| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2005;36:2571.)
© 2005 American Heart Association, Inc.
Original Contributions |
From the Department of Cardiac and Vascular Sciences (Q.X., K.M., M.M., Q.X.), St Georges University of London, United Kingdom; Department of Internal Medicine (G.S.), University of Vienna, Austria; Department of Internal Medicine (F.O.), Hospital of Brunneck, Italy; Division of Experimental Pathology and Immunology (G.W.), Medical University Innsbruck, Austria; and Department of Neurology (J.W., S.K.), Medical University Innsbruck, Austria
Correspondence to Qingbo Xu, MD, PhD, Department of Cardiological Sciences, St Georges University of London, Cranmer Terrace, London SW17 0RE, United Kingdom. E-mail q.xu{at}sghms.ac.uk
| Abstract |
|---|
|
|
|---|
Methods The association of sHSP60 with early atherogenesis (5-year development and progression of nonstenotic carotid plaques) was investigated as part of the population-based prospective Bruneck Study. The current study focused on the follow-up period between 1995 and 2000 and, thus, included 684 subjects.
Results sHSP60 levels measured in 1995 and 2000 were highly correlated (r=0.40; P<0.001), indicating consistency over a 5-year period. Circulating HSP60 levels were significantly correlated with antilipopolysaccharide and anti-HSP60 antibodies. It was also elevated in subjects with chronic infection (top quintile group of HSP60, among subjects with and without chronic infection: 23.8% versus 17.0%; P=0.003 after adjustment for age and sex). HSP60 levels were significantly associated with early atherogenesis, both in the entire population (multivariate odds ratio, for a comparison between quintile group V versus I+II: 2.0 [1.2 to 3.5] and the subgroup free of atherosclerosis at the 1995 baseline: 3.8 [1.6 to 8.9]). The risk of early atherogenesis was additionally amplified when high-sHSP60 and chronic infection were present together.
Conclusions Our study provides the first prospective data confirming an association between high levels of sHSP60 and early carotid atherosclerosis. This possibly indicates an involvement of sHSP60 in activating proinflammatory processes associated with early vessel pathology.
Key Words: heat shock protein atherosclerosis progression follow-up
| Introduction |
|---|
|
|
|---|
The HSP60/65 family of chaperone proteins is a group of evolutionarily conserved proteins with a molecular weight of
60 kDa, which facilitates the folding of newly synthesized polypeptides in an ATP-dependent manner and plays an important role in maintaining the dynamic stability of the intracellular proteome.68 Because of their evolutionary conservation, HSPs show a high degree of structural homology all the way from prokaryotes to humans.68 Human autoantibodies against HSP60/65 have been shown to be cross-reactive against chlamydial, mycobacterial, and other bacterial HSPs.9 HSP60, both human and chlamydial, has also been shown to enhance the production of proinflamatory cytokines, like tumor necrosis factor
and matrix metalloproteinase production in macrophages,10,11 and increase expression of interleukin 6, adhesion molecules like E-selectin, and intercellular adhesion molecule 1 in vascular endothelial cells.12 Elevated levels of soluble HSP60 (sHSP60) have been found in patients with borderline hypertension13 and correlate with high-carotid intimamedia thickness. In a large population-based study, we14 have demonstrated previously that levels of circulating sHSP60 are associated with the presence and severity of carotid atherosclerosis and that subjects with high sHSP60 experienced faster progression of atherosclerosis in the 5 years before the assessment of sHSP. The correlations were independent of age, gender, and other established risk factors. However, these studies do not permit assessment of the temporal relationship between high sHSP60 levels and progression of arteriosclerosis. In the current evaluation, we reassess the association of sHSP60 with progression of arteriosclerosis during follow-up of the same population. The hypothesis was that sHSP60 levels predict the progression of arteriosclerosis during follow-up.
| Methods |
|---|
|
|
|---|
Clinical Examination and Laboratory Methods
All of the participants completed standardized questionnaires on current and past exposure to candidate vascular risk factors and underwent a clinical examination with cardiovascular and neurological priority.14,18 Vascular risk factors were assessed by standard methods as detailed previously.1418 Subjects with chronic infections were identified by extensive clinical and laboratory screening as described previously.18 Briefly, an extensive screening consisting of 2 consecutive phases identified subjects with chronic infections. The first step involved a detailed self-reported medical and medication history, thorough clinical examination, spirometry, extensive laboratory evaluations including urinary analysis, and a review of the Bruneck Hospital databases and other medical records. If the data were inconclusive, in a second step, individuals were referred for additional optional examinations. For example, bronchitis was defined as chronic when cough with expectoration lasted
3 months in 2 consecutive years. Urinary tract infections were regarded as recurrent in the case of
3 documented episodes. Periodontitis was defined by self-report.
sHSP60 Assay
The serum sHSP60 concentrations were analyzed using a slight modification of sandwich ELISA described previously.14,19 The two monoclonal antibodies used were II-13 and N-20 (Santa Cruz Biotechnology), each reacting against separate nonoverlapping epitopes of HSP60.
Assays of Antilipopolysaccharide, Antimycobacterial HSP65, Antichlamydia, and Anticytomegalovirus Antibodies
The procedure for determining antilipopolysaccharide antibodies was similar to that described elsewhere.20 Serum anti-HSP65 antibodies were measured using the ELISA technique described previously.2 Serum antibodies against Chlamydia pneumoniae and cytomegalovirus (CMV) were determined using the following commercially available assays, as per their instructions: SeroCP-IgA (Savyon Diagnostics Ltd) and CMV-IgG ELISA (Medac).
Scanning Protocol and Definition of Ultrasound End Points
The ultrasound protocol involves scanning of the internal (bulbous and distal segments) and common (proximal and distal segments) carotid arteries on either side with a 10-MHz imaging probe and 5-MHz Doppler.17,21 Atherosclerotic lesions were identified by 2 ultrasound criteria: (1) wall surface (protrusion into the lumen or roughness of the arterial boundary); and (2) wall texture (echogenicity). Accuracy of this procedure has been established previously.15 Incident atherosclerosis was defined by the occurrence of atherosclerotic lesions in vessel segments initially free of atherosclerosis and progression of nonstenotic lesions by a relative increase in the plaque diameter exceeding twice the measurement error of the method.15,17,21 Both processes were combined to a single outcome category called "early atherogenesis."15 Incident vessel stenosis was assumed whenever the progression criterion was met and a narrowing of the lumen >40% occurred. This stage of atherogenesis was termed "advanced atherogenesis."1215 The 2 progression categories were highly reproducible (
coefficients >0.8; n=100).
Statistical Analysis
Correlations between sHSP60 and other parameters were estimated by Spearman rank correlation coefficients. Strength and type of association between baseline sHSP60 (1995) and various stages of atherogenesis (1995 to 2000) were assessed by logistic regression analysis, with the test procedure based on maximum likelihood estimators. The accuracy of fit of each model was assessed by the test of Hosmer and Lemeshow. As to early atherogenesis (5-year incidence/progression of carotid atherosclerosis), separate equations were fitted in the entire population (n=684) and in subjects free of atherosclerosis at the 1995 evaluation. Base models were adjusted for age and gender±baseline atherosclerosis. Multiple regression analyses were adjusted for fixed sets of covariates that were assessed in previous analyses of the vascular risk profile of the Bruneck study population.15 sHSP60 was loge transformed and odds ratios (95% CI) were calculated for a 1-SD unit increase in variable levels. In additional analysis, quintile groups of sHSP60 were modeled as a set of indicator variables. We performed a test for linear trend by treating the medians in each category of sHSP60 as a continuous variable.
The current analysis and the manner of risk factor adjustment were prespecified to avoid multiple testing. Calculations were performed using the SPSS 11.5 and BMDP software packages. A 2-sided P<0.05 was considered significant.
| Results |
|---|
|
|
|---|
Of 684 subjects, 330 showed early atherogenesis; that is, they developed new atheroslcerotic lesions or showed progression of preexisting nonstenotic plaques. Among the 350 subjects free of carotid atherosclerosis at the 1995 evaluation, 75 developed first atherosclerotic plaques.
Table 1 lists the means and proportions of selected demographic characteristics and risk factors in the study population. Additional details have been presented previously.14 A large number of epidemiological studies have reported about the association between atherosclerosis and various persistent bacterial and viral infections, including chlamydia and CMV.2225 We were able to show positive correlations between sHSP60 and antipathogen/anti-HSP antibodies in both the 1995 and 2000 evaluation. In both the 1995 and 2000 evaluation, we were able to show positive correlations of sHSP60 with anti-LPS antibodies (r=0.37 and 0.28; P<0.001 each), anti-HSP65 antibodies (r=0.21 and 0.27; P<0.001 each), and anti-HSP60 antibodies [r=0.26 (measured in 2000 only); P<0.001]. Correlations were weak and, in part, inconsistent for antichlamydia IgA (r=0.08 and 0.07; P=0.035 and 0.08) and anti-CMV IgG (r=0.02 and 0.09; P=0.59 and P=0.02). This has been detailed in Table 2. In addition, high levels of sHSP60 were more common among subjects with chronic infections than in those without (top quintile group of HSP60: 23.8% versus 17.0%, P=0.003 after adjustment for age and sex). No correlation was found to exist between C-reactive protein level and sHSP60.
|
|
Loge-transformed sHSP60 levels were significantly associated with early atherogenesis both in the entire population and subgroup free of atherosclerosis at 1995 baseline (Table 3). To rule out confounding by other vascular risk factors, multiple logistic regression models were adjusted for age, gender, baseline atherosclerosis, smoking, hypothyreosis, hypertension, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, microalbuminuria, ferritin, and alcohol consumption (Table 2). The association between sHSP60 and early atherosclerosis progression was similar in both genders and all age groups (data not presented). Figure 1 illustrates the relationship between early atherogenesis and sHSP60 quintile groups (quintile I+II was used as the reference group). Significant associations were obtained in both the base and multivariate models and in the entire population and the subgroup free of atherosclerosis at the 1995 baseline. In the latter group, a gradual increase of risk, across the quintile groups, was found. Presence of chronic infection enhanced the predictive significance of sHSP for early atherogenesis (Figure 2). For example, in the subgroup free of atherosclerosis in 1995, the odds ratio for incident carotid atherosclerosis was 2.0 in subjects with high sHSP60, but this increased to 4.2 when high sHSP60 coexisted with chronic infection (Figure 2A). No association, however, was found to exist between sHSP60 and advanced atherogenesis [multivariate odds ratio (95%CI) calculated for a 1-SD unit increase in loge-transformed sHSP60 0.92 (0.69 to 1.23), P=0.59].
|
|
|
| Discussion |
|---|
|
|
|---|
Proteins of the HSP60 family are mostly located in the intracellular compartment (mitochondria and cytosol) and are only expressed on the cell surface under stressful conditions.2628 In this study, we were able to demonstrate a sustained presence of circulating sHSP60 in human blood, over a 5-year observation period. The exact nature and composition of the measured sHSP60 in human serum is yet to be established, but there are a few possibilities. One of the likely sources could be infectious agents, like chlamydia, especially during the lytic phases of their life cycle.29 The association between sHSP60 and antipathogen antibodies and chronic infection in our study and in other reports14 and their colocalization in human atheromas10 lend additional support to this theory. Alternatively, surface-expressed HSP6026,30 on stressed cells undergoing apoptosis may be released into the circulation as microparticles. This is corroborated by the finding of circulating microparticles in patients with acute coronary syndrome31 and its correlation with the degree of endothelial dysfunction in these patients.32
In our study, we were able to demonstrate an association between sHSP60 and early but not advanced carotid atherosclerosis. It is interesting that a positive correlation has been demonstrated to exist between peripheral blood T cells reactive to HSP60 and the extent of early vessel pathology as indicated by a high-carotid intimamedia thickness in young army recruits.33 However, a similar association could not be demonstrated in middle-aged and elderly people with advanced atherosclerosis, suggesting that immunity may have a preferential role in early stages of atherosclerosis.
Previous studies have shown that T cells in human atheromas are mostly T-helper 1 cells bearing the
/ß receptor;34 however, in the earliest stages of atherosclerosis, there is a relative abundance of T cells bearing the
/
receptor.34,35 One of the probable antigens recognized by these T cells is HSP, a hypothesis supported by the isolation of T cells from rabbit atheromas5 and atherosclerotic plaques obtained from humans,36 specifically responding to HSP60/65. Interleukin 2-expanded T-cell lines derived from atherosclerotic lesions showed a significantly higher HSP60/65 reactivity compared with the cells derived from peripheral blood of the same donor. Recent work by Zal et al37 demonstrated increased numbers of CD4+CD28null T cells in the peripheral blood of patients with acute coronary syndrome in comparison to those with chronic stable angina. These natural killer cells were found to be activated by human HSP60 but not cytotomegalovirus, C. pneumonia, oxidized low-density lipoprotein, or lipopolysaccharide in a MHC class II-dependent manner. This observation additionally supports the role of HSP60 as an important autoantigen involved in atherosclerosis.
The presence of chronic infection further enhanced the predictive significance of sHSP for early atherogenesis (Figure 2). sHSP60 is known to mediate adhesion of monocytes to endothelial cells via the CD14 receptor.38 It has also been suggested that HSP60 has a cytokine-like function39 and elicits a release of tumor necrosis factor
from macrophages, causing expression of E-selectin, intercellular adhesion molecule 1, and vascular cell adhesion molecule 1.11,12 A total of 5.4% of the subjects in our study had concentrations >1000 ng/mL. Such high concentrations have been demonstrated in various in vitro studies to maximally activate macrophages, endothelial cells,11,12 and T cells.37 Similarly, the presence of infections will not only supplement these proinflammatory responses but may also enhance the pool of circulating HSPs2225 and lead to the development of early atherosclerosis.40
In summary, we provide the first truly prospective data confirming an association of sustained elevation of sHSP60 levels with early carotid atherosclerosis. This finding is consistent with the involvement of innate and adaptive immunity in atheroslcerosis and lends additional support to the role of HSP60 as an important candidate autoantigen.
| Acknowledgments |
|---|
Received June 8, 2005; revision received September 8, 2005; accepted September 16, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. RIGANỎ, E. PROFUMO, B. BUTTARI, A. TAGLIANI, L. PETRONE, G. D'AMATI, F. IPPOLITI, R. CAPOANO, L. FUMAGALLI, B. SALVATI, et al. Heat Shock Proteins and Autoimmunity in Patients with Carotid Atherosclerosis Ann. N.Y. Acad. Sci., June 1, 2007; 1107(1): 1 - 10. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2005 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |