Background: Lectin-like oxidized LDL receptor 1 (LOX-1) is implicated in atherothrombotic diseases. Activation of LOX-1 in humans can be evaluated by use of the LOX index, obtained by multiplying the circulating concentration of LOX-1 ligands containing apolipoprotein B (LAB) times that of the soluble form of LOX-1 (sLOX-1) [LOX index = LAB × sLOX-1]. This study aimed to establish the prognostic value of the LOX index for coronary heart disease (CHD) and stroke in a community-based cohort.
Methods: An 11-year cohort study of 2437 residents age 30–79 years was performed in an urban area located in Japan. Of these, we included in the analysis 1094 men and 1201 women without history of stroke and CHD. We measured LAB and sLOX-1 using ELISAs with recombinant LOX-1 and monoclonal anti–apolipoprotein B antibody and with 2 monoclonal antibodies against LOX-1, respectively.
Results: During the follow-up period, there were 68 incident cases of CHD and 91 cases of stroke (with 60 ischemic strokes). Compared with the bottom quartile, the hazard ratio (HR) of the top quartile of LOX index was 1.74 (95% CI 0.92–3.30) for stroke and 2.09 (1.00–4.35) for CHD after adjusting for sex, age, body mass index, drinking, smoking, hypertension, diabetes, non-HDL cholesterol, and use of lipid-lowering agents. Compared with the bottom quartile of LOX index, the fully adjusted HRs for ischemic stroke were consistently high from the second to the top quartile: 3.39 (95% CI 1.34–8.53), 3.15 (1.22–8.13) and 3.23 (1.24–8.37), respectively.
Conclusions: Higher LOX index values were associated with an increased risk of CHD. Low LOX index values may be protective against ischemic stroke.
Therapeutic interventions for dyslipidemia such as hypercholesterolemia have proven their effectiveness for the primary as well as secondary prevention of coronary heart disease (CHD).1 It is also well known that the risk for CHD is significantly associated with high serum concentrations of LDL cholesterol or low concentrations of HDL cholesterol in both Japanese (1)(2) and Western populations (3). In contrast, dyslipidemia has much weaker relationship to stroke than CHD (4). Although the pathogenesis of ischemic stroke and CHD is based largely on atherosclerotic changes of arteries, there is still much unresolved discrepancy.
Oxidized LDL induces a wide variety of cellular responses, such as induction of the expression of adhesion molecules and proinflammatory cytokines, which enhance progression of atherothrombotic cardiovascular diseases. Using antibodies against oxidation-dependent epitopes of LDL, cross-sectional studies reported association of oxidized LDL concentrations with ischemic heart disease, and a cohort study reported association of oxidized LDL with metabolic syndrome (5)(6)(7)(8).
Lectin-like oxidized LDL receptor 1 (LOX-1) is the receptor for oxidized LDL identified in endothelial cells (9)(10). Activation of LOX-1 in endothelial cells induces various changes relevant to endothelial dysfunction, e.g., superoxide generation, reduction in the release of nitric oxide, and induction of the expression of monocyte chemoattractant protein 1 (MCP-1) and adhesion molecules (11)(12)(13). In addition to oxidized LDL, LOX-1 binds various ligands, e.g., apoptotic cells, activated platelets, leukocytes, and C-reactive protein (14)(15)(16)(17). Accumulating evidence suggests that LOX-1 is involved in endothelial dysfunction, inflammation, atherogenesis, myocardial infarction, and intimal thickening after balloon catheter injury (16)(18)(19)(20)(21)(22)(23).
Recently, we developed a system to measure the biological activity of apolipoprotein B (ApoB)-containing lipoprotein based on binding to LOX-1 (24). The activity of LOX-1 ligand containing ApoB (LAB) might reflect atherogenicity of LDL better than measurements of oxidized lipids, oxidized LDL, and LDL. In addition, recent reports have shown that the serum concentrations of soluble LOX-1 (sLOX-1), which is released from the cell surface by proteolysis of LOX-1, might be a useful biomarker for the diagnosis of acute coronary syndrome (25)(26). Accordingly, we hypothesized that the product of LAB and sLOX-1, here designated “LOX index,” might be an even better marker reflecting the interaction of atherogenic lipoproteins and their receptors.
Materials and Methods
The Suita Study is a population-based cohort study in an urban area performed by the National Cardiovascular Center, the details of which have been reported (2)(27)(28). Briefly, in 1989, 6485 men and women, aged 30–79 years, were enrolled as study participants randomly selected from the community of Suita City. They underwent medical examinations every 2 years. In these participants, we set the baseline of the present study as the medical examination held between April 1994 and February 1995, since at that time serum samples were collected and stored at −80 °C. During this 10-month time period, 2437 participants were followed until December 31, 2007. Of these, 142 participants were excluded or the following reasons: history of CHD or stroke (n = 94), lost to follow-up (n = 17), and other reasons such as missing data (n = 31). Data from the remaining 2295 participants (1094 men and 1201 women) were included in the analysis. Informed consent was obtained from all participants. This study was approved by the institutional review board at the National Cardiovascular Center.
baseline medical examination
A baseline survey included questionnaires, anthropometric measurements, and blood sample testing after overnight fasting (at least 10 h). Height and weight were measured in light clothing, and body mass index (BMI) was calculated as weight (kg) divided by height (m) squared. Blood pressure of participants in a sitting position after at least 5 min of rest was measured 3 times by well-trained physicians, using a standard mercury sphygmomanometer (27). The average of the second and third measurement was used in the analysis. Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, and/or the use of antihypertensive agents. Serum total cholesterol (TC), HDL cholesterol, and fasting serum glucose were analyzed with an automated analyzer at the laboratory of the National Cardiovascular Center. Non-HDL cholesterol was calculated by subtracting HDL cholesterol from TC. Diabetes was defined as serum glucose concentrations ≥7.0 mmol/L (126 mg/dL) in fasting or ≥11.1 mmol/L (200 mg/dL) in nonfasting samples and/or current use of medications for diabetes. Well-trained health nurses obtained information on the smoking, alcohol drinking, and medical histories of the participants.
measurement of lab
A schematic presentation of the detection system for sLOX-1 and LAB is shown in Fig. 1⇓ .
We immobilized recombinant human LOX-1(61–273) (0.25 μg/well) on 384-well plates (Greiner 384 Plate High Bind 781061) by incubating overnight at 4 °C in 50 μL PBS (9)(24). After 3 washes with PBS, we blocked the plates with 80 μL of 3% BSA in HEPES buffer (10 mol/L HEPES, 150 mmol/L NaCl, pH 7.0). After 3 washes with PBS, the plates were incubated for 2 h at room temperature with 40 μL standard oxidized LDL or samples. We prepared samples by 20-fold dilution of serum with EDTA-BSA-HEPES buffer [2 mmol/L EDTA, 5% BSA/HEPES buffer (10 mmol/L HEPES, 150 mmol/L NaCl, pH 7.0)] and standards by dilution of oxidized LDL with EDTA-BSA-HEPES buffer. After 3 washes with PBS, the plates were incubated for 1 h at room temperature with 83 μg/L chicken monoclonal anti-ApoB antibody (HUC20) in EDTA-BSA-HEPES buffer (24). After 3 washes with PBS, the plates were incubated for 1 h at room temperature with peroxidase-conjugated donkey antichicken IgY (AP194P; Chemicon) diluted 6000 times with EDTA-BSA-HEPES buffer. After 5 washes with PBS, we added the substrate solution containing 3,3′,5,5′-tetramethylbenzidine (TMB solution; Bio-Rad) to the plates and incubated them for 30 min at room temperature. The reaction was terminated with 2 mol/L sulfuric acid. We determined peroxidase activity by measuring absorbance at 450 nm; the functional sensitivity of the measurement was 7.8 μg/L, and the range of the measurement was 7.8–500 μg/L oxidized LDL. Imprecision (CV) was 7.5% intraassay and 12.5% interassay at 50 μg/L (n = 10).
measurement of slox-1
We immobilized antihuman LOX-1 antibody (TS92, 0.25 μg/well) on 384-well plates (Corning 384 Plate High Bind 3700) by incubating overnight at 4 °C in 50 μL PBS (9). After 3 washes with PBS, the plates were blocked with 20% ImmunoBlock (DS Pharma). After 3 washes with PBS, the plates were incubated with 40 μL standard oxidized LDL or samples for 2 h at room temperature. We prepared samples by 4-fold dilution of the serum with 1% BSA/PBS containing 0.04% Tween20 and 2 mmol/L EDTA and also by dilution of recombinant extracellular LOX-1(61–273) with the same buffer. After 3 washes with PBS, the plates were incubated with 0.16 μg/mL chicken monoclonal antihuman LOX-1 antibody (HUC5–40) in PBS containing 0.04% Tween20 and 2 mmol/L EDTA for 1 h at room temperature (29). After 3 washes with PBS, the plates were incubated with the peroxidase-conjugated donkey antichicken IgY diluted 5000 times. After 5 washes with PBS, the substrate solution containing TMB solution was added to the plates and incubated for 30 min at room temperature. The reaction was terminated with 2 mol/L sulfuric acid. We determined peroxidase activity by measuring absorbance at 450 nm; the functional sensitivity of the measurement was 15.6 ng/L and the range of the measurement of sLOX-1 was 15.6–2500 ng/L. Imprecision (CV) was 8.5% intraassay and 14.7% interassay at 150 ng/L (n = 10).
The method of endpoint determination for the Suita Study has been reported (2)(27)(28). The endpoints of the current follow-up study were (1) date of first CHD or stroke event; (2) date of death; (3) date of leaving Suita city; and (4) December 31, 2007.
The first step in the survey for CHD and stroke involved checking the health status of all participants by repeated clinical visits every 2 years and yearly questionnaires sent by mail or conducted by telephone. In the second step, in-hospital medical records of participants who were suspected of having CHD or stroke were reviewed by registered hospital physicians or research physicians who were blinded to the baseline information. To complete the surveillance for fatal CHD and stroke, we conducted a systematic search for death certificates. The criteria for stroke were defined according to US National Survey of Stroke criteria (30). Classification of patients into stroke subtypes (ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage) was based on examination of computed tomography, magnetic resonance imaging, or autopsy. Definite and probable myocardial infarction (MI) were defined according to the criteria of the MONICA (Monitoring Trends and Determinants of Cardiovascular Disease) project (31). The criteria for a diagnosis of CHD included first-ever MI, coronary artery bypass surgery, or angioplasty. Sudden deaths of unknown origin that occurred within 24 h from onset were classified as CHD in the present study. We also defined cardiovascular disease (CVD) as a composite outcome of CHD or stroke.
In addition to sLOX and LAB, we calculated the LOX index by multiplying serum concentrations of sLOX-1 by those of LAB. We set the cutoff points of sLOX, LAB, and LOX index according to the quartile ranges. Statistical methods of data analysis included ANOVA for assessing mean differences between groups and χ2 tests for proportions. Multivariable analysis combining patients of both sexes was performed because there was no interaction between sex and LOX index (or LAB, sLOX) on the incidence of CHD or stroke. We calculated the multivariable-adjusted hazard ratios (HRs) of sLOX, LAB, and LOX index for CHD or stroke using a proportional hazards regression model after adjusting for sex, age, hypertension, diabetes, use of lipid-lowering agent, BMI, and current smoking and alcohol drinking (model 1). Further adjustment for non-HDL cholesterol was also performed (model 2). All CIs were estimated at the 95% level, and significance was set at P < 0.05. We used the SAS Statistical Package (release version 8.2, SAS Institute) for all the analyses.
LAB and sLOX-1 concentrations at baseline [mean (SE)] were 516.1 (17.1) μg/L and 1060.1 (8.6) ng/L in men and 782.3 (23.7) μg/L and 797.8 (0.2) ng/L in women. The mean baseline serum TC was 181.5 (1.3) mg/dL [4.70 (0.034) mmol/L] in men and 224.5 (2.0) mg/dL [5.81 (0.052) mmol/L] in women in this population. Table 1⇓ shows the baseline characteristics of the participants in each LAB or sLOX-1 quartile. In both sexes, there were significant differences in the mean concentrations for TC and non-HDL cholesterol according to LAB quartile, with higher concentrations in the higher LAB quartiles. Conversely, serum TC and non-HDL cholesterol were not found to be associated with sLOX-1 quartiles. In women, HDL cholesterol was lower in the higher LOX-1 quartiles. The prevalence of smoking was higher in the upper sLOX-1 quartiles but was not associated with LAB. There were no significant differences across quartiles in the prevalence of hypertension and diabetes.
During the mean follow-up period of 11 years, there were 68 incident cases of CHD and 91 cases of stroke, including 60 cases of ischemic stroke. The number of incident cases and multivariable-adjusted HRs for CVD, stroke, ischemic stroke, and CHD stratified by LAB and sLOX-1 are shown in Table 2⇓ . The HRs for stroke, ischemic stroke, and CHD were highest in the highest LAB quartile, and except for CHD, the trends in HRs across quartiles were statistically significant. The HR for CVD was highest in the top quartile of LAB, and the trend across quartiles was statistically significant. For sLOX-1, however, the across-quartile trends in HRs did not reach statistical significance.
The number of incident cases and multivariable-adjusted HRs for CVD, stroke, ischemic stroke, and CHD stratified by LOX index (LAB × sLOX-1) are shown in Table 3⇓ . The HR for ischemic stroke was constantly high from the second to the highest quartile: 3.39 (95% CI 1.34–8.53), 3.15 (1.22–8.13), and 3.23 (1.24–8.37), respectively. Furthermore, the HR of the highest quartile of LOX index was 2.09 (1.00–4.35) for CHD. In the highest LOX index quartile, the incidence of CVD was approximately 2-fold that in the lowest LOX index quartile, and the associated HR was 1.83 (1.03–2.96).
After additional adjustment for HDL cholesterol or the exclusion of sudden cardiac death from CHD, the results of all the analyses listed above remained the same (data not shown).
In the present study, we followed 1094 men and 1201 women for a mean period of 11 years to investigate the impact of LAB, sLOX-1, and LOX index (LAB × sLOX) on the incidence of CVD. We found LAB and LOX index to be significantly associated with the incidence of CVD and CHD, especially ischemic stroke. This investigation is the first cohort study on the relationship between CVD and LOX index–related or oxidized LDL–related parameters in a general population.
Instead of oxidized LDL, here we measured the serum concentrations of LAB. Researchers have applied several methods to measure circulating concentrations of oxidized LDL, including measurement of oxidation-dependent epitopes in the ApoB moiety of LDL. These methods, however, which evaluate the amount of oxidized moiety on LDL, do not necessarily reflect biological activity. In contrast, the present assay system using recombinant LOX-1 and anti-ApoB antibody has the ability to evaluate the biological activity of the atherogenic lipoproteins (Fig. 1⇑ ). On the other hand, circulating sLOX-1 concentrations might reflect the expression levels of LOX-1, the target site of the atherogenic lipoproteins in vascular wall. Therefore, LOX index (LAB × sLOX-1) could represent ligand (LAB)–receptor (LOX-1) interaction leading to vascular dysfunction. The present results confirmed LOX index as a predictor of the incidence of CVD. This suggests that LOX-1 may be important in the pathogenesis of CVD, and indicates that the evaluation of LOX-1–mediated signaling may serve as a potential tool for risk stratification.
It is well known that increased blood pressure, smoking, diabetes, and atrial fibrillation are major risk factors for stroke, especially for ischemic stroke (32). In contrast with these risk factors, we found either no relationship or a weakly positive one between TC or LDL cholesterol and ischemic stroke in several cohort studies performed in the Japanese population (32)(33)(34). A large metaanalysis of individual data from 61 prospective studies performed mainly in Western populations also showed no association between TC or non-HDL cholesterol and stroke mortality (4). We recently reported that we found no association between LDL cholesterol or non-HDL cholesterol concentrations and the incidence of ischemic stroke in this cohort using another baseline survey, shortly before that of the present study (2). In contrast, the present investigation demonstrated that the LOX index is a predictor of not only CHD but also ischemic stroke. A strong association of LOX-1 with ischemic stroke in experimental models has been reported. For example, expression of LOX-1 and MCP-1 is increased in the early stage of atherosclerotic changes of common carotid arteries in spontaneously hypertensive rats (35). Schwarz et al.(36) reported that LOX-1 expression was induced >10-fold at ischemic core sites during experimental stroke. Furthermore, we found that LOX-1 contributed to the formation of arterial thrombus (unpublished data). Thus, activation of LOX-1 might facilitate the pathophysiological conditions leading to stroke. In the present study, the risk for ischemic stroke was significantly increased from the second to the fourth quartile of LOX index, which suggests a protective role against ischemic stroke. Additional epidemiologic studies to establish clinical cut points of LOX index are warranted.
The positive relationship between LOX index and ischemic stroke bridged, for the first time, the missing link between stroke and cholesterol-related parameters. In those with high LOX index, breaking the interaction between LAB and LOX-1 might be effective in preventing stroke. The most straightforward approach would be to apply a LOX-1 antagonist, which is yet to be developed. Although most of the randomized controlled trials failed to find a beneficial effect of antioxidants for the prevention of CVD (21), we may reduce LAB concentration per se by statin therapy, thereby increasing LDL receptor expression, leading to an increase in the turnover of LDL and a decrease in the chance of LDL modification (37). Actually, in the present study, serum concentrations of LAB showed a significant association with TC and non-HDL cholesterol. In addition, a positive relationship between smoking and LOX-1 suggests the possibility of smoking cessation to reduce sLOX-1 concentration by decreasing the expression level of LOX-1.
The present study has some limitations. First, a recent report from the Hisayama study showed a positive relationship between LDL cholesterol and atherothrombotic infarction, which accounts for one fourth of all ischemic stroke (38). Therefore, the relation between LOX index and each subtype of ischemic stroke would be worth analyzing; however, the relatively small sample size of the current study precludes such analysis. Second, the participants in the present investigation were all Japanese; therefore, the study should be repeated in other ethnic populations. Because the incidence of stroke in Japan is much greater than in other countries (39), some Japanese-specific factors might be affecting the present findings. Finally, since the number of cardiovascular events was not sufficiently large to enable a sex-specific analysis, especially in women, we did not perform such analysis.
In conclusion, LOX index is associated with an increased risk of CVD, especially ischemic stroke, in a Japanese urban population. From a public health viewpoint, the novel biochemical marker may provide new insights into not only risk stratification but also therapeutic strategy for CVD.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors’ Disclosures of Potential Conflicts of Interest: Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest:
Employment or Leadership: None declared.
Consultant or Advisory Role: None declared.
Stock Ownership: None declared.
Honoraria: None declared.
Research Funding: This study was supported in part by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan; the Ministry of Health, Labour and Welfare of Japan; the National Institute of Biomedical Innovation; Japan Science and Technology Agency; and the New Energy and Industrial Technology Development Organization. The authors have declared that no conflict of interests exist.
Expert Testimony: None declared.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
Acknowledgments: T. Sawamura had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
We acknowledge the members of Suita City Health Center, the Suita Medical Association, and Satsuki-Junyukai, the volunteers involved in the administration of the Suita Study.
1 Data are mean (SE) unless noted otherwise.
2 Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or use of antihypertensive agents.
3 Diabetes was defined as fasting serum glucose ≥7.0 mmol/L (126 mg/dL), use of antidiabetic agents, or both.
4 Alcohol drinking was defined as consuming at least 1 drink per week.
1 Model 1 adjusted for age, sex, BMI, smoking, drinking, hypertension, diabetes, and use of lipid-lowering agents; model 2 as model 1 with the addition of non-HDL cholesterol.
1 Model 1 adjusted for age, sex, BMI, smoking, drinking, hypertension, diabetes, and use of lipid-lowering agents; model 2 as model 1 with the addition of non-HDL cholesterol.
↵1 Nonstandard abbreviations: CHD, coronary heart disease; LOX-1, lectin-like oxidized LDL receptor 1; MCP-1, monocyte chemoattractant protein 1; ApoB, apolipoprotein B; LAB, LOX-1 ligand containing ApoB; sLOX-1, soluble LOX-1; TC, total cholesterol; MI, myocardial infarction; MONICA, Monitoring Trends and Determinants of Cardiovascular Disease; CVD, cardiovascular disease; HR, hazard ratio; BMI, body mass index.
- © 2010 The American Association for Clinical Chemistry