Native C-Reactive Protein Increases Whereas Modified C-Reactive Protein Reduces Atherosclerosis in Apolipoprotein E–Knockout Mice
http://www.100md.com
《循环学杂志》
the Department of Medicine, Division of Nephrology, University of Würzburg, Würzburg, Germany (S.B.S., C.W., J.G.)
Institute of Pathology, University of Erlangen-Nürnberg, Erlangen, Germany (K.A., K.W.)
Clinical Biochemistry, University of Greifswald, Greifswald, Germany (A.K., M.N.)
Immtech International, Inc, Vernon Hills, Ill (L.A.P.).
Abstract
Background— C-reactive protein (CRP) may have proatherogenic but also vasoprotective properties. We tested the hypothesis that the configuration of CRP (pentameric, or native [nCRP], versus monomeric, or modified [mCRP]) determines these different characteristics in an in vivo model.
Methods and Results— We investigated the effects of human nCRP and mCRP on the development of atherosclerosis in apolipoprotein E–knockout (ApoE–/–) mice. Treatment with nCRP for 8 weeks (2.5 mg/kg SC weekly) resulted in a 4-fold-higher mean aortic plaque area in 14-week-old female ApoE–/– mice compared with the saline controls. In contrast, mean plaque size was decreased by 50% in mCRP-treated ApoE–/– mice (2.5 mg/kg SC weekly). Using immunohistochemistry, we report the natural presence of the mCRP antigen in saline controls. mCRP antigen was expressed in smooth muscle cells and extracellularly in the vicinity of the plaques to a similar level in both CRP-treated groups and saline controls. mCRP and ApoB colocalized with macrophages and were equally upregulated in all aortic plaques. Vascular cell adhesion molecule expression was increased, and CD154 and intercellular adhesion molecule showed a trend for higher expression in nCRP-treated compared with mCRP-treated mice. CD154 expression in the vessel wall and plaque size correlated significantly. mCRP-treated ApoE–/– exhibited higher serum levels of the antiinflammatory interleukin-10 compared with the other 2 groups.
Conclusions— Here, we show that mCRP and nCRP have opposite effects on atherosclerosis in ApoE–/– mice. These data may explain in part the conflicting activities previously reported for CRP in models of atherogenesis.
Key Words: atherosclerosis ; C-reactive protein ; immunohistochemistry ; inflammation ; mice, knockout
Introduction
C-reactive protein (CRP) represents one of the strongest independent predictors in cardiovascular disease in several settings.1,2 Originally considered to be an innocent bystander, numerous reports suggest that it plays an active part in the development of vascular inflammation (for review, see elsewhere3,4). Proinflammatory effects on cardiovascular cells include activation of the complement system,5 decrease in NO release,6 and upregulation of cell adhesion molecules.7 However, CRP also elicits effects considered to be vasoprotective such as endothelium-independent vasorelaxing effects in human vessels in vitro.8 Verma et al6 showed that CRP inhibited both basal and vascular endothelial growth factor (VEGF)–stimulated angiogenesis, which would in turn inhibit plaque neovascularization.
These opposing observations may result partly from differences in CRP configuration. The classically studied serum CRP is reported to be a pentamer composed of 5 identical globular subunits arranged as an annular disk (referred to here as native [nCRP]).9 nCRP can undergo subunit dissociation into individual monomeric units, eg, when associating with a cell membrane.10 These subunits undergo a conformational change that significantly modifies CRP structure, solubility, and antigenicity. This form of CRP is called modified or monomeric CRP (mCRP). mCRP is a naturally occurring stable protein but, in contrast to nCRP, is not found in serum but in fibrous tissues of normal human blood vessel intima.11 The formation of mCRP from CRP is nonproteolytic and irreversible.12 Numerous studies have identified CRP mRNA expression in many different extrahepatic cells such as islet cells of the pancreas,13 neurons,14 adipocytes,15 and renal tubular epithelial cells.16 Whether the "CRP" that is translated at each site is the nCRP conformer or the mCRP conformer needs to be clarified.
Recently, Paul et al17 showed that human CRP transgene expression caused accelerated atherosclerosis in apolipoprotein E–knockout (ApoE–/–) mice, a hypercholesterolemic mouse model of accelerated atherosclerosis. The aim of the present study was to compare the effects of human nCRP and mCRP on the development of early atherosclerosis by using directly injected proteins, thereby testing the hypothesis that the configuration of CRP may determine different effects on atherosclerosis. Using immunohistochemistry, we studied proinflammatory mediators thought to initiate atherosclerosis such as vascular cell adhesion molecule (VCAM), intracellular adhesion molecule (ICAM), and the CD40 ligand CD154, a member of the tumor necrosis factor gene superfamily. Furthermore, to elucidate the role of CRP within the plaque, immunostainings for mCRP, nCRP, ApoB, and macrophages were performed. Finally, cytokines in the serum that activate (interferon [IFN]-) and inhibit (interleukin [IL]-10) macrophages were measured.
Methods
Animal Model
Female homozygous ApoE–/– mice and their genetic background mice (C57BL/6) were obtained from Jackson Laboratories (Bar Harbor, Maine) at 6 weeks of age. They were maintained in a light- and temperature-controlled environment and had free access to water and standard laboratory chow containing 4% fat (Altromin). Weight-matched ApoE–/– and C57BL/6 mice (n=7) were randomly assigned to 3 treatment groups: Group 1 received human nCRP 2.5 mg/kg SC; group 2 received a recombinant form of human mCRP 2.5 mg/kg SC; and group 3 received 0.9% saline as control (identical volume given subcutaneously). Treatment was applied once weekly for 8 weeks. Animal weight was determined every week. Before the fifth injection, 50 μL whole blood was collected from the tail vein for white blood cell count.
At 14 weeks of age, mice were anesthetized with ketamine (100 mg/kg) and xylazine-HCl (4 mg/kg), and blood samples (between 200 and 250 μL) were collected from the vena cava for lipid analysis. Heart and aorta were carefully removed and fixed in 10% buffered formalin for morphometric analysis and immunohistochemistry. All experiments and measurements were performed in accordance with national guidelines for the care and use of research animals.
CRP Preparations
nCRP
Crude CRP was purchased from SeraCare Inc. It was bound to Q-Sepharose FF and was eluted with a step gradient of NaCl. Eluted protein was pooled, and CaCl2 was immediately added to a concentration of 2 mmol/L. The ion exchange step with CRP preparations is critically important not only to reduce endotoxin but also to remove any mCRP protein that may be present in an nCRP preparation. mCRP will strongly bind to the agarose resin. It resists elution with increasing salt concentration but elutes with 5 mol/L guanidinium hydrochloride. Eluted nCRP protein was dialyzed against 25 mmol/L-Tris-HCl and 0.15 mol/L NaCl (pH 7.4) containing 2 mmol/L CaCl2, sterile filtered, and stored at 4°C. Storage in calcium-containing buffers is critical to prevent the spontaneous formation of mCRP from nCRP.
mCRP
mCRP was prepared from nCRP by urea chelation as described previously.18 By electron microscopic analysis, mCRP molecules in the absence of added salt associate into a diffuse matrix distinct from the annular pentameric disk that defines nCRP. In a normal saline solution, mCRP forms an insoluble suspension of self-aggregated subunits. A recombinant form of mCRP (rmCRP) with both cysteine residues mutated to alanine residues and with an added N-terminal formylmethionine residue was expressed in Escherichia coli and isolated from inclusion bodies to >95% purity. To facilitate purification of rmCRP, it was acylated with citraconic anhydride to enhance solubility (C-rmCRP). C-rmCRP was directly comparable to mCRP as analyzed by SDS-PAGE, amino acid composition, and N-terminal sequence analysis. Once purified, the citraconyl groups were removed from C-rmCRP by dialysis at room temperature for 20 hours in 0.1 mol/L citrate (pH 3.5). The decitraconylated protein (Cx-rmCRP) was dialyzed extensively at 4°C against 25 mmol/L sodium PBS without calcium (pH 7.4), yielding an insoluble suspended protein similar to mCRP in saline-based buffers. This form was used in the present study. Monoclonal antibodies directed against the C-terminal octapeptide of the CRP subunit (mAb 9C9), which is expressed only in mCRP and not in nCRP, reacted with the same specificity and affinity with mCRP, C-rmCRP, and Cx-rmCRP.
Biochemistry
White blood cell counts were determined automatically with the Coulter method (Beckman) designed for small sample volumes. Serum cholesterol and triglycerides were measured enzymatically. For quantification of lipoprotein cholesterol, 30 μL serum was subjected to agarose gel electrophoresis, followed by enzymatic staining of cholesterol as previously described.19 IFN- and IL-10 levels in the serum of ApoE–/– mice were measured by ELISA in duplicates using commercially available kits (Mouse Quantikine ELISA kits, R&D Systems Inc).
Morphological Evaluation
Morphometric analysis of the aorta was performed by planimetry and a semiautomatic image analyzing system (AnalysisPRO, SIS), combined with a counting grid eyepiece for defined movement over the section plane (Zeiss Co) at a magnification of x400 as described previously.20 Contours of aortic plaques in the aortic root through the aortic valve and in the aortic arch were marked manually with a cursor, and plaque area was automatically calculated. All investigations were performed in a blinded manner.
Immunohistochemistry
Immunohistochemistry was performed on paraffin sections with the following antibodies: rabbit anti–human-CD154 (Santa Cruz Biotechnology, 1:600), rabbit anti–mouse-ApoB48/100 (USBiological, 1:500), rat anti–mouse-macrophages/monocytes (MAB1852, Chemicon International Inc, 1:25), goat anti–human-VCAM-1 (Santa Cruz Biotechnology, 1:300), goat anti–mouse-ICAM-1 (Santa Cruz Biotechnology, 1:100), and 2 mouse anti–human-CRP antibodies as previously described,21 which were provided by L.A. Potempa: mAb 9C9, which reacts exclusively with mCRP, and mAb 8D8, showing only reactivity with nCRP. Briefly, freshly cut 4-μm sections were mounted on silane-coated slides, deparaffinized, and rehydrated through a graded ethanol series, followed by blocking of nonspecific binding with 3% H2O2 (20 minutes, room temperature). An antigen retrieval was performed by heating the slides in citrate buffer (20 minutes in a microwave at 900 W). For antigen visualization, the CSA Dako and Goat Vectastain Elite ABC kits with AEC as the substrate (CD154, VCAM, and ICAM stainings, red deposits), the Rat and Rabbit Vectastain Elite ABC, and the MOM Vectastain kits (Vector Labs) with DAB and histogreen as substrates (CRP and macrophage stainings, brown deposits; ApoB staining, green deposits) were used. Color development was stopped by adding water, and sections were finally counterstained with hematoxylin. Negative controls were performed by omitting the primary antibody. Specific blocking experiments were carried out by preincubating mAb 9C9 and 8D8 with a 10-fold excess of the respective antigens. A double staining was performed with mAb 9C9 as the first and anti-ApoB as the second antibody. The sections were examined under light microscopy at a magnification of x400 with a semiquantitative scoring system (0 to 4): 0=no expression, 1=mild expression, 2=moderate expression, 3=strong expression, and 4=very strong expression. At least 5 animals per group were stained. All analyses were performed in a blinded manner.
Statistical Analysis
Data are presented as mean±SEM. For comparisons between treatment groups, the Kruskal-Wallis H test for nonnormally distributed variables was performed. Bivariate regression and correlation analyses were performed by use of the Spearman rank method. A value of P<0.05 was considered statistically significant. Data analysis was performed with the SPSS/PC+ package (SPSS Inc).
Results
Animal Data
Initial body weights were not different between groups. At the end of the study, no differences in body weights were detectable among the ApoE–/– groups (17.1±1.6, 17.3±1.6, and 17.3±1.6 g for saline, nCRP, and mCRP, respectively), whereas mCRP-treated C57BL/6 control mice (18.8±1.8 g) were slightly heavier compared with the other 2 groups (saline and nCRP, both 17.9±1.7 g; P<0.05). Other 3-month-old ApoE–/– mice in the animal facility receiving a normal chow had similar body weights. Animals treated with either form of CRP showed no difference in activity or behavior compared with their controls.
Biochemistry
White blood cell counts were identical in all ApoE–/– and C57BL/6 groups (data not shown), indicating that no treatment induced a systemically detectable inflammatory response. In all ApoE–/– groups, IFN- levels in the serum were below the detection limit of 4 pg/mL. In contrast, all ApoE–/– mice had detectable serum IL-10 levels >4 pg/mL (14±3, 10±2, and 26±11 pg/mL for saline, nCRP, and mCRP, respectively; range, 5 to 92 pg/mL). All control C57BL/6 mice had IL-10 levels >20 pg/mL (35±4 pg/mL; range, 21 to 49 pg/mL; P<0.01 between control C57BL/6 and control ApoE–/– mice). In Table 1, results are presented according to low (<10 pg/mL), middle (10 and <20 pg/mL), and high ( 20 pg/mL) IL-10 levels. Significant differences were observed between groups of ApoE–/– mice (P<0.05). Although control ApoE–/– mice had equally distributed IL-10 levels, all mCRP-treated ApoE–/– mice exhibited IL-10 levels in the middle and upper ranges, whereas IL-10 levels of all nCRP-treated ApoE–/– mice were <20 pg/mL. Total cholesterol and triglycerides levels, as well as the lipoprotein profiles, were unaffected by either nCRP or mCRP treatment in various groups of ApoE–/– mice (Table 2).
Histology and Morphology
Aortic Root
In the aortic root, all ApoE–/– mice developed early atherosclerotic lesions characterized by the predominance of foam cells accumulating in the subendothelial space, with no obvious changes in the underlying intima (Figure 1a through 1c). However, treatment with nCRP resulted in 4-fold-higher mean aortic plaque area in the aortic root in ApoE–/– mice compared with controls (38.6±18.3 versus 9.7±4.0x103 μm2), and 2 of 7 animals had much larger aortic plaques of 62.5 and 136.9x103 μm2. In contrast, mean plaque area was decreased by 50% in ApoE–/– mice treated with mCRP compared with controls (4.2±1.3 versus 9.7±4.0x103 μm2). Differences between the 3 groups were statistically significant (P<0.05; Figure 2).
Aortic Arch
The same trend within the groups was observed when plaque area in the aortic arch was quantified, although statistical significance was not reached (4.0±1.4 and 1.5±0.9 versus 2.8±0.8x103 μm2 for nCRP, mCRP, and saline, respectively; 0.2>P>0.05). No lesions were detectable in the control C57BL/6 mice.
Immunohistochemistry
Discussion
In this study, we show that 2 clearly defined different conformers of human CRP cause opposite effects in an animal model of early atherosclerosis. nCRP increased and mCRP reduced plaque formation in ApoE–/– mice with early lesions, processes that may involve the production of the antiinflammatory cytokine IL-10, the proinflammatory cytokine CD154, and the expression of VCAM and ICAM.
The role that CRP plays in vivo with respect to acceleration or inhibition of atherogenesis is an important issue. In recent years, a multitude of in vitro data have been generated showing proatherogenic effects of CRP (for review, see elsewhere3,4). We propose that it is essential to address several methodical aspects in interpreting these data. First, previous studies have neither discussed nor distinguished the observed effects in terms of 2 distinct conformational forms of CRP. Because prolonged storage of purified CRP in the absence of calcium or in the presence of chelating agents will cause a spontaneous conversion of nCRP to mCRP, this issue is most relevant.22 Because mCRP does not require fragmentation of the nCRP protein, an SDS-PAGE assay to define the "purity" of a nCRP preparation would be insufficient to prove that there is no mCRP in such preparations. Second, many, if not all, CRP-polyclonal antisera contain specificity to both the nCRP antigen and mCRP antigen.23 Also, monoclonal anti-CRP– clone 8 has clearly been shown to detect predominantly the mCRP protein and not the nCRP protein.24 Hence, any immunohistological studies evaluating tissue-associated CRP may, in fact, more accurately report on the expression of tissue associated mCRP. Third, many previously reported studies using CRP used test protein concentrations >5 μg/mL. Using high concentrations of CRP protein would favor the effects of any minor component found in any isolated preparation. In the present study, we tried to take these limiting factors into consideration.
Three recently published in vivo studies support an active role of human CRP in vascular injury in mice. First, arterial injury in human CRP-transgenic mice resulted in a higher rate of thrombosis.25 Second, human CRP-transgenic mice crossed with ApoE–/– mice developed accelerated atherosclerosis.17 Third, in rats, injection of human CRP enhanced tissue damage in myocardial infarction.26 The last 2 studies described immunohistochemical expression of human CRP in the tissue. Our finding that nCRP accelerated plaque formation in ApoE–/– mice is in line with the results of the above-mentioned models. However, our model differs in several important ways. Concern has been addressed through the use of heterologous systems.27 Exposing mice to huge amounts of human CRP might indeed not be physiological because mouse CRP is only a trace protein, not an acute-phase reactant. Taking this into consideration, we designed a model of low dosing over a long period of time. Our results most likely do not reflect systemic CRP levels and the direct effect of circulating CRP on atherogenic plaque formation. Kresl et al28 reported that when mice were injected with 100 μg IV (5 mg/kg) of mCRP, little or no mCRP was detected in a number of serum specimens. Because mCRP is hydrophobic and thus tends to self-aggregate, its movement away from the injection site as a free protein is unlikely.
The mCRP antigenic reactivity found in the plaques of all mouse aortas could reflect the natural presence of mouse mCRP. We did not directly determine to what extent the mAb 9C9 antibody also recognizes mouse "CRP" epitopes. But because the primary structure of mouse CRP is known to be 70% homologous to human CRP29 and residues 201 to 205 (KPQLW), which define the epitope for mAb 9C9 specificity,21 are conserved in human and mouse CRP,9 it is highly likely that a naturally occurring form of mouse mCRP exists. The facts that saline-treated ApoE–/– mice also stained positively and that antibody reactivity could be completely adsorbed with the human mCRP protein also suggest that human and mouse proteins are structurally similar. Our animal model identifies, for the first time, the existence of a mouse mCRP protein that is found at vessel walls and sites of atherogenic plaque buildup. Furthermore, plaque development is regulated by injection of human CRP proteins at distant tissue sites.
Staining of mCRP could be attributed to the cytoplasma of foam cells. In an in vitro study, it was proposed that CRP aggregates may bind and cluster LDL particles phagocytized by macrophages, leading to foam cell formation.30 The colocalization of mCRP, ApoB, and macrophages found here supports this hypothesis. Comparison of nCRP and mCRP staining in the plaque showing a predominance of mCRP reactivity within the foam cells suggests that CRP aggregates are related to the modified isoform of CRP. Extrahepatic expression of CRP in smooth muscle cells has been described.31,32 However, whether mCRP is locally produced or derived from serum CRP needs to be clarified. The lack of mCRP staining in normal vessel sections without plaque formation is in contrast to previous findings in humans in whom it is constitutively expressed in several healthy tissues.11 Further studies are needed to determine the exact role of mouse mCRP at the site of plaque development. However, the presence of mCRP in the vessel wall as soon as plaques develop suggests an important role in early atherogenesis in mice.
Extending previous work by Paul et al,17 who showed proatherogenic effects of CRP in ApoE–/– mice with well-established lesions, our animal model starting with CRP injections as early as 6 weeks, before lesions were established, suggests effects on early atherogenesis. To provide further evidence for this assumption, we studied proinflammatory pathways involved in the initiation of the disease, ie, expression of VCAM and ICAM, which regulate leukocyte adhesion to the endothelium,33 and the expression of the CD40 ligand CD154, which is described in endothelial and smooth muscle cells, macrophages, platelets, and T lymphocytes and thought to contribute to the initial recruitment of inflammatory cells to damaged endothelium (for review, see work by Schonbeck and Libby34). It was shown that the disruption of the CD154-CD40 system retarded the initiation of arterial plaque formation.35 We observed a trend toward lower CD154 expression in endothelium and smooth muscle cells in mCRP-treated ApoE–/– mice compared with the other 2 groups. In addition, CD154 expression correlated with plaque size. Treatment with human mCRP at a distant injection site might have reduced the tumor necrosis factor-–related inflammatory process at the aortic site. The observation that enhanced plaque formation induced by nCRP was paralleled by increased VCAM and a trend toward increased ICAM expression supports the critical role of ICAM and VCAM found by others in hypercholesterolemic animal models.36,37
At first glance, the results presented here appear to be contradictory to reports in vitro. In those studies, mCRP triggered proinflammatory activities in isolated human neutrophils and endothelial cells, including increased production of monocyte chemoattractant protein-1, IL-8, ICAM, VCAM, and E-selectin.18,22,38 In contrast, nCRP was without detectable effects on these signaling cascades. It was suggested that endothelial injury might expose the naturally occurring mCRP, leading to the attraction of neutrophils to the area of injury. Alternatively, mCRP may also be formed from nCRP at sites of injury or infection by some as-yet-unknown mechanism as part of the acute activation and/or resolution of the inflammatory process.39
The fact that human mCRP-treated mice showed less plaque is not necessarily in contradiction to the aforementioned studies that addressed immediate responses (over minutes to hours) using isolated cells. The long-term effect of a stimulatory protein on the overall immune response in a live animal cannot be predicted precisely from isolated systems. Such opposite stimulatory and inhibitory actions, depending on the dose used, have been described for anti-CD3 monoclonal antibody effects on the T-cell receptor in vivo.40 In an animal model of systemic lupus erythematodes, a single injection of 200 μg SC of CRP prevented or reversed lupus nephritis.41 Because IL-10 was required for this protection, we studied serum levels of this antiinflammatory cytokine with potent deactivating properties on both macrophages and T lymphocytes.42 Although ApoE–/– mice generally had lower IL-10 levels than the C57BL/6 control mice, mCRP-treated ApoE–/– mice had higher IL-10 levels than the other 2 groups. Therefore, mCRP injections could be beneficial through upregulation of IL-10. Namiki et al43 proposed IL-10 as a new therapeutic tool for the treatment of atherosclerosis. Gene transfer of IL-10 cDNA reduced atherosclerosis in male ApoE–/– mice and decreased serum levels of IFN- associated with macrophage activation. Seven days after the last injection, we were unable to detect IFN- in either ApoE–/– or C57BL/6 mice. Because of the limited availability of serum material, it was not possible to measure other proinflammatory or antiinflammatory cytokines.
In summary, we describe a model of early atherogenesis in which human nCRP accelerated atherosclerosis and mCRP partly prevented plaque formation. Results of IL-10 measurements are consistent with the interpretation that small amounts of mCRP could heighten normal immune surveillance in the mouse, slowing the process of atherosclerotic plaque formation. Immunostainings for CD154, VCAM, and ICAM may indicate that nCRP and mCRP influence the initial steps of atherogenesis differentially. Further studies are needed to elucidate the signaling cascade of mCRP and its relevance to the processes of atherosclerosis in humans.
Acknowledgments
For this project, Dr Schwedler received a 2002 scholarship from the Deutsche Nierenstiftung. She is currently funded by a research program of the University Würzburg (HWP, Programm "Chancengleichheit für Frauen in Forschung und Lehre"). This work was also supported by the Deutsche Forschungsgemeinschaft (grants Ga431/5-3 and SFB 355, TP B11), and the IZKF Erlangen (Interdisziplin;res Zentrum für Klinische Forschung, project A11). We thank Marita Bartrow, Carmen Bauer, Monika Klewer, and Stefan S;llner for technical assistance.
Disclosure
Dr Potempa was employed by NextEra Therapeutics, Inc, and is currently employed by Immtech International, Inc. NextEra and Immtech are biotechnology companies that (in aggregate) are the primary owners of intellectual property developed on the modified form of C-reactive protein.
References
Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation. 2003; 107: 363–369.
Yeh ET, Willerson JT. Coming of age of C-reactive protein: using inflammation markers in cardiology. Circulation. 2003; 107: 370–371.
Jialal I, Devaraj S, Venugopal SK. C-reactive protein: risk marker or mediator in atherothrombosis; Hypertension. 2004; 44: 6–11.
Mazer SP, Rabbani LE. Evidence for C-reactive protein’s role in (CRP) vascular disease: atherothrombosis, immuno-regulation and CRP. J Thromb Thrombolysis. 2004; 17: 95–105.
Wolbink GJ, Brouwer MC, Buysmann S, ten Berge IJ, Hack CE. CRP-mediated activation of complement in vivo: assessment by measuring circulating complement-C-reactive protein complexes. J Immunol. 1996; 157: 473–479.
Verma S, Wang CH, Li SH, Dumont AS, Fedak PW, Badiwala MV, Dhillon B, Weisel RD, Li RK, Mickle DA, Stewart DJ. A self-fulfilling prophecy: C-reactive protein attenuates nitric oxide production and inhibits angiogenesis. Circulation. 2002; 106: 913–919.
Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation. 2000; 102: 2165–2168.
Sternik L, Samee S, Schaff HV, Zehr KJ, Lerman LO, Holmes DR, Herrmann J, Lerman A. C-reactive protein relaxes human vessels in vitro. Arterioscler Thromb Vasc Biol. 2002; 22: 1865–1868.
Shrive AK, Cheetham GM, Holden D, Myles DA, Turnell WG, Volanakis JE, Pepys MB, Bloomer AC, Greenhough TJ. Three dimensional structure of human C-reactive protein. Nat Struct Biol. 1996; 3: 346–354.
Wu Y, Wang HW, Ji SR, Sui SF. Two-dimensional crystallization of rabbit C-reactive protein monomeric subunits. Acta Crystallogr D Biol Crystallogr. 2003; 59: 922–926.
Diehl EE, Haines GK, III, Radosevich JA, Potempa LA. Immunohistochemical localization of modified C-reactive protein antigen in normal vascular tissue. Am J Med Sci. 2000; 319: 79–83.
Kresl JJ, Potempa LA, Anderson BE. Conversion of native oligomeric to a modified monomeric form of human C-reactive protein. Int J Biochem Cell Biol. 1998; 30: 1415–1426.
Fehsel K, Plewe D, Kolb-Bachofen V. Nitric oxide-induced expression of C-reactive protein in islet cells as a very early marker for islet stress in the rat pancreas. Nitric Oxide. 1997; 1: 254–262.
Yasojima K, Schwab C, McGeer EG, McGeer PL. Human neurons generate C-reactive protein and amyloid P: upregulation in Alzheimer’s disease. Brain Res. 2000; 887: 80–89.
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.
Jabs WJ, Logering BA, Gerke P, Kreft B, Wolber EM, Klinger MH, Fricke L, Steinhoff J. The kidney as a second site of human C-reactive protein formation in vivo. Eur J Immunol. 2003; 33: 152–161.
Paul A, Ko KW, Li L, Yechoor V, McCrory MA, Szalai AJ, Chan L. C-reactive protein accelerates the progression of atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2004; 109: 647–655.
Khreiss T, Jozsef L, Hossain S, Chan JS, 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.
Nauck M, Winkler K, M;rz W, Wieland H. Quantitative determination of high-, low-, and very-low-density lipoproteins and lipoprotein(a) by agarose gel electrophoresis and enzymatic cholesterol staining. Clin Chem. 1995; 41: 1761–1767.
Nabokov AV, Amann K, Wessels S, Munter K, Wagner J, Ritz E. Endothelin receptor antagonists influence cardiovascular morphology in uremic rats. Kidney Int. 1999; 55: 512–519.
Ying SC, Gewurz H, Kinoshita CM, Potempa LA, Siegel JN. 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.
Zouki C, Haas B, Chan JS, 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.
Samberg NL, Bray RA, Gewurz H, Landay AL, Potempa LA. Preferential expression of neo-CRP epitopes on the surface of human peripheral blood lymphocytes. Cell Immunol. 1988; 116: 86–98.
Schwedler SB, Guderian F, D;mmrich J, Potempa LA, Wanner C. Tubular staining of modified C-reactive protein in diabetic chronic kidney disease. Nephrol Dial Transplant. 2003; 18: 2300–2307.
Danenberg HD, Szalai AJ, Swaminathan RV, Peng L, Chen Z, Seifert P, Fay WP, Simon DI, Edelman ER. Increased thrombosis after arterial injury in human C-reactive protein–transgenic mice. Circulation. 2003; 108: 512–515.
Griselli M, Herbert J, Hutchinson WL, Taylor KM, Sohail M, Krausz T, Pepys MB. C-reactive protein and complement are important mediators of tissue damage in acute myocardial infarction. J Exp Med. 1999; 190: 1733–1740.
Pepys MB, Hirschfield GM. C-reactive protein: a critical update. J Clin Invest. 2003; 111: 1805–1812.
Kresl JJ, Potempa LA, Anderson B, Radosevich JA. Inhibition of mouse mammary adenocarcinoma (EMT6) growth and metastases in mice by a modified form of C-reactive protein. Tumour Biol. 1999; 20: 72–87.
Pepys MB, Dash AC, Fletcher TC, Richardson N, Munn EA, Feinstein A. Analogues in other mammals and in fish of human plasma proteins, C-reactive protein and amyloid P component. Nature. 1978; 273: 168–170.
Fu T, Borensztajn J. Macrophage uptake of low-density lipoprotein bound to aggregated C-reactive protein: possible mechanism of foam-cell formation in atherosclerotic lesions. Biochem J. 2002; 366: 195–201.
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.
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 artery venous bypass grafts and normal vascular tissue. Circulation. 2003; 108: 1428–1431.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Schonbeck U, Libby P. The CD40/CD154 receptor/ligand dyad. Cell Mol Life Sci. 2001; 58: 4–43.
Schonbeck U, Sukhova GK, Shimizu K, Mach F, Libby P. Inhibition of CD40 signaling limits evolution of established atherosclerosis in mice. Proc Natl Acad Sci U S A. 2000; 97: 7458–7463.
Collins RG, Velji R, Guevara NV, Hicks MJ, Chan L, Beaudet AL. P-Selectin or intercellular adhesion molecule (ICAM)-1 deficiency substantially protects against atherosclerosis in apolipoprotein E-deficient mice. J Exp Med. 2000; 191: 189–194.
Cybulsky MI, Iiyama K, Li H, Zhu S, Chen M, Iiyama M, Davis V, Gutierrez-Ramos JC, Connelly PW, Milstone DS. A major role for VCAM-1, but not ICAM-1, in early atherosclerosis. J Clin Invest. 2001; 107: 1255–1262.
Khreiss T, Jozsef 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 ET. C-reactive protein: structure affects function. Circulation. 2004; 109: 1914–1917.
Nakane A, Okamoto M, Asano M, Kohanawa M, Minagawa T. An anti-CD3 monoclonal antibody protects mice against a lethal infection with Listeria monocytogenes through induction of endogenous cytokines. Infect Immun. 1993; 61: 2786–2792.
Rodriguez W, Mold C, Kataranovski M, Hutt J, Marnell LL, Du Clos TW. Reversal of ongoing proteinuria in autoimmune mice by treatment with C-reactive protein. Arthritis Rheum. 2005; 52: 642–650.
de Vries JE. Immunosuppressive and anti-inflammatory properties of interleukin 10. Ann Med. 1995; 27: 537–541.
Namiki M, Kawashima S, Yamashita T, Ozaki M, Sakoda T, Inoue N, Hirata K, Morishita R, Kaneda Y, Yokoyama M. Intramuscular gene transfer of interleukin-10 cDNA reduces atherosclerosis in apolipoprotein E-knockout mice. Atherosclerosis. 2004; 172: 21–29.(Susanne B. Schwedler, MD;)
Institute of Pathology, University of Erlangen-Nürnberg, Erlangen, Germany (K.A., K.W.)
Clinical Biochemistry, University of Greifswald, Greifswald, Germany (A.K., M.N.)
Immtech International, Inc, Vernon Hills, Ill (L.A.P.).
Abstract
Background— C-reactive protein (CRP) may have proatherogenic but also vasoprotective properties. We tested the hypothesis that the configuration of CRP (pentameric, or native [nCRP], versus monomeric, or modified [mCRP]) determines these different characteristics in an in vivo model.
Methods and Results— We investigated the effects of human nCRP and mCRP on the development of atherosclerosis in apolipoprotein E–knockout (ApoE–/–) mice. Treatment with nCRP for 8 weeks (2.5 mg/kg SC weekly) resulted in a 4-fold-higher mean aortic plaque area in 14-week-old female ApoE–/– mice compared with the saline controls. In contrast, mean plaque size was decreased by 50% in mCRP-treated ApoE–/– mice (2.5 mg/kg SC weekly). Using immunohistochemistry, we report the natural presence of the mCRP antigen in saline controls. mCRP antigen was expressed in smooth muscle cells and extracellularly in the vicinity of the plaques to a similar level in both CRP-treated groups and saline controls. mCRP and ApoB colocalized with macrophages and were equally upregulated in all aortic plaques. Vascular cell adhesion molecule expression was increased, and CD154 and intercellular adhesion molecule showed a trend for higher expression in nCRP-treated compared with mCRP-treated mice. CD154 expression in the vessel wall and plaque size correlated significantly. mCRP-treated ApoE–/– exhibited higher serum levels of the antiinflammatory interleukin-10 compared with the other 2 groups.
Conclusions— Here, we show that mCRP and nCRP have opposite effects on atherosclerosis in ApoE–/– mice. These data may explain in part the conflicting activities previously reported for CRP in models of atherogenesis.
Key Words: atherosclerosis ; C-reactive protein ; immunohistochemistry ; inflammation ; mice, knockout
Introduction
C-reactive protein (CRP) represents one of the strongest independent predictors in cardiovascular disease in several settings.1,2 Originally considered to be an innocent bystander, numerous reports suggest that it plays an active part in the development of vascular inflammation (for review, see elsewhere3,4). Proinflammatory effects on cardiovascular cells include activation of the complement system,5 decrease in NO release,6 and upregulation of cell adhesion molecules.7 However, CRP also elicits effects considered to be vasoprotective such as endothelium-independent vasorelaxing effects in human vessels in vitro.8 Verma et al6 showed that CRP inhibited both basal and vascular endothelial growth factor (VEGF)–stimulated angiogenesis, which would in turn inhibit plaque neovascularization.
These opposing observations may result partly from differences in CRP configuration. The classically studied serum CRP is reported to be a pentamer composed of 5 identical globular subunits arranged as an annular disk (referred to here as native [nCRP]).9 nCRP can undergo subunit dissociation into individual monomeric units, eg, when associating with a cell membrane.10 These subunits undergo a conformational change that significantly modifies CRP structure, solubility, and antigenicity. This form of CRP is called modified or monomeric CRP (mCRP). mCRP is a naturally occurring stable protein but, in contrast to nCRP, is not found in serum but in fibrous tissues of normal human blood vessel intima.11 The formation of mCRP from CRP is nonproteolytic and irreversible.12 Numerous studies have identified CRP mRNA expression in many different extrahepatic cells such as islet cells of the pancreas,13 neurons,14 adipocytes,15 and renal tubular epithelial cells.16 Whether the "CRP" that is translated at each site is the nCRP conformer or the mCRP conformer needs to be clarified.
Recently, Paul et al17 showed that human CRP transgene expression caused accelerated atherosclerosis in apolipoprotein E–knockout (ApoE–/–) mice, a hypercholesterolemic mouse model of accelerated atherosclerosis. The aim of the present study was to compare the effects of human nCRP and mCRP on the development of early atherosclerosis by using directly injected proteins, thereby testing the hypothesis that the configuration of CRP may determine different effects on atherosclerosis. Using immunohistochemistry, we studied proinflammatory mediators thought to initiate atherosclerosis such as vascular cell adhesion molecule (VCAM), intracellular adhesion molecule (ICAM), and the CD40 ligand CD154, a member of the tumor necrosis factor gene superfamily. Furthermore, to elucidate the role of CRP within the plaque, immunostainings for mCRP, nCRP, ApoB, and macrophages were performed. Finally, cytokines in the serum that activate (interferon [IFN]-) and inhibit (interleukin [IL]-10) macrophages were measured.
Methods
Animal Model
Female homozygous ApoE–/– mice and their genetic background mice (C57BL/6) were obtained from Jackson Laboratories (Bar Harbor, Maine) at 6 weeks of age. They were maintained in a light- and temperature-controlled environment and had free access to water and standard laboratory chow containing 4% fat (Altromin). Weight-matched ApoE–/– and C57BL/6 mice (n=7) were randomly assigned to 3 treatment groups: Group 1 received human nCRP 2.5 mg/kg SC; group 2 received a recombinant form of human mCRP 2.5 mg/kg SC; and group 3 received 0.9% saline as control (identical volume given subcutaneously). Treatment was applied once weekly for 8 weeks. Animal weight was determined every week. Before the fifth injection, 50 μL whole blood was collected from the tail vein for white blood cell count.
At 14 weeks of age, mice were anesthetized with ketamine (100 mg/kg) and xylazine-HCl (4 mg/kg), and blood samples (between 200 and 250 μL) were collected from the vena cava for lipid analysis. Heart and aorta were carefully removed and fixed in 10% buffered formalin for morphometric analysis and immunohistochemistry. All experiments and measurements were performed in accordance with national guidelines for the care and use of research animals.
CRP Preparations
nCRP
Crude CRP was purchased from SeraCare Inc. It was bound to Q-Sepharose FF and was eluted with a step gradient of NaCl. Eluted protein was pooled, and CaCl2 was immediately added to a concentration of 2 mmol/L. The ion exchange step with CRP preparations is critically important not only to reduce endotoxin but also to remove any mCRP protein that may be present in an nCRP preparation. mCRP will strongly bind to the agarose resin. It resists elution with increasing salt concentration but elutes with 5 mol/L guanidinium hydrochloride. Eluted nCRP protein was dialyzed against 25 mmol/L-Tris-HCl and 0.15 mol/L NaCl (pH 7.4) containing 2 mmol/L CaCl2, sterile filtered, and stored at 4°C. Storage in calcium-containing buffers is critical to prevent the spontaneous formation of mCRP from nCRP.
mCRP
mCRP was prepared from nCRP by urea chelation as described previously.18 By electron microscopic analysis, mCRP molecules in the absence of added salt associate into a diffuse matrix distinct from the annular pentameric disk that defines nCRP. In a normal saline solution, mCRP forms an insoluble suspension of self-aggregated subunits. A recombinant form of mCRP (rmCRP) with both cysteine residues mutated to alanine residues and with an added N-terminal formylmethionine residue was expressed in Escherichia coli and isolated from inclusion bodies to >95% purity. To facilitate purification of rmCRP, it was acylated with citraconic anhydride to enhance solubility (C-rmCRP). C-rmCRP was directly comparable to mCRP as analyzed by SDS-PAGE, amino acid composition, and N-terminal sequence analysis. Once purified, the citraconyl groups were removed from C-rmCRP by dialysis at room temperature for 20 hours in 0.1 mol/L citrate (pH 3.5). The decitraconylated protein (Cx-rmCRP) was dialyzed extensively at 4°C against 25 mmol/L sodium PBS without calcium (pH 7.4), yielding an insoluble suspended protein similar to mCRP in saline-based buffers. This form was used in the present study. Monoclonal antibodies directed against the C-terminal octapeptide of the CRP subunit (mAb 9C9), which is expressed only in mCRP and not in nCRP, reacted with the same specificity and affinity with mCRP, C-rmCRP, and Cx-rmCRP.
Biochemistry
White blood cell counts were determined automatically with the Coulter method (Beckman) designed for small sample volumes. Serum cholesterol and triglycerides were measured enzymatically. For quantification of lipoprotein cholesterol, 30 μL serum was subjected to agarose gel electrophoresis, followed by enzymatic staining of cholesterol as previously described.19 IFN- and IL-10 levels in the serum of ApoE–/– mice were measured by ELISA in duplicates using commercially available kits (Mouse Quantikine ELISA kits, R&D Systems Inc).
Morphological Evaluation
Morphometric analysis of the aorta was performed by planimetry and a semiautomatic image analyzing system (AnalysisPRO, SIS), combined with a counting grid eyepiece for defined movement over the section plane (Zeiss Co) at a magnification of x400 as described previously.20 Contours of aortic plaques in the aortic root through the aortic valve and in the aortic arch were marked manually with a cursor, and plaque area was automatically calculated. All investigations were performed in a blinded manner.
Immunohistochemistry
Immunohistochemistry was performed on paraffin sections with the following antibodies: rabbit anti–human-CD154 (Santa Cruz Biotechnology, 1:600), rabbit anti–mouse-ApoB48/100 (USBiological, 1:500), rat anti–mouse-macrophages/monocytes (MAB1852, Chemicon International Inc, 1:25), goat anti–human-VCAM-1 (Santa Cruz Biotechnology, 1:300), goat anti–mouse-ICAM-1 (Santa Cruz Biotechnology, 1:100), and 2 mouse anti–human-CRP antibodies as previously described,21 which were provided by L.A. Potempa: mAb 9C9, which reacts exclusively with mCRP, and mAb 8D8, showing only reactivity with nCRP. Briefly, freshly cut 4-μm sections were mounted on silane-coated slides, deparaffinized, and rehydrated through a graded ethanol series, followed by blocking of nonspecific binding with 3% H2O2 (20 minutes, room temperature). An antigen retrieval was performed by heating the slides in citrate buffer (20 minutes in a microwave at 900 W). For antigen visualization, the CSA Dako and Goat Vectastain Elite ABC kits with AEC as the substrate (CD154, VCAM, and ICAM stainings, red deposits), the Rat and Rabbit Vectastain Elite ABC, and the MOM Vectastain kits (Vector Labs) with DAB and histogreen as substrates (CRP and macrophage stainings, brown deposits; ApoB staining, green deposits) were used. Color development was stopped by adding water, and sections were finally counterstained with hematoxylin. Negative controls were performed by omitting the primary antibody. Specific blocking experiments were carried out by preincubating mAb 9C9 and 8D8 with a 10-fold excess of the respective antigens. A double staining was performed with mAb 9C9 as the first and anti-ApoB as the second antibody. The sections were examined under light microscopy at a magnification of x400 with a semiquantitative scoring system (0 to 4): 0=no expression, 1=mild expression, 2=moderate expression, 3=strong expression, and 4=very strong expression. At least 5 animals per group were stained. All analyses were performed in a blinded manner.
Statistical Analysis
Data are presented as mean±SEM. For comparisons between treatment groups, the Kruskal-Wallis H test for nonnormally distributed variables was performed. Bivariate regression and correlation analyses were performed by use of the Spearman rank method. A value of P<0.05 was considered statistically significant. Data analysis was performed with the SPSS/PC+ package (SPSS Inc).
Results
Animal Data
Initial body weights were not different between groups. At the end of the study, no differences in body weights were detectable among the ApoE–/– groups (17.1±1.6, 17.3±1.6, and 17.3±1.6 g for saline, nCRP, and mCRP, respectively), whereas mCRP-treated C57BL/6 control mice (18.8±1.8 g) were slightly heavier compared with the other 2 groups (saline and nCRP, both 17.9±1.7 g; P<0.05). Other 3-month-old ApoE–/– mice in the animal facility receiving a normal chow had similar body weights. Animals treated with either form of CRP showed no difference in activity or behavior compared with their controls.
Biochemistry
White blood cell counts were identical in all ApoE–/– and C57BL/6 groups (data not shown), indicating that no treatment induced a systemically detectable inflammatory response. In all ApoE–/– groups, IFN- levels in the serum were below the detection limit of 4 pg/mL. In contrast, all ApoE–/– mice had detectable serum IL-10 levels >4 pg/mL (14±3, 10±2, and 26±11 pg/mL for saline, nCRP, and mCRP, respectively; range, 5 to 92 pg/mL). All control C57BL/6 mice had IL-10 levels >20 pg/mL (35±4 pg/mL; range, 21 to 49 pg/mL; P<0.01 between control C57BL/6 and control ApoE–/– mice). In Table 1, results are presented according to low (<10 pg/mL), middle (10 and <20 pg/mL), and high ( 20 pg/mL) IL-10 levels. Significant differences were observed between groups of ApoE–/– mice (P<0.05). Although control ApoE–/– mice had equally distributed IL-10 levels, all mCRP-treated ApoE–/– mice exhibited IL-10 levels in the middle and upper ranges, whereas IL-10 levels of all nCRP-treated ApoE–/– mice were <20 pg/mL. Total cholesterol and triglycerides levels, as well as the lipoprotein profiles, were unaffected by either nCRP or mCRP treatment in various groups of ApoE–/– mice (Table 2).
Histology and Morphology
Aortic Root
In the aortic root, all ApoE–/– mice developed early atherosclerotic lesions characterized by the predominance of foam cells accumulating in the subendothelial space, with no obvious changes in the underlying intima (Figure 1a through 1c). However, treatment with nCRP resulted in 4-fold-higher mean aortic plaque area in the aortic root in ApoE–/– mice compared with controls (38.6±18.3 versus 9.7±4.0x103 μm2), and 2 of 7 animals had much larger aortic plaques of 62.5 and 136.9x103 μm2. In contrast, mean plaque area was decreased by 50% in ApoE–/– mice treated with mCRP compared with controls (4.2±1.3 versus 9.7±4.0x103 μm2). Differences between the 3 groups were statistically significant (P<0.05; Figure 2).
Aortic Arch
The same trend within the groups was observed when plaque area in the aortic arch was quantified, although statistical significance was not reached (4.0±1.4 and 1.5±0.9 versus 2.8±0.8x103 μm2 for nCRP, mCRP, and saline, respectively; 0.2>P>0.05). No lesions were detectable in the control C57BL/6 mice.
Immunohistochemistry
Discussion
In this study, we show that 2 clearly defined different conformers of human CRP cause opposite effects in an animal model of early atherosclerosis. nCRP increased and mCRP reduced plaque formation in ApoE–/– mice with early lesions, processes that may involve the production of the antiinflammatory cytokine IL-10, the proinflammatory cytokine CD154, and the expression of VCAM and ICAM.
The role that CRP plays in vivo with respect to acceleration or inhibition of atherogenesis is an important issue. In recent years, a multitude of in vitro data have been generated showing proatherogenic effects of CRP (for review, see elsewhere3,4). We propose that it is essential to address several methodical aspects in interpreting these data. First, previous studies have neither discussed nor distinguished the observed effects in terms of 2 distinct conformational forms of CRP. Because prolonged storage of purified CRP in the absence of calcium or in the presence of chelating agents will cause a spontaneous conversion of nCRP to mCRP, this issue is most relevant.22 Because mCRP does not require fragmentation of the nCRP protein, an SDS-PAGE assay to define the "purity" of a nCRP preparation would be insufficient to prove that there is no mCRP in such preparations. Second, many, if not all, CRP-polyclonal antisera contain specificity to both the nCRP antigen and mCRP antigen.23 Also, monoclonal anti-CRP– clone 8 has clearly been shown to detect predominantly the mCRP protein and not the nCRP protein.24 Hence, any immunohistological studies evaluating tissue-associated CRP may, in fact, more accurately report on the expression of tissue associated mCRP. Third, many previously reported studies using CRP used test protein concentrations >5 μg/mL. Using high concentrations of CRP protein would favor the effects of any minor component found in any isolated preparation. In the present study, we tried to take these limiting factors into consideration.
Three recently published in vivo studies support an active role of human CRP in vascular injury in mice. First, arterial injury in human CRP-transgenic mice resulted in a higher rate of thrombosis.25 Second, human CRP-transgenic mice crossed with ApoE–/– mice developed accelerated atherosclerosis.17 Third, in rats, injection of human CRP enhanced tissue damage in myocardial infarction.26 The last 2 studies described immunohistochemical expression of human CRP in the tissue. Our finding that nCRP accelerated plaque formation in ApoE–/– mice is in line with the results of the above-mentioned models. However, our model differs in several important ways. Concern has been addressed through the use of heterologous systems.27 Exposing mice to huge amounts of human CRP might indeed not be physiological because mouse CRP is only a trace protein, not an acute-phase reactant. Taking this into consideration, we designed a model of low dosing over a long period of time. Our results most likely do not reflect systemic CRP levels and the direct effect of circulating CRP on atherogenic plaque formation. Kresl et al28 reported that when mice were injected with 100 μg IV (5 mg/kg) of mCRP, little or no mCRP was detected in a number of serum specimens. Because mCRP is hydrophobic and thus tends to self-aggregate, its movement away from the injection site as a free protein is unlikely.
The mCRP antigenic reactivity found in the plaques of all mouse aortas could reflect the natural presence of mouse mCRP. We did not directly determine to what extent the mAb 9C9 antibody also recognizes mouse "CRP" epitopes. But because the primary structure of mouse CRP is known to be 70% homologous to human CRP29 and residues 201 to 205 (KPQLW), which define the epitope for mAb 9C9 specificity,21 are conserved in human and mouse CRP,9 it is highly likely that a naturally occurring form of mouse mCRP exists. The facts that saline-treated ApoE–/– mice also stained positively and that antibody reactivity could be completely adsorbed with the human mCRP protein also suggest that human and mouse proteins are structurally similar. Our animal model identifies, for the first time, the existence of a mouse mCRP protein that is found at vessel walls and sites of atherogenic plaque buildup. Furthermore, plaque development is regulated by injection of human CRP proteins at distant tissue sites.
Staining of mCRP could be attributed to the cytoplasma of foam cells. In an in vitro study, it was proposed that CRP aggregates may bind and cluster LDL particles phagocytized by macrophages, leading to foam cell formation.30 The colocalization of mCRP, ApoB, and macrophages found here supports this hypothesis. Comparison of nCRP and mCRP staining in the plaque showing a predominance of mCRP reactivity within the foam cells suggests that CRP aggregates are related to the modified isoform of CRP. Extrahepatic expression of CRP in smooth muscle cells has been described.31,32 However, whether mCRP is locally produced or derived from serum CRP needs to be clarified. The lack of mCRP staining in normal vessel sections without plaque formation is in contrast to previous findings in humans in whom it is constitutively expressed in several healthy tissues.11 Further studies are needed to determine the exact role of mouse mCRP at the site of plaque development. However, the presence of mCRP in the vessel wall as soon as plaques develop suggests an important role in early atherogenesis in mice.
Extending previous work by Paul et al,17 who showed proatherogenic effects of CRP in ApoE–/– mice with well-established lesions, our animal model starting with CRP injections as early as 6 weeks, before lesions were established, suggests effects on early atherogenesis. To provide further evidence for this assumption, we studied proinflammatory pathways involved in the initiation of the disease, ie, expression of VCAM and ICAM, which regulate leukocyte adhesion to the endothelium,33 and the expression of the CD40 ligand CD154, which is described in endothelial and smooth muscle cells, macrophages, platelets, and T lymphocytes and thought to contribute to the initial recruitment of inflammatory cells to damaged endothelium (for review, see work by Schonbeck and Libby34). It was shown that the disruption of the CD154-CD40 system retarded the initiation of arterial plaque formation.35 We observed a trend toward lower CD154 expression in endothelium and smooth muscle cells in mCRP-treated ApoE–/– mice compared with the other 2 groups. In addition, CD154 expression correlated with plaque size. Treatment with human mCRP at a distant injection site might have reduced the tumor necrosis factor-–related inflammatory process at the aortic site. The observation that enhanced plaque formation induced by nCRP was paralleled by increased VCAM and a trend toward increased ICAM expression supports the critical role of ICAM and VCAM found by others in hypercholesterolemic animal models.36,37
At first glance, the results presented here appear to be contradictory to reports in vitro. In those studies, mCRP triggered proinflammatory activities in isolated human neutrophils and endothelial cells, including increased production of monocyte chemoattractant protein-1, IL-8, ICAM, VCAM, and E-selectin.18,22,38 In contrast, nCRP was without detectable effects on these signaling cascades. It was suggested that endothelial injury might expose the naturally occurring mCRP, leading to the attraction of neutrophils to the area of injury. Alternatively, mCRP may also be formed from nCRP at sites of injury or infection by some as-yet-unknown mechanism as part of the acute activation and/or resolution of the inflammatory process.39
The fact that human mCRP-treated mice showed less plaque is not necessarily in contradiction to the aforementioned studies that addressed immediate responses (over minutes to hours) using isolated cells. The long-term effect of a stimulatory protein on the overall immune response in a live animal cannot be predicted precisely from isolated systems. Such opposite stimulatory and inhibitory actions, depending on the dose used, have been described for anti-CD3 monoclonal antibody effects on the T-cell receptor in vivo.40 In an animal model of systemic lupus erythematodes, a single injection of 200 μg SC of CRP prevented or reversed lupus nephritis.41 Because IL-10 was required for this protection, we studied serum levels of this antiinflammatory cytokine with potent deactivating properties on both macrophages and T lymphocytes.42 Although ApoE–/– mice generally had lower IL-10 levels than the C57BL/6 control mice, mCRP-treated ApoE–/– mice had higher IL-10 levels than the other 2 groups. Therefore, mCRP injections could be beneficial through upregulation of IL-10. Namiki et al43 proposed IL-10 as a new therapeutic tool for the treatment of atherosclerosis. Gene transfer of IL-10 cDNA reduced atherosclerosis in male ApoE–/– mice and decreased serum levels of IFN- associated with macrophage activation. Seven days after the last injection, we were unable to detect IFN- in either ApoE–/– or C57BL/6 mice. Because of the limited availability of serum material, it was not possible to measure other proinflammatory or antiinflammatory cytokines.
In summary, we describe a model of early atherogenesis in which human nCRP accelerated atherosclerosis and mCRP partly prevented plaque formation. Results of IL-10 measurements are consistent with the interpretation that small amounts of mCRP could heighten normal immune surveillance in the mouse, slowing the process of atherosclerotic plaque formation. Immunostainings for CD154, VCAM, and ICAM may indicate that nCRP and mCRP influence the initial steps of atherogenesis differentially. Further studies are needed to elucidate the signaling cascade of mCRP and its relevance to the processes of atherosclerosis in humans.
Acknowledgments
For this project, Dr Schwedler received a 2002 scholarship from the Deutsche Nierenstiftung. She is currently funded by a research program of the University Würzburg (HWP, Programm "Chancengleichheit für Frauen in Forschung und Lehre"). This work was also supported by the Deutsche Forschungsgemeinschaft (grants Ga431/5-3 and SFB 355, TP B11), and the IZKF Erlangen (Interdisziplin;res Zentrum für Klinische Forschung, project A11). We thank Marita Bartrow, Carmen Bauer, Monika Klewer, and Stefan S;llner for technical assistance.
Disclosure
Dr Potempa was employed by NextEra Therapeutics, Inc, and is currently employed by Immtech International, Inc. NextEra and Immtech are biotechnology companies that (in aggregate) are the primary owners of intellectual property developed on the modified form of C-reactive protein.
References
Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation. 2003; 107: 363–369.
Yeh ET, Willerson JT. Coming of age of C-reactive protein: using inflammation markers in cardiology. Circulation. 2003; 107: 370–371.
Jialal I, Devaraj S, Venugopal SK. C-reactive protein: risk marker or mediator in atherothrombosis; Hypertension. 2004; 44: 6–11.
Mazer SP, Rabbani LE. Evidence for C-reactive protein’s role in (CRP) vascular disease: atherothrombosis, immuno-regulation and CRP. J Thromb Thrombolysis. 2004; 17: 95–105.
Wolbink GJ, Brouwer MC, Buysmann S, ten Berge IJ, Hack CE. CRP-mediated activation of complement in vivo: assessment by measuring circulating complement-C-reactive protein complexes. J Immunol. 1996; 157: 473–479.
Verma S, Wang CH, Li SH, Dumont AS, Fedak PW, Badiwala MV, Dhillon B, Weisel RD, Li RK, Mickle DA, Stewart DJ. A self-fulfilling prophecy: C-reactive protein attenuates nitric oxide production and inhibits angiogenesis. Circulation. 2002; 106: 913–919.
Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation. 2000; 102: 2165–2168.
Sternik L, Samee S, Schaff HV, Zehr KJ, Lerman LO, Holmes DR, Herrmann J, Lerman A. C-reactive protein relaxes human vessels in vitro. Arterioscler Thromb Vasc Biol. 2002; 22: 1865–1868.
Shrive AK, Cheetham GM, Holden D, Myles DA, Turnell WG, Volanakis JE, Pepys MB, Bloomer AC, Greenhough TJ. Three dimensional structure of human C-reactive protein. Nat Struct Biol. 1996; 3: 346–354.
Wu Y, Wang HW, Ji SR, Sui SF. Two-dimensional crystallization of rabbit C-reactive protein monomeric subunits. Acta Crystallogr D Biol Crystallogr. 2003; 59: 922–926.
Diehl EE, Haines GK, III, Radosevich JA, Potempa LA. Immunohistochemical localization of modified C-reactive protein antigen in normal vascular tissue. Am J Med Sci. 2000; 319: 79–83.
Kresl JJ, Potempa LA, Anderson BE. Conversion of native oligomeric to a modified monomeric form of human C-reactive protein. Int J Biochem Cell Biol. 1998; 30: 1415–1426.
Fehsel K, Plewe D, Kolb-Bachofen V. Nitric oxide-induced expression of C-reactive protein in islet cells as a very early marker for islet stress in the rat pancreas. Nitric Oxide. 1997; 1: 254–262.
Yasojima K, Schwab C, McGeer EG, McGeer PL. Human neurons generate C-reactive protein and amyloid P: upregulation in Alzheimer’s disease. Brain Res. 2000; 887: 80–89.
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.
Jabs WJ, Logering BA, Gerke P, Kreft B, Wolber EM, Klinger MH, Fricke L, Steinhoff J. The kidney as a second site of human C-reactive protein formation in vivo. Eur J Immunol. 2003; 33: 152–161.
Paul A, Ko KW, Li L, Yechoor V, McCrory MA, Szalai AJ, Chan L. C-reactive protein accelerates the progression of atherosclerosis in apolipoprotein E-deficient mice. Circulation. 2004; 109: 647–655.
Khreiss T, Jozsef L, Hossain S, Chan JS, 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.
Nauck M, Winkler K, M;rz W, Wieland H. Quantitative determination of high-, low-, and very-low-density lipoproteins and lipoprotein(a) by agarose gel electrophoresis and enzymatic cholesterol staining. Clin Chem. 1995; 41: 1761–1767.
Nabokov AV, Amann K, Wessels S, Munter K, Wagner J, Ritz E. Endothelin receptor antagonists influence cardiovascular morphology in uremic rats. Kidney Int. 1999; 55: 512–519.
Ying SC, Gewurz H, Kinoshita CM, Potempa LA, Siegel JN. 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.
Zouki C, Haas B, Chan JS, 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.
Samberg NL, Bray RA, Gewurz H, Landay AL, Potempa LA. Preferential expression of neo-CRP epitopes on the surface of human peripheral blood lymphocytes. Cell Immunol. 1988; 116: 86–98.
Schwedler SB, Guderian F, D;mmrich J, Potempa LA, Wanner C. Tubular staining of modified C-reactive protein in diabetic chronic kidney disease. Nephrol Dial Transplant. 2003; 18: 2300–2307.
Danenberg HD, Szalai AJ, Swaminathan RV, Peng L, Chen Z, Seifert P, Fay WP, Simon DI, Edelman ER. Increased thrombosis after arterial injury in human C-reactive protein–transgenic mice. Circulation. 2003; 108: 512–515.
Griselli M, Herbert J, Hutchinson WL, Taylor KM, Sohail M, Krausz T, Pepys MB. C-reactive protein and complement are important mediators of tissue damage in acute myocardial infarction. J Exp Med. 1999; 190: 1733–1740.
Pepys MB, Hirschfield GM. C-reactive protein: a critical update. J Clin Invest. 2003; 111: 1805–1812.
Kresl JJ, Potempa LA, Anderson B, Radosevich JA. Inhibition of mouse mammary adenocarcinoma (EMT6) growth and metastases in mice by a modified form of C-reactive protein. Tumour Biol. 1999; 20: 72–87.
Pepys MB, Dash AC, Fletcher TC, Richardson N, Munn EA, Feinstein A. Analogues in other mammals and in fish of human plasma proteins, C-reactive protein and amyloid P component. Nature. 1978; 273: 168–170.
Fu T, Borensztajn J. Macrophage uptake of low-density lipoprotein bound to aggregated C-reactive protein: possible mechanism of foam-cell formation in atherosclerotic lesions. Biochem J. 2002; 366: 195–201.
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.
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 artery venous bypass grafts and normal vascular tissue. Circulation. 2003; 108: 1428–1431.
Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Schonbeck U, Libby P. The CD40/CD154 receptor/ligand dyad. Cell Mol Life Sci. 2001; 58: 4–43.
Schonbeck U, Sukhova GK, Shimizu K, Mach F, Libby P. Inhibition of CD40 signaling limits evolution of established atherosclerosis in mice. Proc Natl Acad Sci U S A. 2000; 97: 7458–7463.
Collins RG, Velji R, Guevara NV, Hicks MJ, Chan L, Beaudet AL. P-Selectin or intercellular adhesion molecule (ICAM)-1 deficiency substantially protects against atherosclerosis in apolipoprotein E-deficient mice. J Exp Med. 2000; 191: 189–194.
Cybulsky MI, Iiyama K, Li H, Zhu S, Chen M, Iiyama M, Davis V, Gutierrez-Ramos JC, Connelly PW, Milstone DS. A major role for VCAM-1, but not ICAM-1, in early atherosclerosis. J Clin Invest. 2001; 107: 1255–1262.
Khreiss T, Jozsef 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 ET. C-reactive protein: structure affects function. Circulation. 2004; 109: 1914–1917.
Nakane A, Okamoto M, Asano M, Kohanawa M, Minagawa T. An anti-CD3 monoclonal antibody protects mice against a lethal infection with Listeria monocytogenes through induction of endogenous cytokines. Infect Immun. 1993; 61: 2786–2792.
Rodriguez W, Mold C, Kataranovski M, Hutt J, Marnell LL, Du Clos TW. Reversal of ongoing proteinuria in autoimmune mice by treatment with C-reactive protein. Arthritis Rheum. 2005; 52: 642–650.
de Vries JE. Immunosuppressive and anti-inflammatory properties of interleukin 10. Ann Med. 1995; 27: 537–541.
Namiki M, Kawashima S, Yamashita T, Ozaki M, Sakoda T, Inoue N, Hirata K, Morishita R, Kaneda Y, Yokoyama M. Intramuscular gene transfer of interleukin-10 cDNA reduces atherosclerosis in apolipoprotein E-knockout mice. Atherosclerosis. 2004; 172: 21–29.(Susanne B. Schwedler, MD;)