| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Department of Pathology,* Cardiovascular Disease Laboratory, Institute of Basic Medical Sciences, University of Tsukuba, Tsukuba, Japan; the First Department of Pathology, Faculty of Medicine, University of Miyazaki, Miyazaki, Japan; Institute for Experimental Animals, Kobe University School of Medicine, Kobe, Japan; the Department of Cardiology,¶ Fukuoka University School of Medicine, Fukuoka, Japan; Analytical Research Center for Experimental Sciences|| and President Office,** Saga University, Saga, Japan; Cardiovascular Research Institute, Morehouse School of Medicine, Atlanta, Georgia; and the Department of Pharmacology, Dalian Medical University, Dalian, China
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, we used both rabbit and human specimens to examine CRP deposition and the relationship of CRP with other cellular components in atherosclerotic lesions. For rabbit experiments, we used two kinds of rabbits with atherosclerosis: 45 cholesterol-diet fed rabbits32 (fed a diet containing 0.3% cholesterol and 3% soybean oil for 16 weeks) and 31 WHHL rabbits33 (fed a chow diet, aged from 8 to 12 months) and 62 normolipidemic rabbits (fed a chow diet, aged from 4 to 10 months) as a control group. All rabbits were raised in a specific pathogen free-equivalent barrier facility at either the University of Tsukuba or Saga University and given food and water ad libitum. Human coronary arteries and aortas were obtained from patients who died of myocardial infarction within 6 hours after death. Autopsy was performed at the Department of Pathology, Faculty of Medicine, University of Miyazaki. All specimens were fixed in formalin for histological examinations or snap-frozen in liquid nitrogen and stored at –80°C for protein isolation and RNA extraction. All protocols for animal studies and use of human specimens were performed according to the guidelines and with the approval of the ethics review committee of the University of Tsukuba.
Histology and Immunohistochemical Staining
All specimens were embedded in paraffin and cut into 3-µm-thick serial sections. The specimens were mounted on slides and deparaffinized with graded concentrations of xylene and ethanol, and then stained with hematoxylin and eosin (H&E). For immunohistochemical staining, all slides were immersed in 3% H2O2 in methanol for 45 minutes at room temperature to block endogenous peroxidase activity. To stain human macrophages and complement 3 (C3), the slides were predigested with 0.4% proteinase K for 5 minutes at 37°C. The sections were then incubated with 10% normal goat serum for 60 minutes to reduce nonspecific background staining and reacted with each appropriately diluted antibody (Ab) at 4°C overnight. The panel of Abs used for examining rabbit and human CRP, apoB, C3, and macrophages and SMCs in the lesions is listed in Table 1
. CRP antibodies were solid phase absorbed and then immunoaffinity purified of CRP proteins. Affinity purified CRP antibodies give a single band by immunoelectrophoresis and gel diffusion against fluids containing CRP. The specificity of CRP antibodies has been confirmed by Western blots. This antibody does not cross-react with serum amyloid P component, haptoglobin, 
-1-acid glycoprotein and IgG. It can detect both native and denatured-reduced CRP proteins. After rinsing with phosphate-buffered saline (pH 7.4), the sections were incubated with peroxidase-conjugated secondary Abs at room temperature for 60 minutes and then the staining was visualized by reaction with 3-amino-9-ethylcarbazole (Nichirei, Tokyo) as described previously.34
For negative controls, primary Abs were either omitted or replaced by nonspecific IgG.
|
To examine the expression of CRP proteins in the atherosclerotic lesions, aortic specimens were divided into lesions and nonlesions based on the gross observation. The aortic lesions were further separated into the intima, media, and adventitia under a dissecting microscope. In addition, alveolar macrophages, thioglycollate-elicited peritoneal macrophages, and liver were used. One hundred mg of each tissue were homogenized in ice-cold suspension buffer (0.02 mol/L Tris-HCl, pH 7.5) supplemented with phenylmethyl sulfonyl fluoride (1 µg/ml), aprotinin (1 µg/ml), and ethylenediamine tetraacetic acid (1 mmol/L). Ten-µg aliquots of the crude proteins from each sample were separated by 4 to 20% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gradient gel electrophoresis for Western blotting and probed with a mouse monoclonal Ab against human CRP.35
Northern Blot and Real-Time Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)
Total RNA was isolated from various tissues of both normal and cholesterol-fed and WHHL rabbits using Trizol reagent (Invitrogen, Carlsbad, CA). Thioglycollate-elicited peritoneal and alveolar macrophages were collected as described.35
Human specimens (aorta and liver) were obtained at autopsy. Ten µg of RNA was then subjected to electrophoresis in a 1.2% agarose gel and transferred to a Nytran nylon membrane. The membrane was hybridized in turn with the 32P-labeled human CRP cDNA probe. The membrane was rehybridized with human ß-actin cDNA probe as an internal standard. Total RNA was reverse-transcribed into cDNA by using Invitrogen reverse transcription reagents. Expression of CRP and a housekeeping gene (GAPDH) in the liver, aortic lesions, nonlesion regions of aortas, and macrophages by real-time RT-PCR (DNA Engine Option; MJ Research, Tokyo, Japan) were performed using DyNAmo SYBR Green qPCR kits (Finnzymes) according to the manufacturers’ instructions.35
The primers for rabbit and human CRP and GAPDH are listed in Table 2
.
|
To determine whether CRP was elevated in cholesterol-fed or WHHL rabbits, levels of rabbit CRP in the plasma were quantified using two rabbit high-sensitive CRP ELISA kits from Kamiya Biomedical (Seattle, WA) and Shibayagi Co. (Gunma, Japan), and compared with normal rabbits. All measurements were performed independently by two researchers using different ELISA kits. The plasma specimens were mixed thoroughly before assays.
In Vitro Investigation of CRP Expression
To examine the effects of cytokines on the regulation of CRP expression in hepatocytes and possibly in macrophages, we used the Huh7 cell line (derived from human hepatoma) and U937 monocytic cells. Huh7 cells were cultured in serum-free Dulbecco’s modified Eagle’s medium and incubated with various concentrations of interleukin (IL)-1ß, IL-6, and tumor necrosis factor- for 48 hours. U937 cells were cultured in RPMI 1640 medium and treated with phorbol 12-myristate 13-acetate (100 ng/ml) for 3 days before incubation with cytokines.36
After that, these cells were collected for the extraction of total RNA using Trizol agent and CRP expression was evaluated using real-time RT-PCR as described above.
Statistical Analysis
All values are expressed as mean ± SE. Plasma CRP, total cholesterol, and aortic en face lesion area were compared with Student’s t-test, Welch’s t-test, or Mann-Whitney’s U-test for nonparametric analysis. Correlations were analyzed using Spearman’s correlation coefficient by rank test.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In normal rabbits, the range of plasma CRP levels was narrow with an average of 3.15 ± 0.8 mg/L. It seems that rabbit CRP levels are close to the mean values of healthy middle-aged humans (2.82 mg/L) reported in the literature.37
Rabbits with hypercholesterolemia, either induced by cholesterol diet feeding or caused by LDL receptor deficiency, had significantly higher plasma CRP levels than normal rabbits: 2.5-fold higher in cholesterol-fed rabbits and 28-fold higher in WHHL rabbits on average (Figure 1A)
. We also examined the relationship between plasma CRP and aortic atherosclerosis en face size or plasma total cholesterol levels in hyperlipidemic rabbits. As shown in Figure 1B and C
, the CRP level in plasma was well correlated with the lesion size in the aorta, and weakly associated with the plasma total cholesterol level.
|
Because a large range of lesions were obtained from a sufficient number of rabbits in this study, we were able to select both early stage lesions (referred to as foam cell-rich lesions) as well as relatively advanced lesions (including fibrous plaques and complicated lesions). Immunohistochemical staining revealed that in the early-stage lesions, where a few layers of macrophages were present, CRP-immunoreactive proteins were found either on the surface or beneath the foam cells (Figure 2A)
. In advanced lesions, CRP was primarily localized beneath the fibrous cap and closely associated with apoB deposition (Figure 2B)
. In either case, CRP did not apparently overlap with the cytoplasm of macrophages and foam cells (Figure 4)
. CRP tended to be localized around the extracellular matrix and was seldom, if ever, associated with macrophages or SMCs (also see supplemental data for high magnification at http://ajp.amjpathol.org).
|
|
. CRP was co-localized with apoB but seemed not to be associated with the cytoplasm of macrophages and foam cells. In these lesions, CRP was not deposited in any particular areas such as the cap or core. We also performed Western blot analysis. For this purpose, we separated the lesion from nonlesion areas (based on gross observation) and furthermore, we isolated the intima, media, and adventitia of the aorta from WHHL rabbits under a dissecting microscope. We were especially interested in examining whether macrophages could express CRP per se, and therefore we used macrophages in parallel. As shown in Figure 5
, CRP was detected in the lesional intima but not in the media or adventitia. Isolated macrophages contained smaller amounts of CRP but the immunoreactive bands from macrophages showed faster mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, presumably they were degraded CRP proteins or taken up by macrophages.
|
|
Next, we investigated CRP deposition in human coronary plaques. CRP was detected in all types of the lesions, intimal thickening, unstable plaques, and ruptured plaques, and especially in fragile or ruptured caps. In the vicinity of these lesions, strong CRP staining was seen around or concentrated on the surface, but did not overlap with the cytoplasm of macrophages (Figure 6)
. In these CRP-positive areas, C3 was also detected (Figure 6B)
, which was consistent with a previous report.38
In stable plaques, CRP was also detected, either on the surface or in the deep area of thickened intima (Figure 7)
. The immunohistochemical findings were further confirmed by Western blot analysis and lesions of aortic atherosclerosis contained substantial amounts of CRP proteins regardless of the lesion characteristics (Figure 8)
.
|
|
|
The fact that CRP-immunoreactive proteins were invariably present in atherosclerotic lesions raised the question of whether CRP was exclusively derived from the circulation or/and produced locally. To address this issue, we initially screened the CRP mRNA expression of multiple tissues from both chow- and cholesterol-fed rabbits using Northern blotting. As shown in Figure 9A
, the liver was the only organ that expressed detectable mRNA by Northern blot hybridization. There was no expression of CRP mRNA in either atherosclerotic lesions or isolated macrophages. This result suggests that the liver is the exclusive organ for CRP production, while the arterial wall and macrophages do not express CRP mRNA or the levels of CRP mRNA expressed by the arterial wall and macrophages are too low to be detected. To further explore this possibility, we applied a real-time RT-PCR analysis, which is considered to be more sensitive than Northern blotting analysis and more specific than traditional quantitative RT-PCR.39
In these experiments, we analyzed the CRP expression of the liver (n = 22 rabbits), lesions (n = 16 rabbits), nonlesion regions of the aorta (n = 12 rabbits), and macrophages (n = 28 rabbits). Using the liver expression of CRP as a standard, we calculated the CRP expression level using the 2–
CT method by Livak and Schmittgen.40
We found that isolated macrophages did not express CRP mRNA (90% of sample values were actually 0). The average relative CRP expression levels (arbitrary units) of lesions or nonlesions were less than 1/100th of the hepatic expression of CRP (Figure 9B)
, suggesting that lesional CRP expression was indeed extremely low and not comparable with the hepatic expression. This conclusion was further strengthened by examining human aortic atherosclerotic lesions. Analysis of eight freshly isolated human aortic lesions showed no mRNA expression (5 sample values were 0 and 3 were <0.02) by real-time RT-PCR (average value of livers of the same patients: 121 ± 48 arbitrary units). We also compared CRP mRNA expression using human Huh7 hepatoma cells and U937 macrophages. Huh7 hepatocytes expressed a much lower level of CRP under basal conditions (compared to normal human liver tissue) but the level was discernibly up-regulated by incubation with IL-1ß, IL-6, and tumor necrosis factor- (data not shown). U937 macrophages did not express CRP mRNA in either normal state or when stimulated by various cytokines at different concentrations (all sample values were 0). Therefore, these results demonstrate that macrophages from both rabbits and humans did not express CRP mRNA.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In accordance with previous reports,12-14,24 our study clearly demonstrated that CRP protein was invariably present in atherosclerotic lesions, regardless of lesion type (early versus advanced) or nature (foam cell-rich versus extracellular matrix-rich) and was closely co-localized with apoB, suggesting that CRP may interact with apoB-containing particles in the intima, thereby facilitating their binding to extracellular matrix, increasing retention time and oxidation in the intima, and enhancing foam cell formation.43,44 As reviewed in the Introduction, these premises or assumptions regarding the proatherogenic effects of CRP, however, should be considered cautiously, and need to be verified in the future. Despite this, CRP deposition rarely overlapped with the cytoplasm of macrophages and foam cells in our studies of both rabbit and human specimens. Our findings showed that although some macrophages could be positively stained by CRP antibodies, this was not a frequent and universal phenomenon and may occur only when a particular ligand is exposed on these cells. Three methods of analysis (immunohistochemical staining, Western blotting, and real-time RT-PCR) provided consistent observations, and therefore, these results rebut the contention that macrophages16,17,38 are the major sources of CRP in the arterial wall or that macrophages are responsible for the increased plasma CRP levels. This notion is also consistent with a recent report using CRP transgenic mice showing that no CRP mRNA was detected in apoE knockout mouse lesions.45 Our results are not in agreement with the report by Yasojima and colleagues,38 who showed that human atherosclerotic plaques (10 human postmortem samples) contained higher (7.2-fold) amounts of CRP mRNA than liver and proposed that macrophages and SMCs are the main sources of plasma CRP. The discrepancy between our study and previous studies regarding the macrophage-produced CRP in the lesions is clearly due to the differences in the samples and methods applied. In sharp contrast to the previous reports, we used a large number and variety of atherosclerotic samples from two kinds of rabbits along with human coronary arteries and aortas as well as macrophages and livers. More importantly, we analyzed CRP mRNA expression using the most sensitive method available, real-time RT-PCR, whereas the previous studies performed only conventional semiquantitative RT-PCR,38 which has been criticized as being unreliable.39 It should be emphasized that none of the previous studies that concluded that CRP was expressed by vascular macrophages or SMCs hitherto were performed using real-time RT-PCR. In addition, our refutation was also supported by studies on cultured human Hu7h hepatocytes and U937 macrophages. It is noteworthy that CRP-immunoreactive proteins were also detected in unstable plaques or ruptured plaques, particularly on the fibrotic cap. CRP proteins were concentrated on the surface of macrophages and foam cells in these areas and co-localized with C3. Whether CRP in this particular region is directly involved in or triggers the plaque rupture remains to be determined.
It should be pointed out that there are several differences between human CRP and rabbit CRP in terms of their structures and functional properties. For example, native rabbit CRP has higher affinity ability to apoB-containing lipoproteins and formed a complex whereas human CRP can only bind to apoB-containing lipoproteins when they are aggregated. Human CRP can activate human complement pathway whereas it is completely unknown whether rabbit CRP can activate rabbit complement pathway (M. Pepys, personal communication). It remains unclear whether these differences (rabbit versus human, bound CRP versus free CRP, native versus aggregated CRP) may affect their physiological functions such as in atherosclerosis. Perhaps a precise elucidation of CRP pathological roles in atherogenesis awaits the generation of appropriate transgenic animals that can express high levels of inducible CRP in plasma or high levels of CRP specifically in lesions in the future. Recently, four studies using transgenic mice expressing either human45-47 or rabbit48 (K. Reifenberg, personal communication) CRP did not lead to a consistent conclusion regarding the atherogenic effect of CRP. One study showed that increased CRP moderately accelerated atherosclerosis in male apoE knockout mice45 whereas other studies failed to demonstrate this effect. In mice, endogenous plasma CRP is not inducible and never increases more than 2 to 3 mg/L49 and also apoE knockout mice have some defects in their complement system that cannot be activated by transgenic CRP.48 In this regard, rabbits are apparently superior to mice because rabbit CRP levels are highly inducible and responsive during the inflammatory reaction.31 Our separate study also showed that, as in human studies,6,50 administration of statin in rabbits could effectively normalize plasma CRP levels and reduce the CRP levels in both liver and aortic lesions (J.F., unpublished data). This result suggests that the rabbit may be an alternative animal model to evaluate therapeutic effects of decreasing the CRP level.51 Further studies will be required to clarify whether decreasing CRP alone without changing the plasma cholesterol level can be beneficial for the treatment of atherosclerosis. Recently, our laboratory has successfully created a line of human CRP transgenic rabbits (S.K. and J.F., unpublished data). In view of the multiple and complicated biological functions of CRP in many cardiovascular diseases and other inflammatory processes, CRP transgenic rabbits may be extremely valuable for elucidating both the pathophysiological functional roles and mechanisms of this protein in atherosclerosis, and for testing the efficacy of CRP-lowering drugs in the future.
In conclusion, our studies reported here demonstrate that there is a close relationship between the CRP plasma level and atherosclerosis in hypercholesterolemic rabbits. CRP immunoproteins were frequently present in the lesions of atherosclerosis regardless of the severity and degree of the lesions but this CRP was basically derived from the circulation rather than synthesized by vascular cells such as macrophages. In human coronary plaques, CRP was also frequently present in the ruptured fibrotic caps. In future studies, we are expecting to investigate whether elevated CRP levels are directly atherogenic or whether CRP is involved in the progression of atherosclerosis in vivo.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by grants-in-aid for scientific research from Ministry of Education, Culture, Sports and Technology, Japan (KAKENHI.16390089 and 16659099 to J.F.), the Takeda Science Foundation, the Ono Medical Research Foundation, and the Novartis Foundation for the Promotion of Science (to J.F.); and a grant from the Center for Tsukuba Advanced Research Alliance at the University of Tsukuba (to J.F.).
H.S. and T.K. contributed equally to this work.
Accepted for publication June 27, 2005.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  P. R. J. Ames, I. Antinolfi, A. Ciampa, J. Batuca, G. Scenna, L. R. Lopez, J. Delgado Alves, L. Iannaccone, and E. Matsuura Primary antiphospholipid syndrome: a low-grade auto-inflammatory disease? Rheumatology, December 1, 2008; 47(12): 1832 - 1837. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Montecucco, S. Steffens, F. Burger, G. Pelli, C. Monaco, and F. Mach C-reactive protein (CRP) induces chemokine secretion via CD11b/ICAM-1 interaction in human adherent monocytes J. Leukoc. Biol., October 1, 2008; 84(4): 1109 - 1119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Singh, M. V. Suresh, D. C. Prayther, J. P. Moorman, A. E. Rusinol, and A. Agrawal C-Reactive Protein-Bound Enzymatically Modified Low-Density Lipoprotein Does Not Transform Macrophages into Foam Cells J. Immunol., March 15, 2008; 180(6): 4316 - 4322. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Van Vre, H. Bult, V. Y. Hoymans, V. F.I. Van Tendeloo, C. J. Vrints, and J. M. Bosmans Human C-Reactive Protein Activates Monocyte-Derived Dendritic Cells and Induces Dendritic Cell-Mediated T-Cell Activation Arterioscler. Thromb. Vasc. Biol., March 1, 2008; 28(3): 511 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Singh, M. Hoffmann, R. Wolk, A. S.M. Shamsuzzaman, and V. K. Somers Leptin Induces C-Reactive Protein Expression in Vascular Endothelial Cells Arterioscler. Thromb. Vasc. Biol., September 1, 2007; 27(9): e302 - e307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Ryu, C. W. Lee, J.-A. Shin, C.-S. Park, J. J. Kim, S.-J. Park, and K. H. Han Fc{gamma}RIIa mediates C-reactive protein-induced inflammatory responses of human vascular smooth muscle cells by activating NADPH oxidase 4 Cardiovasc Res, August 1, 2007; 75(3): 555 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Liu, M. Morimoto, S. Kitajima, T. Koike, Y. Yu, H. Shiiki, M. Nagata, T. Watanabe, and J. Fan Increased Expression of Vascular Endothelial Growth Factor in Kidney Leads to Progressive Impairment of Glomerular Functions J. Am. Soc. Nephrol., July 1, 2007; 18(7): 2094 - 2104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Norja, L. Nuutila, P. J Karhunen, and S. Goebeler C-reactive protein in vulnerable coronary plaques J. Clin. Pathol., May 1, 2007; 60(5): 545 - 548. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Rocker, D. E. Manolov, E. V. Kuzmenkina, K. Tron, H. Slatosch, J. Torzewski, and G. U. Nienhaus Affinity of C-Reactive Protein toward Fc{gamma}RI Is Strongly Enhanced by the {gamma}-Chain Am. J. Pathol., February 1, 2007; 170(2): 755 - 763. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Fujii, S.-H. Li, P. E. Szmitko, P. W.M. Fedak, and S. Verma C-Reactive Protein Alters Antioxidant Defenses and Promotes Apoptosis in Endothelial Progenitor Cells Arterioscler. Thromb. Vasc. Biol., November 1, 2006; 26(11): 2476 - 2482. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. Sjoberg, L. A. Trouw, F. D. G. McGrath, C. E. Hack, and A. M. Blom Regulation of complement activation by C-reactive protein: targeting of the inhibitory activity of c4b-binding protein. J. Immunol., June 15, 2006; 176(12): 7612 - 7620. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Scirica, D. A. Morrow, S. Verma, S. Devaraj, I. Jialal, B. M. Scirica, D. A. Morrow, S. Verma, S. Devaraj, and I. Jialal The Verdict Is Still Out Circulation, May 2, 2006; 113(17): 2128 - 2151. [Full Text] [PDF]
|
|
|
|
|
|
J. Liang, E. Liu, Y. Yu, S. Kitajima, T. Koike, Y. Jin, M. Morimoto, K. Hatakeyama, Y. Asada, T. Watanabe, et al. Macrophage Metalloelastase Accelerates the Progression of Atherosclerosis in Transgenic Rabbits Circulation, April 25, 2006; 113(16): 1993 - 2001. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Jialal, S. Devaraj, U. Singh, J. Fan, and Y. E. Chen Sources of CRP in Atherosclerotic Lesions Am. J. Pathol., March 1, 2006; 168(3): 1054 - 1056. [Full Text] [PDF]
|
|
|
|
 |
|
 L. Rush NOTEWORTHY Vet. Pathol., January 1, 2006; 43(1): 84 - 85. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||
), SMCs, and apoB. A: Early-stage lesion, which is composed of a single layer of macrophage-derived foam cells. B: Advanced lesion, which is covered by a fibrotic cap. CRP is co-localized with apoB staining. See supplemental data for sections stained with CRP Abs shown in higher magnification at http://ajp.amjpathol.org.
1/100th of the liver) of CRP mRNA whereas macrophages did not produce detectable CRP mRNA.