Effect of Modified C-Reactive Protein on Complement Activation

【摘要】  Objective- The capacity of human C-reactive protein (CRP) to activate/regulate complement may be an important characteristic that links CRP and inflammation with atherosclerosis. Recent advances suggest that in addition to classical pentameric CRP, a conformationally distinct isoform of CRP, termed modified or monomeric CRP (mCRP), may also play an active role in atherosclerosis. Although the complement activation behavior of CRP has been well established, the capacity of mCRP to interact with and activate the complement cascade is unknown.

Methods and Results- mCRP bound avidly to purified C1q, and this binding occurred primarily through collagen-like region of C1q. Fluid phase mCRP inhibited the activation of complement cascade via engaging C1q from binding with other complement activators. In contrast, when immobilized or bound to oxidized or enzymatically modified low-density lipoprotein, mCRP could activate classical complement pathway. Low-level generation of sC5b-9 indicated that the activation largely bypassed the terminal sequence of complement, which appears to involve recruitment of Factor H.

Conclusions- These results indicate that mCRP can both inhibit and activate the classical complement pathway by binding C1q, depending on whether it is in fluid phase or surface-bound state.

The current work studied interaction of mCRP with complement. Our data showed that mCRP can both inhibit and activate the classical complement pathway by binding C1q, depending on whether it is in fluid phase or surface bound.

【关键词】  modified Creactive protein complement lowdensity lipoprotein

Introduction

Several lines of evidence demonstrate the involvement of the complement cascade in atherogenesis. Immunohistochemical staining revealed the presence of complement components 1-3 in both early and advance human atherosclerotic lesions, indicating that complement activation occurred at an early stage in atherosclerosis. In addition, RNA analyses have indicated that complement genes are expressed locally within plaques. 4 A proatherosclerotic role of terminal complement sequence was suggested because C6 5 and C3 6 deficiency showed significant protective effect in diet-induced atherosclerotic rabbit and low-density lipoprotein receptor -/- (LDLR -/- ) mice, respectively. However, C5 deficiency had no effect on apolipoprotein E-deficient (apoE -/- ) mice. 7 Overall, although complement activation is considered to be of pathogenetic relevance, it remains to be established how complement is activated and how complement directly influences the atherosclerotic process.

Several components in the arterial wall may trigger complement activation and C5b-9 assembly. 4 Among these components, C-reactive protein (CRP), a typical pentraxin found in both normal arteries and atherosclerotic lesions, 8,9 is of special interest. Although CRP had long been regarded only as a sensitive prognostic marker, evidence is now accumulating to suggest its direct participation in the atherosclerotic process. 10,11 CRP is known to activate the classical complement pathway (CCP) when bound to multivalent ligands and is capable of regulating alternative complement pathway (ACP) in a mechanism involving recruiting Factor H. 12 CRP can modulate complement activation properties of enzymatically modified LDL (E-LDL) 13 and colocalize with activated complement components 2 and various LDL derivates such as oxidized LDL (ox-LDL) and E-LDL 13-15 in human atherosclerotic lesions. The capacity of CRP to bind LDL and activate/modulate the complement cascade suggests that CRP may play an important role in regulation of complement activation during atherosclerosis process. Recent advances 16,17 indicate that transgenic CRP has no effect on atherosclerosis development in apoE -/- mice and propose that this animal model is not adequate for exploring the atherosclerotic effect of CRP, which may be partly related to the oddities of the mouse complement system. 17

CRP is composed of 5 identical globular subunits arranged as a cyclic pentamer. When the subunits are separated into monomers, they undergo a conformational change that significantly modifies solubility and antigenicity and presents very distinct biological activities. 18-25 This alternative form of CRP is termed modified or monomeric CRP (mCRP). An mCRP antigen is preferentially expressed in tissues 18,26 in contrast to CRP, which is a serum-based protein. mCRP is a predominant antigen detected in human arterial intima. 26 In addition, although the expression of CRP mRNA in both normal and plaque arterial tissues has been identified, 4,27,28 it remains to be proven whether CRP or mCRP isoform is finally produced by extrahepatic cells. In fact, recent evidence indicates that CRP mRNA is processed into mCRP isoform rather than CRP isoform by U937 macrophage. 29 Together, with the emerging knowledge that CRP can naturally exist in 2 isomeric forms and that each has distinct bioactivities, studies that clearly define and differentiate how each isoform may participate in underlying process of atherosclerosis are needed.

The capacity to activate/regulate complement is an important characteristic that links CRP with inflammation and with the suggested role of inflammation in atherosclerosis. Because the monoclonal antibody (mAb; Clone CRP-8) used in several studies 2,13-15 to illustrate colocalization of CRP with complement predominantly recognizes mCRP, we investigated the direct capacity of mCRP to regulate the complement system in the present work.

Materials and Methods

CRP/mCRP batches were dialyzed to remove NaN 3 and assayed for endotoxin contaminant by Limulus assay (Sigma) before use. The final endotoxin level of all protein solutions was <0.06 EU/mL. For details on protein preparation, please refer to the online supplemental Materials and Methods, available at http://atvb.ahajournals.org.

Fluid Phase Binding of mCRP to Immobilized C1q

A total of 100 µL C1q (5 µg/mL) in coating buffer (10 mmol/L sodium bicarbonate, pH 9.6) was incubated in microtiter wells at 37°C for 2 hours. The wells were washed 3 times with TBS (10 mmol/L Tris, 140 mmol/L NaCl, pH 7.4) containing 0.02% Nonidet P-40 for 5 minutes and blocked with 1% BSA/TBS for 1 hour at 37°C. Biotinylated mCRP or CRP in 1% BSA/TBS containing 2 mmol/L calcium was then added for 1 hour at 37°C. For inhibition assays, C1q-coated wells were preincubated with inhibitor for 30 minutes at 37°C. After washes, horseradish peroxidase (HRP)-conjugated avidin (0.1 µg/mL; Pierce) in 1% BSA was added for 1 hour at 37°C. After washing, wells were developed using 0.4 mg/mL o-Phenylenediamine substrate, with quantification at an OD of 492 nm after addition of 1 mol/L H 2 SO 4.

Fluid Phase Binding of C1q to Immobilized mCRP

A total of 100 µL (nonbiotinylated) CRP or mCRP at 5 µg/mL was coated overnight at 4°C. After blocking with 1% BSA for 1 hour, C1q was added and the binding was detected with goat polyclonal antibody against human C1q (10 µg/mL; Calbiochem) and HRP-conjugated rabbit anti-goat antibody (10 µg/mL; Sigma).

Fluid Phase Binding of C1q to mCRP-Coated LDL

A total of 10 µg/mL LDL was immobilized onto microtiter wells overnight at 4°C. After blocking, 5 µg/mL mCRP was added for 1 hour at 37°C. After washing, 5 µg/mL C1q was added for another 1 hour, and then the binding was detected.

C3 Deposition Assays

A total of 5 µg/mL mCRP was immobilized in microtiter wells overnight at 4°C. After blocking with 1% BSA, 1% normal human serum (NHS) diluted in veranol-buffered saline (VBS) Ca+Mg (VBS containing 0.15 mmol/L calcium and 0.5 mmol/L magnesium), VBS Mg+EGTA (VBS containing 0.5 mmol/L magnesium and 5 mmol/L EGTA), or VBS EDTA (VBS containing 5 mmol/L EDTA) was added in protein-coated wells at 37°C for 1 hour. C3d deposition was detected by sheep anti-human C3d polyclonal antibody (1 µg/mL; Binding Site) and HRP-conjugated donkey anti-sheep antibody (1:1000; Sigma).

Other Complement Assays

To examine CCP activation, EA (antibody sensitized sheep erythrocytes) hemolytic activity assays were performed. Various amounts of mCRP, LDL, or mCRP-LDL complex were preincubated with 5% NHS diluted in VBS Ca+Mg for 1 hour at 37°C before addition of 1 x 10 8 EA in VBS Ca+Mg. After 30 minutes of incubation at 37°C, VBS EDTA was added to stop the reaction. After centrifugation, released hemoglobin was quantified at an OD of 412 nm. mCRP-LDL complex was produced by coincubation of mCRP and LDL in VBS Ca (VBS containing 0.15 mmol/L calcium) for 1 hour at 37°C.

To examine ACP activation, RRBC (rabbit red blood cell) hemolytic activity assays were performed. Various amounts of CRP or mCRP were preincubated with 50% NHS diluted in VBS Ca+Mg for 1 hour at 37°C and then added to 2 x 10 7 RRBCs in VBS Mg+EGTA for 30 minutes. The reaction was stopped by addition of VBS EDTA, and 412-nm OD was measured.

For detection of inactivated C3b (iC3b) and soluble C5b-C9 complexes (sC5b-9), mCRP, LDL, or mCRP-LDL complex was preincubated with 5% NHS, and then iC3b and sC5b-9 were measured by commercial ELISA kits (Quidel).

Statistical Analysis

All measurements were performed in duplicate, and each experiment was repeated 3 to 5 times. Data were compared between experimental groups by 1-way ANOVA followed by the Bonferroni multiple range test using OriginPro 7.5 software (OriginLab Cooperation). The normality of data were verified by Shapiro-Wilk normality test. A value of P <0.05 was considered significant. Data are presented as mean±SEM.

Results

Binding of mCRP to C1q

We first investigated the interactions between mCRP and C1q, the initial component of the CCP. Microtiter wells were coated with fixed amount of C1q and assessed for binding of biotinylated mCRP ( Figure 1 A); alternatively, wells were coated with mCRP, and C1q binding was detected ( Figure 1 B). The results indicated that mCRP, either in immobilized phase or fluid phase, bound C1q avidly in a dose-dependent fashion. As negative control, no binding of C1q to BSA was observed.

Figure 1. A, Binding of mCRP ( ) or agg-IgG ( ) to immobilized C1q. A total of 5 µg/mL C1q was immobilized onto microtiter wells for 2 hours at 37°C. After blocking with 1% BSA for 1 hour, biotinylated mCRP was added and the binding was detected with HRP-conjugated avidin (0.1 µg/mL). Biotin-labeled aggregated IgG served as positive control. Background HRP-conjugated avidin binding to coated C1q was <0.1 in all experiments. B, Binding of C1q to immobilized mCRP ( ), agg-IgG ( ), or BSA (). A total of 5 µg/mL mCRP, agg-IgG, or BSA was immobilized onto microtiter wells overnight at 4°C. After blocking with 1% BSA for 1 hour, C1q was added and the binding was detected with goat polyclonal antibody against human C1q (10 µg/mL) and HRP-conjugated rabbit anti-goat antibody (10 µg/mL). Each data point represents the mean of 3 independent experiments.

In contrast to mCRP, CRP exhibited rather weak binding to C1q and was transformed to a potent ligand for C1q binding after cross-linking (ie, Bis-aggCRP; Figure 2 A), in agreement with Jiang et al. 30 A recent study reported that after immobilized onto microtiter wells, rabbit CRP bound C1q strongly. 31 Our results using CRP directly absorbed onto polystyrene (latex) microtiter plates also supported this observation ( Figure 2 B). However, further experiments indicated that substantially conformational change occurred in CRP after direct immobilization such that 95%) and is not recognized by mAbs against CRP determinants 36 ( Figure 2 C). As positive controls, CRP retained pentameric CRP antigenicity by ELISA when captured as a function of calcium onto phosphorylcholine-conjugated BSA and by dot-blot assays when CRP is adsorbed onto nitrocellulose (an ionized surface; data not shown), consistent with previous reports. 32,33 These results indicated that conversion from CRP to mCRP rendered this molecule more active in C1q binding and abolished the prerequisite of complexed with multivalent ligand or cross-linking.

Figure 2. A, The binding of 5 µg/mL biotinylated CRP, mCRP, and cross-linked CRP (Bis-aggCRP; supplemental Materials) to immobilized C1q. Experimental procedure was the same as described in Figure 1 A. **P <0.01 against binding of CRP to coated C1q. B, The binding of fluid phase C1q to immobilized mCRP, CRP, and phosphorylcholine-conjugated BSA (PC-BSA)+CRP. For PC-BSA+CRP, 15 µg/mL PC-BSA was first immobilized overnight at 4°C. After washing and blocking, 5 µg/mL CRP was added in the presence of 2 mmol/L calcium for 2 hours at 37°C. Other experimental procedures were the same as described in Figure 1 95% of the immobilized CRP showed mCRP antigenicity. Each data point represents the mean of 3 independent experiments.

Additional experiments were performed to locate the site of C1q involved in the interaction with mCRP. In solid phase C1q-binding assays, preincubation with agg-IgG (aggregated human IgG) did not inhibit the binding of biotinylated mCRP to immobilized C1q ( Figure 3 A). However, preincubation with mCRP partially inhibited binding of biotinylated agg-IgG to immobilized C1q. Because agg-IgG is known to bind globular region of C1q, the above results suggested mCRP might primarily interact with collagen-like region of C1q. To further address this point, we assayed the binding of mCRP to separated collagen-like region and globular region of C1q. As shown in Figure 3 B, agg-IgG only bound to globular region, whereas mCRP bound both fragments, with predominant binding to collagen-like region.

Figure 3. A, Cross-competition assays of mCRP and agg-IgG binding to immobilized C1q. A total of 5 µg/mL nonbiotinylated mCRP (black bar), nonbiotinylated agg-IgG (gray bar), or buffer alone (white bar) was first added to C1q-coated wells for 30 minutes at 37°C. After washing, 0.5 µg/mL biotinylated mCRP or biotinylated agg-IgG was added and binding was detected by HRP-conjugated avidin. *P <0.05; **P <0.01 against binding without competitor. B, Binding of mCRP (white bar) or agg-IgG (black bar) to immobilized globular region and collagen-like region of C1q. A total of 5 µg/mL globular region or collagen-like region was immobilized in microtiter wells for 2 hours at 37°C followed by blocking. Then biotinylated mCRP or agg-IgG (5 µg/mL) was added and detected for protein binding. Each data point represents the mean of 3 independent experiments.

Fluid Phase mCRP Block Complement Activation Via Engaging C1q

To examine whether mCRP-C1q binding can activate complement, hemolytical assays were performed. Preincubation of NHS with mCRP resulted in dose-dependent decrease in hemolytic activity of EA through CCP ( Figure 4 A). On the other hand, mCRP addition to NHS had only a marginal effect on hemolytic activity of RRBCs through ACP ( Figure 4 B). These results are consistent with the interpretation that mCRP can consume the complement components in NHS necessary for CCP processing but not for ACP processing. Because C3 is the converging point of both CCP and ACP, the effect of fluid mCRP on C3 activation was accessed by evaluating either the formation of mCRP-C3d complex or the generation of iC3b. Comparable generation of iC3b was observed in both NHS and NHS preincubated with mCRP ( Figure 4 C), and only trace levels of mCRP-C3d complex could be detected (data not shown). This indicated that addition of mCRP in NHS did not result in C3 turnover, suggesting successive complement cascade was not activated by fluid phase mCRP-C1q binding. Rather, it is most likely that such binding engages C1q from interacting with other activators such as EA and hence blocks further activation of complement system. In line with this speculation, mCRP dose-dependently inhibits the binding of C1q to EA ( Figure 4 D). Together, these findings support the concept that the binding of free mCRP molecule to C1q does not activate complement, possibly caused by inefficiency in activating C1q hexamer. Similar fluid phase inhibition of complement has also been reported for PTX3, 34 a prototypic long pentraxin.

Figure 4. A, Inhibition of CCP EA hemolytic activity by CRP ( ) or mCRP ( ). Various amounts of CRP or mCRP were preincubated with 5% NHS diluted in VBS Ca+Mg for 1 hour at 37°C and were then added to 1 x 10 8 EA in VBS Ca+Mg. After 30 minutes of incubation at 37°C, VBS EDTA was added to stop the reaction. After centrifugation, 412-nm OD of the reaction mixture was measured. *P <0.05; **P <0.01 against of EA inhibition by CRP at equal concentration. B, Inhibition of ACP RRBC hemolytic activity by CRP ( ) or mCRP ( ).Various amounts of CRP or mCRP were preincubated with 50% NHS diluted in VBS Ca+Mg for 1 hour at 37°C and were then added to 2 x 10 7 RRBCs in VBS Mg+EGTA for 30 minutes. The reaction was stopped and 412-nm OD was detected. C, Various amounts of mCRP ( ) or agg-IgG ( ) was preincubated with 5% NHS for 1 hour, and generation of iC3b was determined by an ELISA kit. *P <0.05; **P <0.01 against iC3b generation by mCRP. D, mCRP inhibits binding of C1q to EA. EA was incubated with 5% NHS in the presence of various concentrations of mCRP at 4°C. C1q binding was accessed by goat-anti human C1q and fluoresceinisothiocyanate-conjugated rabbit anti-goat antibody (Table I, available online at http://atvb.ahajournals.org). Similar competition results were also obtained by using EA and isolated C1q (instead of NHS). *P <0.05; **P <0.01 against residue binding of C1q on EA with NHS alone. Each data point represents the mean of 5 independent experiments.

Immobilized mCRP Activates CCP

Because fluid phase (or ligand free) mCRP could interact with and inhibit the hemolytic activity of the CCP on EA ( Figure 4 ), further experiments were performed to identify whether mCRP immobilized on ELISA plates could similarly affect the complement system. Immobilized mCRP was incubated with serum, and C3d deposition was measured by ELISA. Addition of NHS led to C3 activation on coated mCRP but not on BSA ( Figure 5 A). When NHS was added in the presence of Mg-EGTA, C3d deposition was almost abolished, indicating the effect of mCRP on C3d deposition predominantly involves CCP. Further support was provided by the experiments using C1q-deficient human serum (C1qD-HS) as the complement source, which resulted in no deposition of C3d on immobilized mCRP ( Figure 5 B). Addition of purified C1q to the C1qD-HS could restore C3d deposition. As shown in Figure 2 and in a previous report, immobilized CRP exclusively exhibited mCRP antigenicity, thus the complement activation of immobilized CRP was tested. Similar to immobilized mCRP, immobilized CRP also exclusively activated CCP, which was in line with the previous report using rabbit CRP in solid phase complement activation assays. 31

Figure 5. A, C3d deposition on immobilized mCRP (white bar), CRP (black bar), or BSA (gray bar). A total of 5 µg/mL proteins was immobilized overnight at 4°C. After blocking with 1% BSA, 1% NHS diluted in VBS Ca+Mg, VBS Mg+EGTA, or VBS EDTA was added in protein-coated wells at 37°C for 1 hour. C3d deposition was detected by sheep anti-human C3d polyclonal IgG. **P <0.01 against C3d deposition in NHS-EDTA. B, 1% NHS, C1qD-HS, or C1qD-HS supplemented with purified C1q (2 µg/mL) in VBS Ca+Mg was added to mCRP-coated wells, and C3d deposition was measured. Each data point represents the mean of 3 independent experiments.

mCRP Regulate Complement Activation Behavior of Modified LDL

Native LDL (n-LDL) modified by oxidation 15 or enzyme 13,14 are both of atherosclerotic relevance and are shown to be colocalized with CRP (stained with CRP-8 Clone from Sigma) in human atherosclerotic lesions. Because mCRP could avidly bind to these LDL derivatives 35 (Figure I, available online at http://atvb.ahajournals.org), we examined the effect of mCRP-bound LDLs on C1q binding and complement activation. As shown in Figure 6 A, n-LDL and E-LDL did not bind C1q, whereas ox-LDL bound an appreciable amount of C1q. However, when coated with mCRP, all 3 LDL derivatives showed substantial binding to C1q ( Figure 6 A). Complement activation of the different mCRP-LDL complexes was further investigated by measuring EA hemolytic activity and generation of activated complement products. At a concentration of 5 µg/mL, neither mCRP alone ( Figure 4 A) nor LDL derivatives alone ( Figure 6 B) significantly inhibited NHS hemolytic activity on EA. However, addition of LDL derivatives preincubated with 5 µg/mL mCRP did significantly inhibit NHS CCP hemolysis ( Figure 6 B).

Figure 6. A, Binding of C1q to immobilized LDL (white bar) or mCRP-coated LDL (black bar). A total of 10 µg/mL LDL was immobilized onto microtiter wells overnight at 4°C. Then 5 µg/mL mCRP was added for 1 hour at 37°C. After washing, 5 µg/mL C1q was added for another 1 hour, and then the binding was detected as in Figure 1. *P <0.05; **P <0.01 against C1q binding to LDL alone. B, Inhibition of EA hemolytic activity by LDL alone (white bar) or mCRP-LDL complex (black bar). A total of 5 µg/mL LDL with or without 1 hour preincubation with 5 µg/mL mCRP was added to 5% NHS diluted in VBS Ca+Mg for 1 hour at 37°C and then assayed for EA hemolysis as in Figure 4 A. *P <0.05; **P <0.01 against EA inhibition by LDL alone. C, A total of 5 µg/mL LDL with or without 1-hour preincubation with 5 µg/mL mCRP was reacted with 5% NHS, and iC3b (white bar) and sC5b-9 (black bar) generation was determined by ELISA kits. *P <0.05; **P <0.01 against iC3b/sC5b-9 generation in serum alone. D, Various amount of Factor H was added to mCRP-coated wells, and the binding was detected with sheep anti-human Factor H IgG (Binding Site). Each data point represents the mean of 3 to 5 independent experiments.

Unlike the absence of C3 activation by fluid phase (free) mCRP, ox-LDL or E-LDL complexed with mCRP, but not n-LDL complexed with mCRP, induced obvious consumption of C3 ( Figure 6 C), indicating progression of early complement cascade. The exclusive CCP activation of mCRP-LDL complex was demonstrated by lack of iC3b generation when C1qD-HS was used as complement source (data not shown). Further analysis showed that C3 activation was not paralleled by generation of terminal sequence product, only a small increase in sC5b-9 generation could be noted compared with baseline levels. Thus, complement activation by mCRP-LDL complex does not involve the more inflammatory and destructive terminal sequence. Control experiments indicated that coincubation of CRP with n-LDL or ox-LDL did not induce complement consumption, whereas CRP-E-LDL complex showed enhanced complement activation (data not shown). These results were in line with the observation of Bhakdi et al. 13

Mold et al reported that the regulatory role of CRP on complement activation was mediated by Factor H binding. 12 They reported only immobilized CRP could bind Factor H and thus proposed that the immobilization exposed additional sites on CRP for Factor H binding. As depicted in Figure 2 and previous reports, 33 immobilized CRP actually expressed mCRP antigenicity or, in other words, immobilized CRP present additional site(s) usually inaccessible in pentameric structure but exposed in monomeric structure. This prompted us to examine the interaction between mCRP and Factor H. As shown in Figure 6 D, dose-dependent Factor H binding to immobilized mCRP could be observed.

Discussion

CRP is a serum protein mainly found in the circulation that is widely appreciated by the scientific and medical communities as the most reliable diagnostic marker for inflammation. However, although it is linked as a marker for inflammation, significant accumulation of CRP in inflammatory or injured tissues has not been noted, 36 and the role it plays as a stimulator or inhibitor of inflammatory processes continues to be an enigma. In light of the new and evolving appreciation that CRP can exist in 2 conformationally distinct isoforms (ie, native CRP and mCRP; for a detailed discussion about the mechanism of mCRP production please refer to the online supplement), studies are needed to specifically compare and contrast the bioactivities of each conformer. Complement activation is considered to be an important source of inflammation in atherosclerosis, and CRP, because of its known capacity to activate complement and its colocalization to tissues containing atherosclerotic plaques, has been implicated in the disease process. 2,13,14 In several of these studies, the mAb (Clone CRP-8) used to illustrate colocalization of CRP with complement predominantly recognizes mCRP. Hence, we sought to directly explore whether mCRP could also participate in complement activation. We clearly show that mCRP has the potential to interact with complement by binding to C1q and may regulate complement activation via 2 pathways, depending on whether mCRP is in fluid phase or bound to a surface. When mCRP is in ligand-free state, it exhibits complement inhibitory activity, presumably by restricting the binding of C1q with other complement activators. Because the activation of C1q needs 2 of its subunits to be closely engaged, this may account for the inability of fluid phase mCRP to further activate complement cascade after binding to C1q. When mCRP is surface immobilized either alone or bound to ox-LDL or E-LDL, it exhibits complement-activating activity of the early components of the CCP by binding to C1q, leading to the turnover of C3, largely bypassing the more inflammatory and destructive terminal sequence by recruitment of Factor H.

Together, our observations suggest a dual role for mCRP in the innate immune system. On the one hand, mCRP may facilitate the safe clearance of damaged self-materials that mCRP can bind (eg, modified LDL). On the other hand, mCRP can protect against unwanted complement activation in the fluid phase. It may be inferred from the present results that mCRP may have a protective role against complement-mediated inflammation in early stages of atherosclerosis. Unbound mCRP can inhibit complement activation by restricting the binding of C1q to CCP activators (eg, antibody-coated ox-LDL) and therefore negatively regulate the local production of inflammatory molecules generated by complement activation. On the other hand, the binding of mCRP to atherogenic LDL derivatives transform these particles into CCP activators. Because the complement activation by LDL-bound mCRP is largely restricted to early cascade, bypassing the terminal sequence, opsonic effect may play a major role in this pathway to facilitate the clearance of these potentially proatherosclerotic particles, as suggested by Bhakdi et al, concerning the interaction of CRP with E-LDL and complement. 13 However, with the progression of atherosclerosis, the above suggested protective effect of mCRP on complement mediated inflammation may be counteracted by other factors such as localized accumulation of modified LDLs, increased damage to vascular tissues, exposure of subluminal proteins, and activation of neutrophils and endothelial cells. 21,24

Acknowledgments

This work was supported by grants from National Natural Science Foundation of China (30400069 and 30500087) and Specialized Research Fund for the Doctoral Program of Higher Education (20050730028).

【参考文献】
  Seifert PS, Hugo F, Tranum Jensen J, Zahringer U, Muhly SM, Bhakdi S. Isolation and characterization of a complement-activating lipid extracted from human atherosclerotic lesions. J Exp Med. 1990; 172: 547-557.

Torzewski J, Torzewski M, Bowyer DE, Fröhlich M, Koenig W, Waltenberger J, Fitzsimmons C, Hombach V. C-reactive protein frequently colocalizes with the terminal complement complex in the intima of early atherosclerotic lesions of human coronary arteries. Arterioscler Thromb Vasc Biol. 1998; 18: 1386-1392.

Vlaicu R, Niculescu F, Rus HG, Cristea A. Immunohistochemical localization of the terminal C5b-9 complement complex in human aortic fibrous plaque. Atherosclerosis. 1985; 57: 163-177.

Yasojima K, Schwab C, McGeer EG, McGeer PL. Generation of C-reactive protein and complement components in atherosclerotic plaques. Am J Pathol. 2001; 158: 1039-1051.

Schmiedt W, Kinscherf R, Deigner H-P, Kamencic H, Nauen O, Kilo J, Oelert H, Metz J, Bhakdi S. Complement C6 deficiency protects against diet-induced atherosclerosis in rabbits. Arterioscler Thromb Vasc Biol. 1998; 18: 1790-1795.

Buono C, Come CE, Witztum JL, Maguire GF, Connelly PW, Carroll M, Lichtman AH. Influence of C3 deficiency on atherosclerosis. Circulation. 2002; 105: 3025-3031.

Patel S, Thelander EM, Hernandez M, Montenegro J, Hassing H, Burton C, Mundt S, Hermanowski-Vosatka A, Wright SD, Chao Y-S, Detmers PA. ApoE( -/- ) mice develop atherosclerosis in the absence of complement component C5. Biochem Biophys Res Commun. 2001; 286: 164-170.

Reynolds GD, Vance RP. C-reactive protein immunohistochemical localization in normal and atherosclerotic human aortas. Arch Pathol Lab Med. 1987; 111: 265-269.

Zhang YX, Cliff WJ, Schoefl GI, Higgins G. Coronary C-reactive protein distribution: its relation to development of atherosclerosis. Atherosclerosis. 1999; 145: 375-379.

Pepys MB, Hirschfield GM. C-reactive protein: a critical update. J Clin Invest. 2003; 111: 1805-1812.

Jialal I, Devaraj S, Venugopal SK. C-reactive protein: risk marker or mediator in atherothrombosis? Hypertension. 2004; 44: 6-11.

Mold C, Gewurz H, Du Clos TW. Regulation of complement activation by C-reactive protein. Immunopharmacology. 1999; 42: 23-30.

Bhakdi S, Torzewski M, Paprokta K, Schmitt S, Barsoom H, Suriyaphol P, Han S, Lackner KJ, Husmann M. Possible protective role for C-reactive protein in atherogenesis. Complement activation by modified lipoproteins halts before detrimental terminal sequence. Circulation. 2004; 109: 1870-1876.

Bhakdi S, Torzewski M, Klouche M, Hemmes M. Complement and atherogenesis. Binding of CRP to degraded, nonoxidized LDL enhances complement activation. Arterioscler Thromb Vasc Biol. 1999; 19: 2348-2354.

Chang M, Binder CJ, Torzewski M, Witztum JL. C-reactive protein binds to both oxidized LDL and apoptotic cells through recognition of a common ligand: phosphorylcholine of oxidized phospholipids. Proc Natl Acad Sci U S A. 2002; 99: 13043-13048.

Hirschfield GM, Gallimore JR, Kahan MC, Hutchinson WL, Sabin CA, Benson GM, Dhillon AP, Tennent GA, Pepys MB. Transgenic human C-reactive protein is not proatherogenic in apolipoprotein E-deficient mice. Proc Natl Acad Sci U S A. 2005; 102: 8309-8314.

Reifenberg K, Lehr HA, Baskal D, Wiese E, Schaefer SC, Black S, Samols D, Torzewski M, Lackner KJ, Husmann M, Blettner M, Bhakdi S. Role of C-reactive protein in atherogenesis. Can the apolipoprotein E knockout mouse provide the answer? Arterioscler Thromb Vasc Biol. 2005; 25: 1641-1646.

Potempa LA, Maldonado BA, Laurent P, Zemel ES, Gewurz H. Antigenic, electrophoretic and binding alterations of human C-reactive protein modified selectively in the absence of calcium. Mol Immunol. 1983; 20: 1165-1175.

Zouki C, Haas B, Chan JSD, Potempa LA, Filep JG. Loss of pentameric symmetry of C-reactive protein is associated with promotion of neutrophil-endothelial cell adhesion. J Immunol. 2001; 167: 5355-5361.

Khreiss T, Jozsef L, Hossain S, Chan JSD, Potempa LA, Filep JG. Loss of pentameric symmetry of C-reactive protein is associated with delayed apoptosis of human neutrophils. J Biol Chem. 2002; 277: 40775-40781.

Khreiss T, József L, Potempa LA, Filep JG. Conformational rearrangement in C-reactive protein is required for proinflammatory actions on human endothelial cells. Circulation. 2004; 109: 2016-2022.

Verma S, Szmitko PE, Yeh ETH. C-reactive protein? Structure affects function. Circulation. 2004; 109: 1914-1917.

Schwedler SB, Amann K, Wernicke K, Krebs A, Nauck M, Wanner C, Potempa LA, Galle J. Native C-reactive protein (CRP) increases, whereas modified CRP reduces atherosclerosis in ApoE-knockout-mice. Circulation. 2005; 112: 1016-1023.

Khreiss T, Jozsef L, Potempa LA, Filep JG. Loss of pentameric symmetry in C-reactive protein induces interleukin-8 secretion through peroxynitrite signaling in human neutrophils. Circ Res. 2005; 97: 690-697.

Devaraj S, Venugopal S, Jialal I. Native pentameric C-reactive protein displays more potent pro-atherogenic activities in human aortic endothelial cells than modified C-reactive protein. Atherosclerosis. 2005; 184: 48-52.

Diehl EE, Haines GK, Radosevich JA, Potempa LA. Immunohistochemical localization of modified C-reactive protein antigen in normal vascular tissue. Am J Med Sci. 2000; 319: 79-83.

Jabs WJ, Theissing E, Nitschke M, Bechtel JF, Duchrow M, Mohamed S, Jahrbeck B, Sievers HH, Steinhoff J, Bartels C. Local generation of C-reactive protein in diseased coronary venous bypass grafts and normal vascular tissue. Circulation. 2003; 108: 1428-1431.

Ouchi N, Kihara S, Funahashi T, Nakamura T, Nishida M, Kumada M, Okamoto Y, Ohashi K, Nagaretani H, Kishida K, Nishizawa H, Maeda N, Kobayashi H, Hiraoka H, Matsuzawa Y. Reciprocal association of C-reactive protein with adiponectin in blood stream and adipose tissue. Circulation. 2003; 107: 671-674.

Ciubotaru I, Potempa LA, Wander RC. Production of modified C-reactive protein in U937-derived macrophages. Exp Biol Med (Maywood). 2005; 230: 762-770.

Jiang H, Siegel JN, Gewurz H. 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. 1991; 146: 2324-2330.

Black S, Agrawal A, Samols D. The phosphocholine and the polycation-binding sites on rabbit C-reactive protein are structurally and functionally distinct. Mol Immunol. 2003; 39: 1045-1054.

Ying S, Gewurz H, Kinoshita CM, Potempa LA, Gewurz H. Identification and partial characterization of multiple native and neoantigenic epitopes of human C-reactive protein by using monoclonal antibodies. J Immunol. 1989; 143: 221-228.

Potempa LA, Siegel JN, Fiedel BA, Potempa RT, Gewurz H. Expression, detection, and assay of neoantigen (Neo-CRP) associated with a free, human C-reactive protein subunit. Mol Immunol. 1987; 24: 531-541.

Nauta AJ, Bottazzi B, Mantovani A, Salvatori G, Kishore U, Schwaeble WJ, Gingras AR, Tzima S, Vivanco F, Egido J, Tijsma O, Hack EC, Daha MR, Roos A. Biochemical and functional characterization of the interaction between pentraxin 3 and C1q. Eur J Immunol. 2003; 33: 465-473.

Ji SR, Wu Y, Potempa LA, Qiu Q, Zhao J. Interactions of C-reactive protein with low density lipoproteins: implications for an active role of modified C-reactive protein in atherosclerosis. Int J Biochem Cell Biol. 2006; 38: 648-661.

Vigushin DM, Pepys MB, Hawkins PN. Metabolic and scintigraphic studies of radioiodinated human C-reactive protein in health and disease. J Clin Invest. 1993; 91: 1351-1357.

作者单位：Shang-Rong Ji; Yi Wu; Lawrence A. Potempa; Yu-Heng Liang; Jing ZhaoFrom the Institute of Biophysics (S.-R.J., Y.W., Y.-H.L., J.Z.), Lanzhou University, P. R. China; and Immtech International Inc. (L.A.P.), Vernon Hills, Ill.