Advanced glycation end products (AGEs) and their specific receptor, the receptor for AGEs (RAGE), play an important role in atherosclerosis. Recently, a soluble form of RAGE (sRAGE) has been identified in human serum. However, the role of sRAGE in cardiovascular disease is still controversial. There is no information on the association between simultaneous measurements of AGEs and sRAGE and vascular function. In this study, we evaluated the associations between serum levels of AGEs and sRAGE, ratio of AGEs to sRAGE, and vascular function.
We measured serum levels of AGEs and sRAGE and assessed vascular function by measurement of flow-mediated vasodilation (FMD) and nitroglycerine-induced vasodilation in 110 subjects who underwent health examinations. Multivariate regression analyses were performed to identify factors associated with vascular function.
Univariate regression analysis revealed that FMD correlated with age, BMI, systolic blood pressure, diastolic blood pressure, heart rate, triglycerides, HDL cholesterol, glucose, smoking pack-years, nitroglycerine-induced vasodilation, serum levels of AGEs and sRAGE, and ratio of AGEs to sRAGE. Multivariate analysis revealed that the ratio of AGEs to sRAGE remained an independent predictor of FMD, while serum level of AGEs alone or sRAGE alone was not associated with FMD.
These findings suggest that sRAGE may have a counterregulatory mechanism that is activated to counteract the vasotoxic effect of the AGE–RAGE axis. The ratio of AGEs to sRAGE may be a new chemical biomarker of endothelial function.
Introduction
Nonenzymatic modification of proteins by reducing sugars has been reported to play a crucial role in the pathogenesis of various types of diseases (1,2). It is well established, in vivo, that early glycation products react repeatedly with the amino groups of proteins and form irreversible cross-link moieties termed advanced glycation end products (AGEs) (3). There is growing evidence that interaction of AGEs and their specific receptor, the receptor for AGEs (RAGE), stimulates oxidative stress generation and plays an important role in cardiovascular disease and diabetes complications (4–6). Recently, a soluble form of RAGE (sRAGE) has been identified in human serum (7,8). It has been reported that administration of a recombinant sRAGE inhibits the development and progression of atherosclerosis in animal models (8,9). These beneficial effects of exogenous administration of sRAGE on the vasculature are assumed to be due to capture of circulating AGEs by acting as a decoy receptor.
Flow-mediated vasodilation (FMD) as an index of endothelium-dependent vasodilation and nitroglycerine-induced vasodilation as an index of endothelium-independent vasodilation in the brachial artery using high-resolution ultrasound provide useful information on vascular function (10–16). Endothelial dysfunction occurs in the early stage of atherosclerosis, leading to the maintenance and progression of atherosclerosis, resulting in cardiovascular complications (17). Evaluation of FMD is noninvasive and reflects nitric oxide production. Furthermore, previous studies have shown that endothelial function measured by FMD can serve as an independent predictor of cardiovascular events (18–21).
Recent studies have shown that serum levels of sRAGE are decreased in cardiovascular diseases, while other studies have shown that serum level of sRAGE reflects tissue RAGE expression and predicts future cardiovascular events (22–24). The reason for the controversial results remains unclear. Simultaneous measurements of serum levels of AGEs and sRAGE would enable more specific conclusions concerning the role of the AGE–RAGE axis in cardiovascular disease to be drawn. Ratio of AGEs to sRAGE may be a more useful biomarker reflecting the AGE–RAGE axis (25). In this study, we examined the association between the ratio of serum level of AGEs to sRAGE and vascular function.
Research Design And Methods
Subjects
A total of 110 subjects were enrolled from people who underwent health-screening examinations at Hiroshima University Hospital. Hypertension was defined as systolic blood pressure of >140 mmHg or diastolic blood pressure of >90 mmHg, in a sitting position, on at least three different occasions. Patients with secondary forms of hypertension were excluded in all patients with hypertension on the basis of complete history, physical examination, radiological and ultrasound examinations, urinalysis, plasma renin activity, plasma aldosterone and norepinephrine concentrations, serum creatinine, potassium, calcium, and free thyroxine concentrations, and 24-h urinary excretion of 17-hydroxycorticosteroids, 17-ketogenic steroids, and vanillymandelic acid. Diabetes mellitus was defined according to the American Diabetes Association (26). Dyslipidemia was defined according to the third report of the National Cholesterol Education Program (27). The ethical committees in our institutions approved the study protocol. Written informed consent for participation in the study was obtained from all subjects.
Study Protocol
We measured serum levels of AGEs and sRAGE and assessed vascular function by measurements of FMD and nitroglycerine-induced vasodilation in all subjects. Subjects fasted the previous night for at least 12 h. Subjects were instructed to abstain from smoking at least 12 h prior to the measurements. The study began at 8:30 a.m. The subjects were kept in the supine position in a quiet, dark, air-conditioned room (constant temperature of 22–25°C) throughout the study. A 23-gauge polyethylene catheter was inserted into the left deep antecubital vein to obtain blood samples. Thirty minutes after maintaining the supine position, FMD and nitroglycerine-induced vasodilation were measured. The observers were blind to the form of examination. Serum levels of glyceraldehyde-derived AGEs were measured with an ELISA as described previously (28). Serum sRAGE levels were determined with a commercially available ELISA kit (R&D Systems, Minneapolis, MN).
Measurement of FMD and Nitroglycerine-Induced Vasodilation
Vascular response to reactive hyperemia in the brachial artery was used for assessment of endothelium-dependent FMD. A high-resolution linear artery transducer was coupled to computer-assisted analysis software (UNEXEF18G; UNEX Co, Nagoya, Japan) that used an automated edge detection system for measurement of brachial artery diameter (29). A blood pressure cuff was placed around the forearm. The brachial artery was scanned longitudinally 5–10 cm above the elbow. When the clearest B-mode image of the anterior and posterior intimal interfaces between the lumen and vessel wall was obtained, the transducer was held at the same point throughout the scan by a special probe holder (UNEX Co) to ensure consistency of the image. Depth and gain setting were set to optimize the images of the arterial lumen wall interface. When the tracking gate was placed on the intima, the artery diameter was automatically tracked, and the waveform of diameter changes over the cardiac cycle was displayed in real time using the FMD mode of the tracking system. This allowed the ultrasound images to be optimized at the start of the scan and the transducer position to be adjusted immediately for optimal tracking performance throughout the scan. Pulsed Doppler flow was assessed at baseline and during peak hyperemic flow, which was confirmed to occur within 15 s after cuff deflation. Blood flow velocity was calculated from the color Doppler data and displayed as a waveform in real time. The baseline longitudinal image of the artery was acquired for 30 s, and then the blood pressure cuff was inflated to 50 mmHg above systolic pressure for 5 min. The longitudinal image of the artery was recorded continuously until 5 min after cuff deflation. Pulsed Doppler velocity signals were obtained for 20 s at baseline and for 10 s immediately after cuff deflation. Changes in brachial artery diameter were immediately expressed as percentage change relative to the vessel diameter before cuff inflation. FMD was automatically calculated as the percentage change in peak vessel diameter from the baseline value. Percentage of FMD ([peak diameter − baseline diameter]/baseline diameter) was used for analysis. Blood flow volume was calculated by multiplying the Doppler flow velocity (corrected for the angle) by heart rate and vessel cross-sectional area (−r2). Reactive hyperemia was calculated as the maximum percentage increase in flow after cuff deflation compared with baseline flow.
The response to nitroglycerine was used for assessment of endothelium-independent vasodilation. Nitroglycerine-induced vasodilation was measured as described previously (29). Briefly, after acquiring baseline rest images for 30 s, a sublingual tablet (75 μg nitroglycerine) was given, and images of the artery were recorded continuously until the dilation reached a plateau after administration of nitroglycerine. Subjects who had received nitrate treatment and subjects in whom the sublingually administered nitroglycerine tablet was not dissolved during the measurement were excluded from this study. Nitroglycerine-induced vasodilation was automatically calculated as a percent change in peak vessel diameter from the baseline value. Percentage of nitroglycerine-induced vasodilation ([peak diameter − baseline diameter]/baseline diameter) was used for analysis.
Statistical Analysis
We calculated the sample size based on previous data for the ratio of AGEs to sRAGE (25). For the current study, we estimated that 86 subjects were needed with an α = 0.05 and a power of 0.9. Finally, we enrolled 110 subjects with consideration for 20% dropouts. Results are presented as means ± SD for continuous variables and as percentages for categorical variables. Statistical significance was set at a level of P < 0.05. Spearman rank correlation analysis was used to determine associations between FMD and selected parameters. Multivariate linear regression analyses were performed to determine relationships between vascular function and serum levels of AGEs, sRAGE, and ratio of AGEs to sRAGE before (model 1) and after adjustment for age and sex (model 2), further adjusted for BMI, presence of hypertension, dyslipidemia, diabetes, and current smoking (model 3), and further adjusted for BMI, presence of hypertension, dyslipidemia, glucose levels, and current smoking (model 4). The data were processed using the software package Stata version 9 (Stata Co., College Station, TX).
Results
Clinical Characteristics
The baseline characteristics of the 110 subjects are summarized in Table 1. Of the 110 subjects, 80 (72.7%) were men and 30 (27.3%) were women. Forty-eight (43.6%) had hypertension, 37 (33.6%) had dyslipidemia, 13 (11.8%) had diabetes mellitus, and 49 (44.5%) were current smokers. The mean serum levels of AGEs and sRAGE were 9.5 ± 3.5 U/mL and 823 ± 388 pg/mL, respectively.
Clinical characteristics of the subjects
N | 110 |
Age, years | 46 ± 19 |
Sex, men/women | 80/30 |
BMI, kg/m2 | 23.0 ± 3.7 |
Systolic blood pressure, mmHg | 127 ± 19 |
Diastolic blood pressure, mmHg | 74 ± 14 |
Heart rate, bpm | 70 ± 13 |
Total cholesterol, mmol/L | 4.89 ± 0.91 |
Triglycerides, mmol/L | 1.49 ± 1.15 |
HDL cholesterol, mmol/L | 1.50 ± 0.44 |
LDL cholesterol, mmol/L | 2.79 ± 0.80 |
Glucose, mmol/L | 5.88 ± 1.33 |
HbA1c, % | 5.6 ± 0.7 |
HbA1c, mmol/mol | 37.8 ± 7.3 |
Hypertension, n (%) | 48 (43.6) |
Dyslipidemia, n (%) | 37 (33.6) |
Diabetes mellitus, n (%) | 13 (11.8) |
Previous coronary heart disease, n (%) | 6 (5.5) |
Previous stroke, n (%) | 3 (2.7) |
Smoker, n (%) | 49 (44.5) |
Smoking, pack-years | 15.1 ± 20.5 |
FMD, % | 5.4 ± 2.9 |
Nitroglycerine-induced vasodilation, % | 14.1 ± 6.2 |
AGEs, U/mL | 9.5 ± 3.5 |
sRAGE, pg/mL | 823 ± 388 |
Ratio of AGEs to sRAGE (×10−2), U/pg | 1.43 ± 0.90 |
N | 110 |
Age, years | 46 ± 19 |
Sex, men/women | 80/30 |
BMI, kg/m2 | 23.0 ± 3.7 |
Systolic blood pressure, mmHg | 127 ± 19 |
Diastolic blood pressure, mmHg | 74 ± 14 |
Heart rate, bpm | 70 ± 13 |
Total cholesterol, mmol/L | 4.89 ± 0.91 |
Triglycerides, mmol/L | 1.49 ± 1.15 |
HDL cholesterol, mmol/L | 1.50 ± 0.44 |
LDL cholesterol, mmol/L | 2.79 ± 0.80 |
Glucose, mmol/L | 5.88 ± 1.33 |
HbA1c, % | 5.6 ± 0.7 |
HbA1c, mmol/mol | 37.8 ± 7.3 |
Hypertension, n (%) | 48 (43.6) |
Dyslipidemia, n (%) | 37 (33.6) |
Diabetes mellitus, n (%) | 13 (11.8) |
Previous coronary heart disease, n (%) | 6 (5.5) |
Previous stroke, n (%) | 3 (2.7) |
Smoker, n (%) | 49 (44.5) |
Smoking, pack-years | 15.1 ± 20.5 |
FMD, % | 5.4 ± 2.9 |
Nitroglycerine-induced vasodilation, % | 14.1 ± 6.2 |
AGEs, U/mL | 9.5 ± 3.5 |
sRAGE, pg/mL | 823 ± 388 |
Ratio of AGEs to sRAGE (×10−2), U/pg | 1.43 ± 0.90 |
Results are presented as means ± SD for continuous variables and n (%) for categorical variables.
Relationships of Vascular Function With Cardiovascular Risk Factors, Serum Levels of AGEs and sRAGE, and Ratio of AGEs to sRAGE
FMD was negatively correlated with age, BMI, systolic blood pressure, diastolic blood pressure, heart rate, triglycerides, glucose, HbA1c, and smoking pack-years and was positively correlated with HDL cholesterol and nitroglycerine-induced vasodilation (Table 2).
Univariate analysis of relationships among FMD, nitroglycerine-induced vasodilation, and variables (Spearman rank correlation analysis)
Variables . | FMD . | Nitroglycerine-induced vasodilation . |
---|---|---|
Age, years | −0.381** | −0.454** |
Sex, men/women | 0.017 | −0.038 |
BMI, kg/m2 | −0.281** | −0.207* |
Systolic blood pressure, mmHg | −0.443** | −0.418** |
Diastolic blood pressure, mmHg | −0.460** | −0.352** |
Heart rate, bpm | −0.285** | −0.288** |
Total cholesterol, mmol/L | −0.165 | −0.098 |
Triglycerides, mmol/L | −0.371** | −0.191 |
HDL cholesterol, mmol/L | 0.193* | 0.206* |
LDL cholesterol, mmol/L | −0.102 | −0.076 |
Glucose, mmol/L | −0.204* | −0.290** |
HbA1c, % | −0.330** | −0.298* |
Smoking, pack-years | −0.318** | −0.293** |
FMD, % | 0.400** | |
Nitroglycerine-induced vasodilation, % | 0.400** | |
AGEs, U/mL | −0.340** | −0.260** |
sRAGE, pg/mL | 0.200* | 0.119 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.384** | −0.227* |
Variables . | FMD . | Nitroglycerine-induced vasodilation . |
---|---|---|
Age, years | −0.381** | −0.454** |
Sex, men/women | 0.017 | −0.038 |
BMI, kg/m2 | −0.281** | −0.207* |
Systolic blood pressure, mmHg | −0.443** | −0.418** |
Diastolic blood pressure, mmHg | −0.460** | −0.352** |
Heart rate, bpm | −0.285** | −0.288** |
Total cholesterol, mmol/L | −0.165 | −0.098 |
Triglycerides, mmol/L | −0.371** | −0.191 |
HDL cholesterol, mmol/L | 0.193* | 0.206* |
LDL cholesterol, mmol/L | −0.102 | −0.076 |
Glucose, mmol/L | −0.204* | −0.290** |
HbA1c, % | −0.330** | −0.298* |
Smoking, pack-years | −0.318** | −0.293** |
FMD, % | 0.400** | |
Nitroglycerine-induced vasodilation, % | 0.400** | |
AGEs, U/mL | −0.340** | −0.260** |
sRAGE, pg/mL | 0.200* | 0.119 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.384** | −0.227* |
*P < 0.05, **P < 0.01.
There were significant relationships between FMD and serum level of AGEs (r = −0.30; P = 0.001; Fig. 1A), serum level of sRAGE (r = 0.22; P = 0.02; Fig. 1B), and ratio of serum levels of AGEs to sRAGE (r = −0.33; P < 0.001; Fig. 1C). After adjustment for confounders, neither AGEs nor sRAGE correlated with FMD (Table 3). Multiple logistic regression analysis revealed that age and ratio of serum levels of AGEs to sRAGE remained independently associated with FMD (Table 3).
Scatter plots show the relationships between FMD and AGEs (A), sRAGE (B), and ratio of AGEs to sRAGE (C).
Scatter plots show the relationships between FMD and AGEs (A), sRAGE (B), and ratio of AGEs to sRAGE (C).
Multivariate analysis of relationships between vascular function and AGEs, sRAGE, and ratio of AGEs to sRAGE
Models used . | FMD . | Nitroglycerine-induced vasodilation . | ||
---|---|---|---|---|
β . | P value . | β . | P value . | |
Model 1 | ||||
AGEs, U/mL | −0.30 | 0.001 | −0.20 | 0.04 |
sRAGE, pg/mL | 0.22 | 0.02 | 0.13 | 0.20 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.33 | <0.001 | −0.17 | 0.08 |
Model 2 | ||||
AGEs, U/mL | −0.18 | 0.06 | −0.08 | 0.41 |
sRAGE, pg/mL | 0.15 | 0.09 | 0.06 | 0.53 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.21 | 0.02 | −0.04 | 0.67 |
Model 3 | ||||
AGEs, U/mL | −0.14 | 0.13 | 0.003 | 0.97 |
sRAGE, pg/mL | 0.13 | 0.15 | 0.02 | 0.84 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.22 | 0.03 | 0.02 | 0.84 |
Model 4 | ||||
AGEs, U/mL | −0.15 | 0.11 | 0.009 | 0.92 |
sRAGE, pg/mL | 0.14 | 0.13 | 0.03 | 0.78 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.23 | 0.02 | 0.04 | 0.70 |
Models used . | FMD . | Nitroglycerine-induced vasodilation . | ||
---|---|---|---|---|
β . | P value . | β . | P value . | |
Model 1 | ||||
AGEs, U/mL | −0.30 | 0.001 | −0.20 | 0.04 |
sRAGE, pg/mL | 0.22 | 0.02 | 0.13 | 0.20 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.33 | <0.001 | −0.17 | 0.08 |
Model 2 | ||||
AGEs, U/mL | −0.18 | 0.06 | −0.08 | 0.41 |
sRAGE, pg/mL | 0.15 | 0.09 | 0.06 | 0.53 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.21 | 0.02 | −0.04 | 0.67 |
Model 3 | ||||
AGEs, U/mL | −0.14 | 0.13 | 0.003 | 0.97 |
sRAGE, pg/mL | 0.13 | 0.15 | 0.02 | 0.84 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.22 | 0.03 | 0.02 | 0.84 |
Model 4 | ||||
AGEs, U/mL | −0.15 | 0.11 | 0.009 | 0.92 |
sRAGE, pg/mL | 0.14 | 0.13 | 0.03 | 0.78 |
Ratio of AGEs to sRAGE (×10−2), U/pg | −0.23 | 0.02 | 0.04 | 0.70 |
Model 1: unadjusted model; model 2: adjusted for age and sex; model 3: adjusted for age, sex, BMI, presence of hypertension, dyslipidemia, diabetes, and current smoking; and model 4: adjusted for age, sex, BMI, presence of hypertension, dyslipidemia, glucose levels, and current smoking.
Nitroglycerine-induced vasodilation was negatively correlated with age, BMI, systolic blood pressure, diastolic blood pressure, heart rate, glucose, HbA1c, and smoking pack-years and positively correlated with HDL cholesterol and FMD (Table 2).
There was a significant relationship between nitroglycerine-induced vasodilation and serum level of AGEs (r = −0.20; P = 0.04; Fig. 2A). Nitroglycerine-induced vasodilation was not correlated with serum level of sRAGE (r = 0.13; P = 0.20; Fig. 2B) or ratio of serum levels of AGEs to sRAGE (r = −0.17; P = 0.08; Fig. 2C). After adjustment for confounders, AGEs did not correlate with nitroglycerine-induced vasodilation (Table 3).
Scatter plots show the relationships between nitroglycerine-induced vasodilation and AGEs (A), sRAGE (B), and ratio of AGEs to sRAGE (C).
Scatter plots show the relationships between nitroglycerine-induced vasodilation and AGEs (A), sRAGE (B), and ratio of AGEs to sRAGE (C).
In the current study, 14 of the 110 subjects were healthy subjects. Serum levels of AGEs and ratio of AGEs to sRAGE were significantly higher in diabetic patients than in healthy subjects (6.8 ± 1.7 vs. 12.5 ± 3.5 U/mL and 0.83 ± 0.40 vs. 1.78 ± 0.81 ×10−2 U/pg; P < 0.001 and P = 0.001, respectively; Supplementary Fig. 1A and C). There was no significant difference in serum level of sRAGE between the two groups (920 ± 247 vs. 767 ± 242 pg/mL; P = 0.11; Supplementary Fig. 1B). FMD and nitroglycerin-induced vasodilation were significantly lower in diabetic patients than in healthy subjects (3.6 ± 2.8 vs. 8.7 ± 2.6% and 8.4 ± 3.4 vs. 15.6 ± 4.0%; P < 0.001 and P < 0.001, respectively; Supplementary Fig. 1D and E).
Conclusions
In the current study, we demonstrated that 1) the ratio of serum levels of AGEs to sRAGE was an independent predictor of endothelial dysfunction evaluated by FMD; and 2) while serum levels of AGEs and sRAGE were positively and negatively associated with endothelial dysfunction, respectively, the significance was lost after adjusting for various confounders.
Hypertension, dyslipidemia, diabetes, aging, smoking, and obesity are commonly known cardiovascular risk factors (10–12). These cardiovascular risk factors are independent contributing factors of endothelial dysfunction (10–12). Some investigators have shown the relationships among serum levels of carboxymethyllysine (CML), glucose-derived AGEs and sRAGE, and endothelial function (30–32). Furthermore, single oral AGEs challenge acutely impaired endothelial function evaluated by FMD in both diabetic and nondiabetic subjects, which was associated with increase in serum CML levels (30). These observations suggest that both AGEs and sRAGE levels are biomarkers for endothelial dysfunction. However, the potential problem with these studies is that they did not simultaneously measure AGEs and sRAGE levels. In this study, multivariate regression analysis revealed that the ratio of AGEs to sRAGE remained an independent predictor of endothelial dysfunction, although each value was not correlated with endothelial function. The present findings have extended our previous observation showing that the ratio of serum levels of AGEs to sRAGE was inversely associated with adiponectin, an anti-inflammatory and vascular protective adipocytokine with insulin-sensitizing properties (25). The ratio of AGEs to sRAGE rather than each parameter alone might be a useful biomarker of endothelial dysfunction in humans.
There is a growing body of evidence that the AGE–RAGE axis plays an important role in the development and progression of atherosclerosis (4–6). Engagement of RAGE with AGEs elicits oxidative stress generation, which could reduce NO production and/or bioavailability, thereby evoking inflammatory and thrombogenic reactions in a variety of cells (33). These observations support the active participation of the AGE–RAGE axis in endothelial dysfunction. In univariate analysis in the current study, besides an increased AGE level, a low sRAGE value was associated with endothelial dysfunction. However, it is unlikely that endogenous sRAGE could work as a decoy receptor for AGEs and play a protective role against atherosclerosis in humans because sRAGE level is 1,000 times lower than that needed for efficiently binding to and capturing serum AGEs (7). sRAGE is mainly generated from the cleavage of cell-surface full-length RAGE, the process of which is promoted by the engagement of RAGE with ligands such as AGEs and high-mobility group protein box-1 (7). Therefore, a low sRAGE level might be a consequence of impaired shedding of RAGE by ligands. In these circumstances, the AGE–RAGE axis could be further augmented. This is one possible reason why the ratio of AGEs to sRAGE is a more sensitive marker of endothelial dysfunction than each parameter alone.
Several prospective studies have shown that elevated serum level of sRAGE is associated with incidence of cardiovascular disease or all-cause mortality in diabetic subjects (7,22,23). However, there is some controversy about the role of sRAGE in cardiovascular disease in humans. Indeed, Falcone et al. (24) reported that a decreased sRAGE level was correlated with a higher prevalence of cardiovascular risk factors and coronary artery disease in nondiabetic subjects. Furthermore, a low plasma level of sRAGE at baseline was independently associated with the risk of coronary heart disease and all-cause mortality during a median follow-up period of 18 years in a community-based population (34). At present, the exact reason for the discrepant results about the relationship between sRAGE and cardiovascular disease is unclear. Clinical significance of sRAGE as a biomarker might differ considerably depending on the patients’ background.
The characterization of AGEs measured is important to interpret the present data and to make a detailed comparison with previous reports. There are multiple types of immunologically distinct and structurally identified AGEs such as glyoxal and methylglyoxal hydroimidazolones, pyrraline, and pentosidine in humans (35,36). However, they constitute a small percentage of circulating AGEs in vivo, and their biological relevance in endothelial dysfunction has remained unclear (35,36). In this study, we measured glyceraldehyde-derived AGEs to examine the role of AGEs and sRAGE in endothelial dysfunction because we previously found that 1) serum levels of glyceraldehyde-derived AGEs are elevated under inflammatory and/or diabetic conditions and correlated with vascular inflammation evaluated by 18F-fluorodeoxyglucose–positron emission tomography in humans; 2) this type of AGE mimics the biological effects of AGE-rich serum purified from diabetic patients on hemodialysis; and 3) deleterious and cytopathic effects of the AGE-rich serum were neutralized by addition of an antiglyceraldehyde-derived AGE-specific antibody but not by other types of anti-AGE antibodies (35–39). Therefore, although glyceraldehyde, which could be derived from glucose metabolism, is not a major sugar in vivo, and its incubation with proteins will generate a large number of structurally unidentified AGEs, and we cannot identify the structure of AGEs measured in this study, our study suggests the clinical utility of measurement of the ratio of serum levels of glyceraldehyde-derived AGEs to sRAGE for detecting endothelial dysfunction in humans. In regard to the characterization of AGEs measured, we previously reported that antibodies raised against glyceraldehyde-derived AGEs used for the ELISA did not cross-react with early glycation products (HbA1c and/or fructoselysine) or structurally identified AGEs such as glyoxal and methylglyoxal hydroimidazolones, CML, carboxyethyllysine, or pentosidine (35,36). So, these structurally identified AGEs were not evaluated in our ELISA system, although methylglyoxal hydroimidazolone could be formed through both glyceraldehyde-related and methylglyoxal-related pathways. Thus, besides glyceraldehyde-derived AGEs, it would be interesting to further examine whether the ratios of serum levels of these structurally identified AGEs to sRAGE are also associated with endothelial dysfunction.
Study Limitations
First, circulating sRAGE could be generated from cleavage of cell-surface RAGE or novel splice variants of RAGE (esRAGE) (7). In this study, we measured total levels of circulating sRAGE by ELISA. Therefore, although esRAGE levels are approximately three- to fivefold lower than levels of sRAGE (7), we did not know whether the ratio of AGEs to esRAGE was also associated with endothelial dysfunction. Second, to evaluate what percent of serum sRAGE was bound form with AGEs, we performed additional experiments using a Dynabeads Protein A Immunoprecipitation kit (Life Technologies, Carlsbad, CA). In brief, polyclonal antibody raised against AGEs used for ELISA or nonimmune rabbit IgG was first bound to Dynabeads Protein A, and then beads and serum were mixed and incubated at 4°C with tilting and rotation. After 1 h, the captured immune complexes were removed from the supernatant by magnetic separation according to the supplier’s recommendation. Immunoprecipitation of AGEs from serum did not affect sRAGE level; sRAGE values in AGE-cleared and noncleared serum were 100.0 ± 22.4 and 101.8 ± 21.7%, respectively (n = 5; P = 0.79). So, almost all of the sRAGE would be an unbound form with AGEs. Third, we have previously shown that addition of 10 U/ml glyceraldehyde-derived AGEs (which is comparable to the serum concentration in our subjects) to serum samples does not interfere with the assay system for sRAGE (37). Likewise, we also confirmed that addition of 800 pg/mL, a level comparable to the serum concentration, to assay samples did not affect circulating AGE values; addition of sRAGE decreased serum AGE levels to 91.3 ± 19.8% (n = 5; P = 0.42). Therefore, it is unlikely that binding of AGEs to sRAGE in serum could confound the present findings. Fourth, there are multiple other classes of RAGE ligands such as S100s or high-mobility group protein box-1 (7). It would be interesting to study the relationships among AGEs, these ligands, and endothelial function. Fifth, although several cell culture and animal models have shown that soluble AGEs could impair endothelial function (30–33), cross-linked/less soluble AGEs might be linked more specifically to changes in endothelial function. However, due to the difficulty in obtaining tissue samples from all patients, we could not evaluate the correlation of cross-linked AGEs to endothelial function. Sixth, although there was no significant difference in FMD between males and females in a previous study (29), it has been reported that FMD was significantly smaller in males than in females (14). In the current study, there were no significant differences in the serum levels of AGEs (9.6 ± 2.9 vs. 9.2 ± 4.7 U/mL; P = 0.67), sRAGE (824 ± 397 vs. 821 ± 369 pg/mL; P = 0.98), ratio of AGEs to sRAGE (1.44 ± 0.88 vs. 1.39 ± 0.96; P = 0.77), FMD (5.2 ± 2.7 vs. 5.7 ± 3.4%; P = 0.55), and nitroglycerine-induced vasodilation (14.5 ± 6.4 vs. 13.1 ± 5.3%; P = 0.31) between males and females. In addition, after adjustment for sex, our findings did not change. However, we cannot deny the possibility that the results might be confounded with more females in the cohort. Seventh, measurement of FMD was performed in a fasted state and with medication, although medication was discontinued for at least 12 h prior the measurement of FMD. In the current study, 12 (10.9%) of the subjects received antihypertensive drugs, 9 (8.2%) received lipid-lowering drugs, and 11 (10.0%) received hypoglycemic drugs. Since we enrolled subjects who underwent health examinations, it would have been inappropriate to stop medications. Gokce et al. (16) demonstrated that administration of vasoactive medication with the exception of nitrates did not significantly influence the value of FMD and nitroglycerine-induced vasodilation. However, we cannot also exclude the possibility that medication affects FMD and nitroglycerine-induced vasodilation. Finally, this was a cross-sectional study design and did not allow us to establish a definitive causal relationship among serum levels of AGEs and sRAGE, ratio of AGEs to sRAGE, and endothelial function. Previous studies have demonstrated associations between serum levels of AGEs and sRAGE and endothelial function (32,33). However, in the current study, neither the serum level of AGEs nor that of sRAGE correlated with endothelial function after adjustment for confounders. Future studies are needed to confirm the relationships among serum levels of AGEs and sRAGE, ratio of AGEs to sRAGE, and endothelial function in a larger population.
In conclusion, sRAGE might have a counterregulatory mechanism that is activated to counteract the vasotoxic effect of the AGE–RAGE axis. Our present study suggests that the ratio of AGEs to sRAGE may be a new biochemical marker of endothelial function.
Article Information
Acknowledgments. The authors thank Miki Kumiji, Megumi Wakisaka, Ki-ichiro Kawano, and Satoko Michiyama of Hiroshima University, Research Institute for Radiation Biology and Medicine, for excellent secretarial assistance.
Funding. This study was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan (1859081500 and 21590898).
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Author Contributions. M.K., N.F., and Y.H. drafted the manuscript and conceived the study. A.N., T.Mar., Y.I., A.I., T.Matsum., N.O., T.H., C.G., Y.A., and K.N. performed the ultrasonography. M.T. and T.Matsui measured the levels of AGEs and sRAGE. Y.K. and K.C. revised the article critically for important intellectual content. S.-i.Y. measured the levels of AGEs and sRAGE and revised the article critically for important intellectual content. Y.H. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.