted in PBS containing 1% BSA and 0.05% Tween 20. Enzyme activity of HRP was detected using ABTS, followed by measurement of the OD at 415 nm.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Characterization of anti-C1q mAbs

To localize the epitopes of different anti-C1q mAbs on C1q, each of the nine anti-C1q mAbs was immobilized and binding of intact C1q and C1q fragments was studied by ELISA. These binding experiments were conducted at 0.5 M NaCl to prevent nonspecific binding of C1q to the Fc domains of IgG. All nine anti-C1q mAbs strongly recognized the intact C1q molecule (Fig. 4A). Anti-C1q-85, -89, -124, -171, and -195 showed a strong binding to C1qGR, in contrast to anti-C1q-2, -42, -52, and -139 (Fig. 4B). Conversely, mAbs anti-C1q-2, -42, -52, and -139 showed strong reactivity with C1qCLR, whereas anti-C1q-85, -89, -124, -171, and -195 showed hardly any binding to C1qCLR (Fig. 4C). These results indicated that mAbs anti-C1q-2, -42, -52, and -139 are directed against epitopes on the CLR region of C1q, while mAbs anti-C1q-85, -89, -124, -171, and -195 recognize epitopes in the C1qGR region of C1q. Nonspecific mouse Ab yielded negative results with C1q, CLR, and GR (Fig. 4), demonstrating the specificity of these assays.


Figure 4
View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 4. Epitope localization of anti-C1q mAbs. Nine different anti-C1q mAbs were immobilized on ELISA plates (5 µg/ml) and incubated with intact C1q (A), C1qGR (B), or C1qCLR (C), respectively, in the presence of 0.5 M NaCl. Coated mAb anti-IL-8 was used as a nonspecific control (ctrl). Bound C1q and C1q fragments were detected with biotinylated polyclonal rabbit anti-C1q.

 
In experiments not shown here, competition between anti-C1q mAbs for binding to C1q was observed to varying degrees indicating closely related epitopes. With respect to mAbs directed against C1qGR and CLR, respectively, there were at least two noncompeting mAbs in each group: anti-C1q-2 and -42 did not compete for binding to C1qCLR and anti-C1q 85 and 124 did not compete for binding to C1qGR.

Effects of anti-C1q mAbs on the interaction of C1q with IgG-Sepharose

To exclude interference by Fc fragments of the mAbs in the interaction between C1q and IgG or CRP, and to allow the use of buffers at physiologic ionic strength, F(ab')2 of each of the nine anti-C1q mAbs were used in subsequent experiments. These F(ab')2 were studied for their ability to inhibit the binding of radiolabeled C1q to IgG-Sepharose. F(ab')2 of anti-C1q-2, -42, -52, and -139 did not inhibit binding of 125I-C1q to immobilized IgG relative to the control F(ab')2 (Fig. 5A). In contrast, F(ab')2 of mAbs anti-C1q-85, -89, -124, -171, and -195 showed >80% inhibition of this binding (Fig. 5A). These results indicate that the five mAbs directed against C1qGR inhibit the interaction between C1q and IgG.


Figure 5
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 5. Anti-C1q F(ab')2 against C1qGR inhibit interaction of C1q with IgG and CRP. A, Inhibition of 125I -C1q interaction with IgG-Sepharose by anti-C1q mAb F(ab')2. B, Inhibition of 125I -C1q interaction with CRP-PC-Sepharose by anti-C1q mAb F(ab')2. C1q was preincubated with anti-C1q F(ab')2, after which appropriate Sepharose beads were added. After incubation, beads were washed, and radioactivity bound to Sepharose was counted. Results are given as the percent of input of radiolabeled C1q. Anti-FcRII mAb 8G3 was used as a control.

 
Effects of anti-C1q mAbs on interaction of C1q with CRP-PC-Sepharose

rCRP was incubated with PC-Sepharose to allow its immobilization in a physiologic conformation. Using this immobilized CRP, binding of radiolabeled C1q was studied in the presence of F(ab')2 anti-C1q. The results showed that mAbs anti-C1q-2, -42, -52, and -139 did not inhibit the binding of 125I-C1q to CRP relative to the control F(ab')2, whereas anti-C1q-85, -89, -124, -171, and -195 showed >90% inhibition (Fig. 5B). Thus, the F(ab')2 of the anti-C1q mAbs with C1qGR specificity were able to prevent binding of C1q not only to solid-phase IgG but also to ligand-bound CRP. In contrast, F(ab')2 from anti-C1q mAbs with C1qCLR specificity were unable to block these interactions.

Competition between CRP and IgG for binding to C1q

The above results suggested that CRP bound to C1qGR, similar to IgG, and not to C1qCLR as has been reported in the literature (23, 28). To further confirm that both CRP and IgG bind to C1qGR, we performed cross-competition experiments, using the binding of fluid-phase radiolabeled C1q to immobilized IgG and to immobilized CRP, respectively, as readouts. Preincubation of 125I-C1q with increasing concentrations of fluid-phase IgG strongly inhibited binding of 125I-C1q to rCRP-PC-Sepharose, whereas 125I-C1q binding was not inhibited by HSA, used as a negative control (Fig. 6A). Furthermore, in the reverse experiment, fluid-phase rCRP was able to inhibit the binding of I125-C1q to IgG-Sepharose in a dose-dependent manner (Fig. 6B). Binding of I125-C1q to IgG-Sepharose was also dose-dependently inhibited by fluid-phase IgG, but not by HSA (Fig. 6C). Together, these results show that interaction between C1q and CRP in the fluid phase inhibits C1q binding to immobilized IgG, whereas fluid-phase interaction between C1q and IgG inhibits the binding of C1q to immobilized CRP.


Figure 6
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 6. IgG and CRP compete for binding to C1q in a RIA. A, Radiolabeled C1q was preincubated in the presence or absence of human IgG or albumin, using concentrations as indicated, followed by incubation with rCRP-PC-Sepharose. B and C, Radiolabeled C1q was preincubated in the presence or absence of CRP (B), human IgG (C), or albumin (C), using concentrations as indicated, followed by incubation with IgG-Sepharose. Radioactivity bound to Sepharose was counted and results are expressed as the percent of input of radiolabeled C1q.

 
To directly compare competition between CRP and IgG in a single experiment, competition experiments were performed using CRP directly immobilized on Sepharose. In experiments not shown here, we observed that this rCRP-Sepharose bound labeled C1q in a similar way as rCRP-PC-Sepharose, and that this binding was inhibited by the same anti-C1q mAbs as those inhibiting binding of C1q to rCRP-PC-Sepharose. Competition experiments showed that both IgG and aggregated IgG were able to inhibit the binding to 125I-C1q to CRP-Sepharose in a dose-dependent manner, although aggregated IgG was ~10 times more efficient than monomeric IgG (Fig. 7). Furthermore, fluid-phase CRP inhibited binding of 125I-C1q to immobilized CRP with an efficacy that was better than that of IgG but less than that of aggregated IgG (Fig. 7). The specificity of the competition experiments was again shown using albumin, which did not modify 125I-C1q binding.


Figure 7
View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 7. CRP and IgG inhibit the binding of fluid-phase C1q to immobilized CRP. Radiolabeled C1q was preincubated in the presence or absence of human IgG, aggregated human IgG, CRP, or albumin, using concentrations as indicated, followed by incubation with rCRP-Sepharose. Radioactivity bound to Sepharose was counted and results are expressed as the percent of input of radiolabeled C1q.

 
Ability of anti-C1q mAbs to inhibit C1 activation

CRP is known to activate the classical pathway of complement via its interaction with C1q. rCRP as well as CRP purified from human plasma were equally able to activate the classical complement pathway when immobilized on ELISA plates, as demonstrated by the activation of C4 following incubation with increasing concentrations of fresh human serum (Fig. 8A). Control-coated wells did not show C4 deposition (Fig. 8A). To investigate the involvement of the globular region of C1q in complement activation by CRP, we assessed the effects of F(ab')2 of anti-C1q mAbs on C4 activation. The F(ab')2 of anti-C1q mAbs-2, -42, -52, and -139 showed little or no inhibition of C4 activation, whereas those of anti-C1q mAbs -85, -89, -124, -171, and -195 showed complete inhibition of C4 activation, both for rCRP (Fig. 8B) and for CRP purified from human plasma (Fig. 8C). Similar results were obtained when complement activation was induced by immobilized IgG (Fig. 8D). The requirement for complement activation in this assay was clearly shown by experiments with heat-inactivated serum or in the presence of EDTA, resulting into complete inhibition of C4 deposition, as indicated by dashed lines in Fig. 8, B–D. The results thus show that only F(ab')2 from mAbs with C1qGR specificity inhibited C1 activation by both recombinant and native CRP or IgG whereas those with C1qCLR specificity were unable to inhibit C1 activation by CRP or IgG.


Figure 8
View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 8. Complement activation by recombinant and native CRP requires its interaction with C1qGR. A, Wells were coated with recombinant CRP, native CRP purified from human plasma, or coating buffer only, followed by incubation with increasing concentrations of normal human serum as a complement source and assessment of C4 activation. B–D, Normal human serum was preincubated in the presence of anti-C1q mAb F(ab')2 or control F(ab')2 mAb, as indicated, followed by incubation in wells coated with rCRP (B), plasma CRP (C), or human IgG (D). Activation of C4 was assessed by ELISA.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
In the present study, we show that five mAbs against C1qGR inhibited the binding of C1q to CRP, whereas none of the four mAbs against C1qCLR shared this feature. Moreover, binding of C1q to CRP could be inhibited by IgG and conversely, indicating that the binding sites on C1q for IgG and CRP are both on the GR rather than on the CLR, which has earlier been suggested to harbor the binding sites for CRP (28, 29).

The mAbs used for this study were characterized regarding binding to C1qGR and C1qCLR. Five mAbs turned out to be directed to the C1qGR, and these mAbs inhibited the binding of C1q to solid-phase IgG, in agreement with the notion that the binding sites on C1q for IgG are located in the GR (16, 17). Moreover, competition experiments not shown here revealed that these five mAbs were directed against at least two different epitopes on the globular region of C1q. Thus, these mAbs provided unique molecular probes to study the interaction between C1q and CRP.

It should be noted that earlier studies used trimerized CRP (30) as a ligand for C1q interaction studies. The extrapolation of these results to native CRP is unclear and recent information about the interaction of CRP with C1q (31) emphasizes that this chemical cross-linking may influence the characteristics of this interaction. Studies on the topology of the C1q binding site on CRP (31) showed an extended deep cleft that contains the binding site for C1q, which is located on the face of the pentamer opposite that of the PC-binding site. Critical residues for C1q binding in this cleft include Asp112 and Tyr175 whereas the interaction is hindered by a lysine at position 114. Because trimerization of CRP involves chemical cross-linking via lysine residues (30), this modification might have altered its interaction with C1q, possibly contributing to the previous conclusion that CRP binds to C1qCLR. Furthermore, it is interesting to note that in earlier studies CRP trimers were observed to interact with C1qGR although competition studies were not conducted (30).

It is also important to note that many of the interactions via the CLR portion of C1q are ascribed to two regions of the A chain, 14-26 and 76-92. These are highly cationic sequences (20) with, for example, 5 basic residues in the 13 residues of C1qA 14–26. The interactions ascribed to this segment of C1q could be mainly based on low affinity charge interactions, a hypothesis strengthened by the experimental detail that half-physiologic ionic strength buffers were used to detect the binding between C1qCLR and CRP (28).

Next to these possible explanations for the discrepancies in the localization of the C1q binding site for CRP, it is important to note that earlier interaction studies used C1q and C1q fragments coated to microtiter plates, whereas in this situation, interaction with the C1qGR may not be easily accessible, because C1q binding to the plate is primarily via these regions (20). In our experiments, fluid-phase C1q and its fragments were allowed to bind to immobilized CRP.

Finally, strong support that C1q-CRP interaction is via the GR comes from the recent elucidation of the crystal structure of C1qGR, refined to 1.9 Å resolution (26). Using molecular modeling, this study revealed that the tip of C1qGR, which is predominantly basic, would fit into the central pore of the CRP pentamer. Moreover, the residues Asp112 and Tyr175 of CRP, which were experimentally shown to be critical for C1q binding, would be the contact points in this model, which strongly supports this theoretical model. Taken together, much of our present understanding of C1q and its ligands (32) now points to its interaction with CRP via C1qGR. However, we believe that the present study provides the first experimental evidence to demonstrate that this is the case under physiological conditions.

Not only CRP but also PTX3, another member of the pentraxin family, was recently shown to interact with the globular head of C1q (19). Furthermore, competition studies strongly suggested that PTX3, CRP and the pentraxin family member SAP recognize the same domain of C1q (19). Next to the pentraxin molecules, recent evidence was provided that the beta-amyloid peptide, a molecule involved in Alzheimer’s disease, also binds to C1qGR (20). Therefore, taken together, these results indicate that complement-activating C1q ligands generally bind to the GR region, leaving the CLR region available for interaction with the complement-activating enzymes C1r and C1s, and for receptor interactions with cells (Fig. 9).


Figure 9
View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 9. The heads of C1q are required for recognition of complement-activating ligands. A schematic representation of the C1q molecule is provided, containing six trimeric heads and tails. The position of globular heads and collagenous fragments, their recognition by mAbs used in the present study, and the interaction with serine proteases, receptors, and complement-activating ligands including CRP is indicated.

 
Our results show that recombinant and native CRP are both equally able to activate the classical pathway of complement, as assessed at the level of C4 deposition. This complement activation was completely inhibited by Abs against the C1qGR, but not by Abs against C1qCLR, strongly suggesting that the mechanisms of C1 activation by CRP and by IgG are similar.

In the present study, we use fluid-phase competition studies to demonstrate that binding sites for CRP and IgG on C1q are overlapping or in close proximity. In addition, these studies clearly indicate that CRP is able to bind to C1q in the fluid phase. Although in general, complement-activating ligands for C1q were proposed to need presentation in a multimeric fashion to allow C1q binding and subsequent complement activation, also for the classical interaction of C1q with Abs, it has become increasingly clear that this is largely a matter of affinity. In this respect, it has been previously shown that C1q can bind to IgG and IgM in the fluid phase, and that this interaction can compete for the binding of C1q to immobilized IgG (33), as is confirmed in the present study. Furthermore, it has been previously shown that the interaction of C1q with PTX3 also occurs in the fluid phase (19). Therefore, we propose that C1q has a low affinity interaction with CRP in the fluid phase, whereas multimeric immobilized CRP binds to C1q with high affinity. The latter interaction results in complement activation.

Our experiments with CRP and C1q in the fluid phase, as well as experiments using CRP directly conjugated to Sepharose or immobilized in ELISA wells, indicate that binding of C1q and activation of the classical complement pathway by CRP does not require its interaction with the natural CRP-ligand PC. A direct interaction of C1q with CRP (28), PTX3 (19), and SAP (34), without a need for ligand binding, has also been shown previously. Fluid-phase interaction between CRP and C1q, as we now present, excludes the alternative explanation that immobilization of CRP on plastic exposes a previously hidden C1q-binding site on CRP. It has been proposed by Agrawal et al. (31) that attachment of CRP to PC leads to a conformational change in CRP, exposing the C1q-binding site to the surface. However, our data and previously published data indicate that such a conformational change is not necessary for C1q binding to CRP, implying that this conformational change, if it occurs, may rather be involved in increasing the affinity of CRP for C1q.

CRP, once ligand-bound and aggregated, is capable of activating complement with the resultant proinflammatory effects. Many studies now show the importance of CRP levels as a predictive factor in cardiac disease both for the occurrence of a coronary event (35, 36) and postmyocardial infarct survival (37, 38). CRP has been shown to colocalize with complement in acute myocardial infarction (39) and together they have been demonstrated to be important mediators of the resulting tissue damage (2, 36). Inhibition of the early phase of complement activation has been demonstrated in an animal model to have cardioprotective effects following ischemia/reperfusion injury (40), and using complement inhibition as a therapy is the subject of many studies (reviewed in Refs.41 and 42). Given reports of the importance of the C1q-CRP interaction under conditions as wide-ranging as myocardial infarction, ischemia-reperfusion injury, and rheumatoid arthritis (43), it is hoped that a better understanding of this interaction will lead to the formulation of better therapeutics for the treatment of these conditions.


   Acknowledgments
 
We thank Danielle J. van Gijlswijk-Janssen and Nicole Schlagwein (Department of Nephrology, Leiden University Medical Center, Leiden, The Netherlands) for excellent technical assistance.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO-Netherlands Organisation for Scientific Research) and Sanquin Research. A.R. is supported by the Dutch Kidney Foundation. Back

2 Current address: Crucell Holland, Leiden, The Netherlands. Back

3 Address correspondence and reprint requests to Dr. Anja Roos, Department of Nephrology, D3P, Leiden University Medical Center, Albinusdreef 2, 2333 ZA Leiden, The Netherlands. E-mail address: A.Roos{at}LUMC.NL Back

4 Abbreviations used in this paper: CRP, C-reactive protein; CLR, collagen-like region; GR, globular region; 125I-C1q, 125I-labeled C1q; VBS, veronal-buffered saline; AHG, aggregated human IgG; RIA, radioimmunoassay; PC, phosphorylcholine; HSA, human serum albumin. Back

Received for publication November 23, 2004. Accepted for publication December 1, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Arlaud, G. J., C. Gaboriaud, N. M. Thielens, M. Budayova-Spano, V. Rossi, J. C. Fontecilla-Camps. 2002. Structural biology of the C1 complex of complement unveils the mechanisms of its activation and proteolytic activity. Mol. Immunol. 39: 383-394. [Medline]
  2. Griselli, M., J. Herbert, W. L. Hutchinson, K. M. Taylor, M. Sohail, T. Krausz, M. B. Pepys. 1999. C-reactive protein and complement are important mediators of tissue damage in acute myocardial infarction. J. Exp. Med. 190: 1733-1740. [Abstract/Free Full Text]
  3. Fleming, S. D., T. Shea-Donohue, J. M. Guthridge, L. Kulik, T. J. Waldschmidt, M. G. Gipson, G. C. Tsokos, V. M. Holers. 2002. Mice deficient in complement receptors 1 and 2 lack a tissue injury-inducing subset of the natural antibody repertoire. J. Immunol. 169: 2126-2133. [Abstract/Free Full Text]
  4. Weiser, M. R., J. P. Williams, F. D. Moore, Jr, L. Kobzik, M. Ma, H. B. Hechtman, M. C. Carroll. 1996. Reperfusion injury of ischemic skeletal muscle is mediated by natural antibody and complement. J. Exp. Med. 183: 2343-2348. [Abstract/Free Full Text]
  5. Kaplan, M. H., J. E. Volanakis. 1974. Interaction of C-reactive protein complexes with the complement system. I. Consumption of human complement associated with the reaction of C-reactive protein with pneumococcal C-polysaccharide and with the choline phosphatides, lecithin and sphingomyelin. J. Immunol. 112: 2135-2147. [Abstract/Free Full Text]
  6. Volanakis, J. E., M. H. Kaplan. 1974. Interaction of C-reactive protein complexes with the complement system. II. Consumption of guinea pig complement by CRP complexes: requirement for human C1q. J. Immunol. 113: 9-17. [Abstract/Free Full Text]
  7. Volanakis, J. E., A. J. Narkates. 1981. Interaction of C-reactive protein with artificial phosphatidylcholine bilayers and complement. J. Immunol. 126: 1820-1825. [Abstract]
  8. Roos, A., W. Xu, G. Castellano, A. J. Nauta, P. Garred, M. R. Daha, C. van Kooten. 2004. Mini-review: a pivotal role for innate immunity in the clearance of apoptotic cells. Eur. J. Immunol. 34: 921-929. [Medline]
  9. Navratil, J. S., S. C. Watkins, J. J. Wisnieski, J. M. Ahearn. 2001. The globular heads of C1q specifically recognize surface blebs of apoptotic vascular endothelial cells. J. Immunol. 166: 3231-3239. [Abstract/Free Full Text]
  10. Nauta, A. J., L. A. Trouw, M. R. Daha, O. Tijsma, R. Nieuwland, W. J. Schwaeble, A. R. Gingras, A. Mantovani, E. C. Hack, A. Roos. 2002. Direct binding of C1q to apoptotic cells and cell blebs induces complement activation. Eur. J. Immunol. 32: 1726-1736. [Medline]
  11. Ciurana, C. L. F., C. E. Hack. 2002. Molecular mechanisms of complement activation during ischaemia and reperfusion. J.-L. Vincent, Jr, ed. Yearbook of Intensive Care and Emergency Medicine 39-50. Springer, Berlin.
  12. Schumaker, V. N., P. Zavodszky, P. H. Poon. 1987. Activation of the first component of complement. Annu. Rev. Immunol. 5: 21-42. [Medline]
  13. Reid, K. B., R. R. Porter. 1976. Subunit composition and structure of subcomponent C1q of the first component of human complement. Biochem. J. 155: 19-23. [Medline]
  14. Kishore, U., K. B. Reid. 2000. C1q: structure, function, and receptors. Immunopharmacology 49: 159-170. [Medline]
  15. Reid, K. B.. 1976. Isolation, by partial pepsin digestion, of the three collagen-like regions present in subcomponent Clq of the first component of human complement. Biochem. J. 155: 5-17. [Medline]
  16. Hughes-Jones, N. C., B. Gardner. 1979. Reaction between the isolated globular sub-units of the complement component C1q and IgG-complexes. Mol. Immunol. 16: 697-701. [Medline]
  17. Poon, P. H., M. L. Phillips, V. N. Schumaker. 1985. Immunoglobulin M possesses two binding sites for complement subcomponent C1q, and soluble 1:1 and 2:1 complexes are formed in solution at reduced ionic strength. J. Biol. Chem. 260: 9357-9365. [Abstract/Free Full Text]
  18. Bristow, C. L., R. J. Boackle. 1986. Evidence for the binding of human serum amyloid P component to Clq and Fab {gamma}. Mol. Immunol. 23: 1045-1052. [Medline]
  19. Nauta, A. J., B. Bottazzi, A. Mantovani, G. Salvatori, U. Kishore, W. J. Schwaeble, A. R. Gingras, S. Tzima, F. Vivanco, J. Egido, et al 2003. Biochemical and functional characterization of the interaction between pentraxin 3 and C1q. Eur. J. Immunol. 33: 465-473. [Medline]
  20. Tacnet-Delorme, P., S. Chevallier, G. J. Arlaud. 2001. beta-Amyloid fibrils activate the C1 complex of complement under physiological conditions: evidence for a binding site for A beta on the C1q globular regions. J. Immunol. 167: 6374-6381. [Abstract/Free Full Text]
  21. Duncan, A. R., G. Winter. 1988. The binding site for C1q on IgG. Nature 332: 738-740. [Medline]
  22. Perkins, S. J., A. S. Nealis, B. J. Sutton, A. Feinstein. 1991. Solution structure of human and mouse immunoglobulin M by synchrotron X-ray scattering and molecular graphics modelling: a possible mechanism for complement activation. J. Mol. Biol. 221: 1345-1366. [Medline]
  23. Jiang, H. X., J. N. Siegel, H. Gewurz. 1991. Binding and complement activation by C-reactive protein via the collagen-like region of C1q and inhibition of these reactions by monoclonal antibodies to C-reactive protein and C1q. J. Immunol. 146: 2324-2330. [Abstract]
  24. Tenner, A. J., P. H. Lesavre, N. R. Cooper. 1981. Purification and radiolabeling of human C1q. J. Immunol. 127: 648-653. [Abstract]
  25. Hack, C. E., A. J. Eerenberg-Belmer, A. J. Hannema, T. A. Out, R. C. Aalberse. 1981. Polyethylene glycol enhances the binding of C1q to circulating immune complexes. J. Immunol. Methods 44: 211-221. [Medline]
  26. Gaboriaud, C., J. Juanhuix, A. Gruez, M. Lacroix, C. Darnault, D. Pignol, D. Verger, J. C. Fontecilla-Camps, G. J. Arlaud. 2003. The crystal structure of the globular head of complement protein C1q provides a basis for its versatile recognition properties. J. Biol. Chem. 278: 46974-46982. [Abstract/Free Full Text]
  27. Hoekzema, R., M. Martens, M. C. Brouwer, C. E. Hack. 1988. The distortive mechanism for the activation of complement component C1 supported by studies with a monoclonal antibody against the "arms" of C1q. Mol. Immunol. 25: 485-494. [Medline]
  28. Jiang, H., F. A. Robey, H. Gewurz. 1992. Localization of sites through which C-reactive protein binds and activates complement to residues 14–26 and 76–92 of the human C1q A chain. J. Exp. Med. 175: 1373-1379. [Abstract/Free Full Text]
  29. Gewurz, H., S. C. Ying, H. Jiang, T. F. Lint. 1993. Nonimmune activation of the classical complement pathway. Behring. Inst. Mitt. : 138-147.
  30. Jiang, H., T. F. Lint, H. Gewurz. 1991. Defined chemically cross-linked oligomers of human C-reactive protein: characterization and reactivity with the complement system. Immunology 74: 725-731. [Medline]
  31. Agrawal, A., A. K. Shrive, T. J. Greenhough, J. E. Volanakis. 2001. Topology and structure of the C1q-binding site on C-reactive protein. J. Immunol. 166: 3998-4004. [Abstract/Free Full Text]
  32. Kishore, U., R. Ghai, T. J. Greenhough, A. K. Shrive, D. M. Bonifati, M. G. Gadjeva, P. Waters, M. S. Kojouharova, T. Chakraborty, A. Agrawal. 2004. Structural and functional anatomy of the globular domain of complement protein C1q. Immunol. Lett. 95: 113-128. [Medline]
  33. Rieben, R., A. Roos, Y. Muizert, C. Tinguely, A. F. Gerritsen, M. R. Daha. 1999. Immunoglobulin M-enriched human intravenous immunoglobulin prevents complement activation in vitro and in vivo in a rat model of acute inflammation. Blood 93: 942-951. [Abstract/Free Full Text]
  34. Ying, S. C., A. T. Gewurz, H. Jiang, H. Gewurz. 1993. Human serum amyloid P component oligomers bind and activate the classical complement pathway via residues 14–26 and 76–92 of the A chain collagen-like region of C1q. J. Immunol. 150: 169-176. [Abstract]
  35. Lagrand, W. K., C. A. Visser, W. T. Hermens, H. W. Niessen, F. W. Verheugt, G. J. Wolbink, C. E. Hack. 1999. C-reactive protein as a cardiovascular risk factor: more than an epiphenomenon?. Circulation 100: 96-102. [Abstract/Free Full Text]
  36. Nijmeijer, R., W. K. Lagrand, C. A. Visser, C. J. Meijer, H. W. Niessen, C. E. Hack. 2001. CRP, a major culprit in complement-mediated tissue damage in acute myocardial infarction?. Int. Immunopharmacol. 1: 403-414. [Medline]
  37. Pepys, M. B., G. M. Hirschfield. 2003. C-reactive protein: a critical update. J. Clin. Invest. 111: 1805-1812. [Medline]
  38. Hirschfield, G. M., M. B. Pepys. 2003. C-reactive protein and cardiovascular disease: new insights from an old molecule. Q. J. Med. 96: 793-807.
  39. Lagrand, W. K., H. W. Niessen, G. J. Wolbink, L. H. Jaspars, C. A. Visser, F. W. Verheugt, C. J. Meijer, C. E. Hack. 1997. C-reactive protein colocalizes with complement in human hearts during acute myocardial infarction. Circulation 95: 97-103. [Abstract/Free Full Text]
  40. Buerke, M., H. Schwertz, W. Seitz, J. Meyer, H. Darius. 2001. Novel small molecule inhibitor of C1s exerts cardioprotective effects in ischemia-reperfusion injury in rabbits. J. Immunol. 167: 5375-5380. [Abstract/Free Full Text]
  41. Roos, A., T. H. Ramwadhdoebe, A. J. Nauta, C. E. Hack, M. R. Daha. 2002. Therapeutic inhibition of the early phase of complement activation. Immunobiology 205: 595-609. [Medline]
  42. Roos, A., A. J. Nauta, D. Broers, M. C. Faber-Krol, L. A. Trouw, J. W. Drijfhout, M. R. Daha. 2001. Specific inhibition of the classical complement pathway by C1q-binding peptides. J. Immunol. 167: 7052-7059. [Abstract/Free Full Text]
  43. Molenaar, E. T., A. E. Voskuyl, A. Familian, G. J. van Mierlo, B. A. Dijkmans, C. E. Hack. 2001. Complement activation in patients with rheumatoid arthritis mediated in part by C-reactive protein. Arthritis Rheum. 44: 997-1002. [Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
N. Brouwer, F. N. J. Frakking, M. D. van de Wetering, M. van Houdt, M. Hart, I. K. Budde, P. F. W. Strengers, I. Laursen, G. Houen, D. Roos, et al.
Mannose-Binding Lectin (MBL) Substitution: Recovery of Opsonic Function In Vivo Lags behind MBL Serum Levels
J. Immunol., September 1, 2009; 183(5): 3496 - 3504.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
K. E. Happonen, A. P. Sjoberg, M. Morgelin, D. Heinegard, and A. M. Blom
Complement Inhibitor C4b-Binding Protein Interacts Directly with Small Glycoproteins of the Extracellular Matrix
J. Immunol., February 1, 2009; 182(3): 1518 - 1525.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
N. Brouwer, K. M. Dolman, M. van Houdt, M. Sta, D. Roos, and T. W. Kuijpers
Mannose-Binding Lectin (MBL) Facilitates Opsonophagocytosis of Yeasts but Not of Bacteria despite MBL Binding
J. Immunol., March 15, 2008; 180(6): 4124 - 4132.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Endocrinol. Metab.Home page
J. Zhang, W. Wright, D. A. Bernlohr, S. W. Cushman, and X. Chen
Alterations of the classic pathway of complement in adipose tissue of obesity and insulin resistance
Am J Physiol Endocrinol Metab, May 1, 2007; 292(5): E1433 - E1440.
[Abstract] [Full Text] [PDF]

Home page
 Circ. Res.Home page
J. Wehner, C. N. Morrell, T. Reynolds, E. R. Rodriguez, and W. M. Baldwin III
Antibody and Complement in Transplant Vasculopathy
Circ. Res., February 2, 2007; 100(2): 191 - 203.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McGrath, F. D. G.
Right arrow Articles by Roos, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McGrath, F. D. G.
Right arrow Articles by Roos, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS