CLINICAL INVESTIGATION
Retinol-Binding Protein 4 as a Novel Risk Factor for Cardiovascular Disease in Patients With Coronary Artery Disease and Hyperinsulinemia Fei Li, MD, Ke Xia, MD, Chuanchang Li, MD and Tianlun Yang, MD, PhD
Abstract: Background: Retinol-binding protein 4 (RBP4) is an adipokine associated with insulin resistance (IR) and obesity-related metabolic disorders. To evaluate its association with cardiovascular disease, we compared serum RBP4 concentrations in patients with coronary artery disease (CAD) and in controls. Methods: Sixty patients with coronary angiography-confirmed CAD, including 30 with CAD and 30 with CAD and hyperinsulinemia (CAD/HIns group), and 29 healthy subjects were enrolled. Serum RBP4, IR indexes and cardiovascular risk factors were assessed in these subjects. Results: Serum RBP4 concentrations were significantly higher in the CAD/HIns than in the CAD and control groups (P , 0.01 each). RBP4 concentration was significantly associated with cardiovascular risk factors, including body mass index and concentrations of triglycerides, high-density lipoprotein cholesterol, uric acid, highsensitivity C reactive protein and adiponectin (P , 0.01 each). Two-hour postprandial insulin (b 5 0.224), homeostatic model assessment of IR score (b 5 0.456) and adiponectin concentration (b 5 0.294) were independent factors associated with RBP4 (P , 0.01 each). Conclusions: RBP4 concentration is associated with cardiovascular risk factors related to IR and CAD. Circulating RBP4 could be a marker of metabolic complications and atherosclerosis and could be used to assess CAD. Key Indexing Terms Coronary artery disease; Retinol-binding protein 4; Hyperinsulinemia. [Am J Med Sci 2014;348(6):474–479.]
A
dipose tissue is now viewed as an active endocrine organ, secreting many types of adipokines such as adiponectin (APN), resistin, leptin and tumor necrosis factor a.1–3 Retinol-binding protein 4 (RBP4), first identified in 2005 using gene chips and shown to be a retinol carrier protein in blood,4 is an adipokine primarily secreted by the liver and adipose tissue found to be associated with the development of insulin resistance (IR).5–9 Elevated serum RBP4 concentrations increase IR by inhibiting insulin receptor substrate-1 phosphorylation and phosphatidylinositol 3-kinase activation.4 According to the “common soil” hypothesis,10–13 IR may be the fundamental link connecting many metabolic abnormalities and14 atherosclerotic diseases, including ischemic heart disease, coronary artery disease (CAD) and arteriosclerosis obliterans. However, the relaFrom the Department of Cardiology (FL, KX, TY), Xiangya Hospital, Central South University, Changsha, China; Institute of Molecular Medicine and Surgery (MMK) ( KX), Karolinska Hospital, Stockholm, Sweden; Department of Geriatrics (CL), Xiangya Hospital, Central South University, Changsha, China; and Institute of Hypertension (TY), Central South University, Changsha, China. Submitted September 14, 2013; accepted in revised form June 18, 2014. Supported by Medical Research Council: Chinese Medical Association (Grant number: 08010008); Medical Research Council: Doctoral Fund of Ministry of Education of China (Grant number: 200805331178); Medical Research Council: Health Department, Hunan province, China (Grant number: B2011-010). The authors have no financial or other conflicts of interest to disclose. Correspondence: Tianlun Yang, MD, PhD, Department of Cardiology, Xiangya Hospital, Central South University, Changsha 410078, China (E-mail: [email protected]).
474
tionship of RBP4 to atherosclerosis has not yet been determined. Sun et al15 did a study among 468 women who developed coronary heart disease and 472 matched controls during 16 years of follow-up, then found that in this cohort of women, higher circulating full-length and total RBP4 levels were associated with increased risk of coronary heart disease in a timedependent fashion. Intima-media thickness (IMT) was shown to be directly associated with CAD risk factors.5,16 Assessments of flow-mediated dilatation, IMT and other clinical parameters in 50 subjects with T2DM found that serum RBP4 levels reflect endothelial dysfunction in adults newly diagnosed with T2DM.17 Moreover, several other studies showed that circulating RBP4 levels correlate with IMT,18–22 suggesting that measuring serum RBP4 may be a more useful marker of endothelial dysfunction than IMT. Moreover, RBP4 was found to be independent associated with small, dense, low-density lipoprotein (LDL) and with oxidized LDL,23,24 suggesting that RBP4 plays an important role in atherosclerotic disorders. In elderly individuals, RBP4 concentrations were associated with metabolic syndrome (MetS) and its components in both genders and with prior cerebrovascular disease (CVD) in men,25 suggesting that circulating RBP4 may be a marker of metabolic complications and possibly of atherosclerosis and CAD. In contrast, some other studies suggested that serum RBP4 is not predictive of CAD risk and is not a marker for MetS or IR in patients with CAD. Circulating RBP4 levels were reported to be higher in type 2 diabetic (T2DM) subjects with than without previous clinical atherosclerosis, suggesting that RBP4 may be a biomarker of CAD in T2DM subjects.26 In contrast, a study assessing whether circulating RBP4 was predictive of CVD in 1036 patients who developed CAD over a 6-year period found that CVD risk increased as RBP4 levels increased but that this association was lost after adjustment for other CVD risk factors.14 Moreover, no differences in RBP4 concentrations and RBP4-to-transthyrethin ratio were observed between nondiabetic patients with CAD and controls or between individuals with and without the MetS.27 The role of RBP4 in IR and CAD remains unclear. Classic CVD risk factors, including body mass index (BMI), waist-to-hip ratio, blood pressure (BP) and blood lipid and high sensitivity C-reactive protein (hsCRP) concentrations, represent only part of the risk of CVD. Associations between RBP4 and these classic CVD risk factors in patients with CAD and hyperinsulinemia have not yet been evaluated. To investigate the associations between RBP4 and CVD risk factors, serum RBP4 concentrations were compared in groups of patients with CAD and CAD plus hyperinsulinemia (CAD/HIns) and in healthy controls.
SUBJECTS AND METHODS Subjects Subjects were recruited while undergoing routine clinical procedures between August 2009 and May 2010 at the
The American Journal of the Medical Sciences
Volume 348, Number 6, December 2014
RBP4 is a Novel Cardiovascular Disease Risk Factor
XiangYa Hospital in Hunan Province in China. Three groups were studied: (1) 30 patients with CAD; (2) 30 CAD/HIns patients and (3) 29 healthy subjects (control group). CAD was defined as remote myocardial infarction or luminal diameter stenosis .50% in at least 1 coronary artery or major branch. Hyperinsulinemia was defined according to the criteria of the Chinese Diabetes Society as fasting insulin concentration $15 mU/L and insulin concentration after glucose load $80 mU/L. Subjects previously diagnosed with or treated for diabetes; those with malignant tumors, acute or chronic infection, or who experienced recent trauma or surgery; those classified as New York Heart Association level III or IV and subjects with liver or kidney dysfunction were excluded. The study protocol was approved by our local ethics committee, and all subjects provided written informed consent. Methods Anthropometric measurements were performed while subjects were lightly clothed and without shoes. Height was measured to the nearest 0.1 cm and weight was measured to the nearest 0.1 kg using an automatic height-weight scale. BMI was calculated as the weight (in kilograms) divided by the square of height (in meters). Blood samples were collected into chilled tubes containing EDTA-Na2 and centrifuged at 3000 rpm at 4°C for 15 minutes. Fasting plasma glucose (FPG), serum triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), LDL cholesterol, uric acid (UA), hsCRP and free fatty acid concentrations were measured using an autoanalyzer (Beckman CX-7 Biochemical Autoanalyzer; Beckman Coulter, Inc, Fullerton, CA). Serum concentrations of fasting insulin (FIns) and 2-hour postprandial insulin (2hPInS) were measured using a chemiluminescence method. Serum samples were stored frozen at 280°C until assayed. RBP4 and APN were measured by enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN). The intra- and inter-assay coefficients of variation of these tests were ,9% and ,11%, respectively. IR was estimated using homeostatic model assessment of IR (HOMA-IR), calculated as [fasting insulin 3 fasting glucose]/ 22.5.28 Beta cell function was estimated using HOMA-B, calculated as 20 3 FIns/(FPG 2 3.5)29; and insulin sensitivity was estimated using insulin sensitivity index (ISI), calculated as LN [1/(FPG 3 FIns)].30 Statistical Analysis Data are expressed as mean 6 standard deviations (SDs) or as medians and interquartile ranges. Non-normally distributed variables were analyzed using log-transformed values. Differences between pairs of groups were assessed using Student’s t-tests. Spearman’s rank coefficient analysis were performed to determine relationships between RBP4, APN and other variables. Partial correlations analyses were performed to determine associations between RBP4 levels and other factors when controlling some factors such as gender, age or BMI. The Stepwise multiple linear regression analyses was used to select statistically significant variables. P-values ,0.05 were considered significant. Data were analyzed using SPSS 13.0 for Windows (SPSS Inc, Chicago, IL).
RESULTS Participants’ Characteristics There were no significant differences among the CAD, CAD/HIns, and control groups in gender distribution, systolic Ó 2014 Lippincott Williams & Wilkins
BP, diastolic BP (DBP) or waist-to-hip ratio. Controls were significantly younger (P , 0.01). BMI, FIns, 2hPIns, HOMAIR, HOMA-B and ISI (P , 0.01 each), as well as UA and APN (P , 0.05 each) were significantly higher in the CAD/HIns group than in the other 2 groups (Table 1). Serum RBP4 Concentrations Serum RBP4 concentrations were higher in the combined CAD/HIns and CAD groups than in the control group (P , 0.01; Figure 1, Table 1). RBP4 concentrations were similar in the CAD and control groups but were significantly higher in the CAD/HIns than in either of the other 2 groups (P , 0.01). Associations Between Serum RBP4 Levels and Subjects’ Gender and Age Serum RBP4 concentrations were similar in men and women (22.71 6 11.44 mg/mL versus 22.53 6 8.75 mg/mL, P 5 nonsignificant). Age was positively associated with serum RBP4 (r 5 0.361, P 5 0.001), even after adjustment for gender and BMI (r 5 0.339, P 5 0.002). Relationship Between RBP4 and Cardiovascular Risk Factors Serum RBP4 concentrations were positively correlated with BMI, TG, UA and hsCRP concentrations and negatively correlated with HDL-C and APN concentrations (Table 2). HDL-C showed the strongest correlation with RBP4 (r 5 20.290; P , 0.01). Determinants of Serum RBP4 After adjustment for age and gender, serum RBP4 concentrations were modestly correlated with FPG (r 5 0.239), FIns (r 5 0.560), 2hPIns (r 5 0.422) and APN (r 5 20.296) concentrations and with HOMA-IR (r 5 20.567), HOMA-B (r 5 0.393) and ISI (r 5 20.567) scores (P , 0.05 each). Stepwise multiple linear regression analyses showed that 2hPIns (b 5 0.224) and APN (b 5 0.294) concentrations and HOMA-IR score (b 5 0.456) were independent determinants of serum RBP4 concentrations (P , 0.01 each). Clinical Characteristics Relative to Serum RBP4 Quartiles To assess the associations between serum RBP4 concentrations and other CAD risk factors, subjects were divided into serum RBP4 quartiles (Table 3). The proportion of patients with combined CAD/HIns increased as RBP4 levels increased, as did FPG, HDL-C, UA, hsCRP, FIns, 2hPIns, HOMA-IR and ISI.
DISCUSSION Previous studies have suggested that RBP4 concentration is significantly associated with IR,4 T2DM,4 atherosclerosis14,21,22,25,27 and renal dysfunction.26,31,32 The present study examined serum RBP4 levels in patients with CAD and CAD/HIns patients, who have not yet developed glucose metabolism disorders. We confirmed that serum RBP4 levels were elevated in CAD/HIns patients and that the proportion of patients with combined CAD/HIns increased as RBP4 levels increased. In addition, we found that serum RBP4 concentrations were significantly associated with IR, measured as HOMA-IR scores; with beta cell function, measured as HOMA-B score and with insulin sensitivity, measured using ISI score. We also found that HOMA-IR score was an independent determinant of serum RBP4, and that plasma insulin, not
475
Li et al
TABLE 1. Demographic and clinical characteristics of the study subjects Controls N (men/women) Age, yr BMI, kg/m2 SBP, mm Hg DBP, mm Hg WHR TG, mmol/La TC, mmol/L LDL-C, mmol/La HDL-C, mmol/L UA, mmol/L hsCRP, mg/L FFA, mmol/L FPG, mmol/L FIns, mU/La 2hPIns, mU/La HOMA-IRa HOMA-Ba ISI APN, mg/L RBP4, mg/mLa
30 (11/19) 60.9 6 6.2 22.05 6 1.82 138.3 6 17.6 82.4 6 11.0 0.85 6 0.06 0.84 (0.35–1.51)** 4.41 6 0.66 1.90 (1.12–2.37) 2.07 6 0.68** 274.82 6 64.32** 0.32 6 0.15** 0.54 6 0.14 4.73 6 0.42 4.70 (1.7–11.8)** 26.33 (9.80–50.02)** 0.98 (0.38–2.84)** 90.27 (21.66–219.05)** 23.15 6 0.52** 493.60 6 282.73** 16.50 (0.01–31.42)**
CAD
CAD/HIns
30 (19/11) 63.7 6 9.8 22.99 6 3.01 141.5 6 18.6 84.5 6 12.0 0.88 6 0.05 1.34 (0.52–5.10)* 4.32 6 1.36 2.42 (0.72–5.90)* 1.24 6 0.27* 335.04 6 78.51*,*** 3.45 6 3.89** 0.59 6 0.31 4.73 6 0.48 6.06 (1.57–11.70)** 27.19 (1.50–78.6)** 1.26 (0.37–2.28)** 104.2 (13.19–327.46)** 23.2 6 0.54** 225.57 6 108.02*,*** 17.56 (12.80–32.48)**
29 (17/12) 63.7 6 7.2* 24.95 6 3.04* 140.5 6 16.5 84.8 6 11.6 0.89 6 0.05 1.62 (0.44–5.88)* 4.06 6 1.01 2.03 (0.80–3.80) 1.21 6 0.25* 380.10 6 92.94* 6.01 6 7.59** 0.56 6 0.42 5.03 6 0.75 10.43 (5.31–37.40)* 90.8 (19.66–323.90)* 2.22 (1.03–8.43)* 150.00 (71.76–575.79)* 24.08 6 0.55* 271.51 6 93.07* 26.59 (16.49–54.59)*
Data are mean 6 standard deviation (if normally distributed) or median (interquartile range). a These variables were log-transformed. *P , 0.01 versus control group; **P , 0.01 versus CAD/HIns; ***P , 0.05 versus CAD/HIns. 2hPIns, 2-hr postprandial insulin; APN, adiponectin; BMI, body mass index; CAD, coronary artery disease; DBP, diastolic blood pressure; FFA, free fatty acid; FIns, fasting insulin; FPG, fasting plasma glucose; HDL-C, high-density lipoprotein cholesterol; HOMA-B, homeostasis model assessment-B; HOMA-IR, homeostatic model assessment of insulin resistance; hsCRP, high-sensitivity C reactive protein; ISI, insulin sensitivity index; LDL-C, low-density lipoprotein cholesterol; RBP4, retinol-binding protein; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides; UA, uric acid; WHR, waist-to-hip ratio.
plasma glucose, concentrations were independent determinants of circulating RBP4 concentrations. These findings were consistent in showing strong correlations between serum RBP4 and insulin levels, as well as with markers of IR, in agreement with recent results.4,31 We also found that serum RBP4 concentrations are elevated in patients with hyperinsulinemia without glucose metabolism abnormalities and are associated with various clinical parameters associated with IR and CAD. Taken together, these findings indicate that serum RBP4 is an early marker of IR.
FIGURE 1. Serum RBP4 concentrations in CAD and CAD/HIns patients and in controls. Serum RBP4 concentrations were significantly higher in the combined CAD/HIns than the other 2 groups (*P , 0.01 versus Control). CAD, coronary artery disease; RBP4, retinol-binding protein.
476
We also found that traditional CVD risk factors, including BMI and blood lipid (HDL-C, TG), UA and hsCRP concentrations, were significantly associated with serum RBP4 levels, in agreement with previous findings.4 Metabolic lipoprotein abnormalities are important CVD risk factors. During longterm hyperlipidemia, oxidized LDL and TC damage the intima of the arteries, ultimately inducing fibrous plaques within the coronary artery, provoking vascular stenosis or obstruction and leading to myocardial ischemia or necrosis. We have shown here that the concentrations of some lipids are associated with RBP4 concentrations. As we previously suggested,32 retinoids increase the expression of liver apolipoprotein CIII (Apo C-III), an agonist of lipoprotein lipase, through the RAR/RXR pathway. RBP4 may inhibit the catabolism of very low-density lipoprotein (VLDL), increasing VLDL levels and decreasing HDL-C. To maintain VLDL catabolism, cholesterol esterification and transfer protein pathways are activated, increasing TC and TG transport and increasing the concentrations of both in the blood. We also found that UA and RBP4 concentrations were correlated. Hyperuricemia is an independent risk factor for CVD, and being closely related to CAD severity.33 Urate crystals deposit in the walls of the arteries, damaging endothelial cells in the arteries and inducing the formation of oxygen-free radicals. In addition, urate crystals can activate platelets, promoting blood platelet gathering and adhesion and enhancing the likelihood of atherosclerosis. Furthermore, we found that hsCRP and RBP4 concentrations were positively correlated, in agreement with results showing that serum RBP4 levels Volume 348, Number 6, December 2014
RBP4 is a Novel Cardiovascular Disease Risk Factor
TABLE 2. Correlations between RBP4 concentrations and other clinical factors in the included subjects FRG, mmol/L FIns, mU/L 2hPIns, mU/L HOMA-IR HOMA-B BMI, kg/m2 r P
r P
0.259 ,0.05
0.331 ,0.01
0.582 ,0.01
0.452 ,0.01
0.600 ,0.01
ISI 20.600 ,0.01
0.338 ,0.01
TG, mmol/La
TC, mmol/L
LDL-C, mmol/L
HDL-C, mmol/L
UA, mmol/L
hsCRP, mg/L
FFA, mmol/L
APN, mg/L
0.236 ,0.05
0.117 .0.05
0.011 .0.05
20.290 ,0.01
0.243 ,0.05
0.231 ,0.05
0.076 .0.05
20.011 ,0.05
Coefficients (r) and P values were calculated by the Spearman’s rank coefficient test. a These variables were log transformed before analyses. 2hPIns, 2-hour postprandial insulin; APN, adiponectin; FFA, free fatty acid; FIns, fasting insulin; FPG, fasting plasma glucose; HDL-C, highdensity lipoprotein cholesterol; HOMA-B, homeostasis model assessment-B; HOMA-IR, homeostatic model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein; ISI, insulin sensitivity index; LDL-C, low-density lipoprotein cholesterol; RBP4, retinol-binding protein; TC, total cholesterol; TG, triglycerides; UA, uric acid.
are associated with pro-inflammatory factors, not with IR, in patients with T2DM and CAD.5,34 Taken together, our findings suggest that RBP4 might be a useful marker of CVD risk. We therefore hypothesized that RBP4 would be associated with the prevalence of CAD. We found that serum APN concentrations were significantly reduced in subjects with CAD and negatively associated with RBP4 concentrations. APN, the only adipokine with protective effects in humans, has been reported to have antiinflammatory, anti-IR and anti-atherogenic properties.35,36 APN can modulate the biological actions of several growth factors by controlling their bioavailability at a pre-receptor
level.37 APN rapidly accumulates in the subendothelial space of an injured artery38 and inhibits atherogenic processes.38–42 Thus, our findings are consistent with the hypothesis that circulating RBP4 may be a marker of atherosclerosis and CAD. This study had several limitations, including the relatively small sample size and the use of subjects from a single geographic area. Although we found that RBP4 concentrations differed in patients with CAD and controls, the difference was not statistically significant, suggesting the need to measure these serum proteins in larger patient populations. Moreover, our findings, suggesting correlations between serum RBP4
TABLE 3. Subject characteristics according to RBP4 quartiles First quartile Second quartile Number RBP4, mg/mL Control CAD CAD + HIns Age, yr BMI, kg/m2 FPG, mmol/L FIns, mU/La 2hPIns, mU/La TG, mmol/La TC, mmol/L LDL-C, mmol/La HDL, mmol/L UA, mmol/L hsCRP, mg/L FFA, mmol/L HOMA-IRa HOMA-Ba ISI APN, mg/L
22 14.60 (10.01–16.05) 13 9 0 51 6 10 22.71 6 2.54 4.55 6 0.35 4.60 (1.70–11.70) 28.39 (9.80–71.39) 0.99 (0.35–4.74) 4.38 6 0.68 1.90 (0.85–3.28) 1.79 6 0.65 276.01 6 63.89 0.52 6 0.52 0.56 6 0.39 0.95 (0.38–2.28) 95.63 (21.66–296.674.74) 23.03 6 0.53 338.57 6 241.47
23 17.20 (16.33–19.17) 9 13 1 57 6 13 22.79 6 2.59 4.74 6 0.44 5.80 (1.80–11.00) 26.51 (1.50–95.67) 1.13 (0.46–5.88) 4.54 6 1.33 2.21 (1.08–5.90) 1.62 6 0.72 332.74 6 75.05 3.05 6 0.90 0.58 6 0.29 1.12 (0.37–2.18) 80.96 (31.77–333.33) 23.21 6 0.47 352.78 6 257.16
Third quartile
Fourth quartile
22 22.51 (19.59–24.91) 3 6 13 62 6 10 23.36 6 2.39 5.09 6 0.77 9.44 (1.57–18.23) 62.39 (11.36–212.20) 1.38 (0.44–5.10) 4.05 6 1.16 2.08 (0.72–3.68) 1.31 6 0.43 362.10 6 100.17 6.12 6 2.40 0.48 6 0.31 2.03 (0.41–4.69) 132.30 (13.19–575.79) 23.72 6 0.66 251.61 6 160.25
22 32.11 (25.30–54.59) 5 2 15 61 6 11 24.46 6 3.79 4.93 6 0.54 9.65 (3.70–37.40) 83.53 (12.49–323.90) 1.49 (0.49–5.06) 4.09 6 0.88 1.97 (1.01–3.06) 1.30 6 0.41 346.70 6 95.63 3.23 6 1.40 0.64 6 0.39 1.86 (0.84–8.43) 145.63 (46.38–476.43) 23.95 6 0.63 379.58 6 183.47
P ,0.001
0.004 0.17 0.01 ,0.001 ,0.001 0.53 0.35 0.06 0.01 0.007 0.006 0.50 ,0.001 0.03 ,0.001 0.23
Data are presented as mean 6 standard deviation (if normally distributed) or median (interquartile range). a These variables were log transformed before analyses. APN, adiponectin; FFA, free fatty acid; FIns, fasting insulin; 2hPIns, 2-hour postprandial insulin; FPG, fasting plasma glucose; HDL-C, highdensity lipoprotein cholesterol; HOMA-B, homeostasis model assessment-B; HOMA-IR, homeostatic model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein; ISI, insulin sensitivity index; LDL-C, low-density lipoprotein cholesterol; RBP4, retinol-binding protein; TC, total cholesterol; TG, triglycerides; UA, uric acid.
Ó 2014 Lippincott Williams & Wilkins
477
Li et al
levels and CAD development, require confirmation in a larger, prospectively recruited patient sample.
17. Oren A, Vos LE, Uiterwaal CS, et al. Cardiovascular risk factors and increased carotid intima-media thickness in healthy young adults: the Atherosclerosis Risk in Young Adults (ARYA) Study. Arch Intern Med 2003;163:1787–92.
CONCLUSIONS
18. Furumoto T, Fujii S, Saito N, et al. Relationships between brachial artery flow mediated dilation and carotid artery intima-media thickness in patients with suspected coronary artery disease. Jpn Heart J 2002;43: 117–25.
Serum RBP4 levels were associated with IR and CVD risk factors in both men and women. These findings agree with the hypothesis that circulating RBP4 may be a marker of the metabolic syndrome, insulin resistant and CAD. Further studies are required to confirm these associations. REFERENCES 1. Fu Y, Luo N, Klein RL, et al. Adiponectin promotes adipocyte differentiation, insulin sensitivity, and lipid accumulation. J Lipid Res 2005; 46:1369–79. 2. Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 2004;89:2548–56. 3. Trayhurn P, Wood IS. Adipokines: inflammation and the pleiotropic role of white adipose tissue. Br J Nutr 2004;92:347–55. 4. Yang Q, Graham TE, Mody N, et al. Serum retinol binding protein 4 contributes to insulin resistance in obesity and type 2 diabetes. Nature 2005;436:356–62. 5. Cho YM, Youn B-S, Lee H, et al. Plasma retinol-binding protein-4 concentrations are elevated in human subjects with impaired glucose tolerance and type 2 diabetes. Diabetes Care 2006;29:2457–61.
19. Park SE, Kim DH, Lee JH, et al. Retinol-binding protein-4 is associated with endothelial dysfunction in adults with newly diagnosed type 2 diabetes mellitus. Atherosclerosis 2009;204:23–5. 20. Xiao Y, Xu A, Hui X, et al. Circulating lipocalin-2 and retinol-binding protein 4 are associated with intima-media thickness and subclinical atherosclerosis in patients with type 2 diabetes. PLoS One 2013;8: e66607. 21. Bobbert T, Raila J, Schwarz F, et al. Relation between retinol, retinolbinding protein 4, transthyretin and carotid intima media thickness. Atherosclerosis 2010;213:549–51. 22. Ingelsson E, Lind L. Circulating retinol-binding protein 4 and subclinical cardiovascular disease in the elderly. Diabetes Care 2009;32: 733–5. 23. Wu J, Shi YH, Niu DM, et al. Association among retinol-binding protein 4, small dense LDL cholesterol and oxidized LDL levels in dyslipidemia subjects. Clin Biochem 2012;45:619–22.
6. Graham TE, Yang Q, Bluher M, et al. Retinol-binding protein 4 and insulin resistance in lean, obese, and diabetic subjects. N Engl J Med 2006;354:2552–63.
24. von Eynatten M, Lepper PM, Liu D, et al. Retinol-binding protein 4 is associated with components of the metabolic syndrome, but not with insulin resistance, in men with type 2 diabetes or coronary artery disease. Diabetologia 2007;50:1930–7.
7. Kloeting N, Graham TE, Berndt J, et al. Serum retinol-binding protein is more highly expressed in visceral than in subcutaneous adipose tissue and is a marker of intra-abdominal fat mass. Cell Metab 2007;6:79–87.
25. Ingelsson E, Sundström J, Melhus H, et al. Circulating retinolbinding protein 4, cardiovascular risk factors and prevalent cardiovascular disease in elderly. Atherosclerosis 2009;206:239–44.
8. Ribel-Madsen R, Friedrichsen M, Vaag A, et al. Retinol-binding protein 4 in twins regulatory mechanisms and impact of circulating and tissue expression levels on insulin secretion and action. Diabetes 2009;58:54–60.
26. Cabré A, Lázaro I, Girona J, et al. Retinol-binding protein 4 as a plasma biomarker of renal dysfunction and cardiovascular disease in type 2 diabetes. J Intern Med 2007;262:496–503.
9. Stefan N, Schick F, Hennige AM, et al. High circulating retinolbinding protein 4 is associated with elevated liver fat but not with total, subcutaneous, visceral, or intramyocellular fat in humans. Diabetes Care 2007;30:1173–8.
27. Mahmoudi MJ, Mahmoudi M, Siassi F, et al. Circulating retinolbinding protein 4 concentrations in patients with coronary artery disease and patients with type 2 diabetes mellitus. Int J Diabetes Dev Ctries 2012;32:105–10.
10. Stern MP. Diabetes and cardiovascular disease. The “common soil” hypothesis. Diabetes 1995;44:369–74.
28. Radziuk J. Insulin sensitivity and its measurement: structural commonalities among the methods. J Clin Endocrinol Metab 2000;85: 4426–33.
11. Matsumoto K, Miyake S, Yano M, et al. Insulin resistance and arteriosclerosis obliterans in patients with NIDDM. Diabetes Care 1997;20: 1738–43.
29. Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191–2.
12. Hanley AJ, Williams K, Stern MP, et al. Homeostasis model assessment of insulin resistance in relation to the incidence of cardiovascular disease: the San Antonio Heart Study. Diabetes Care 2002;25:1177–84.
30. Belfiore F, Iannello S, Volpicelli G. Insulin sensitivity indices calculated from basal and OGTT-induced insulin, glucose, and FFA levels. Mol Genet Metab 1998;63:134–41.
13. Bonora E, Formentini G, Calcaterra F, et al. HOMA-estimated insulin resistance is an independent predictor of cardiovascular disease in type 2 diabetic subjects: prospective data from the Verona Diabetes Complications Study. Diabetes Care 2002;25:1135–41.
31. Akbay E, Muslu N, Nayir E, et al. Serum retinol binding protein 4 level is related with renal functions in type 2 diabetes. J Endocrinol Invest 2010;33:725–9.
14. Mallat Z, Simon T, Benessiano J, et al. Retinol-binding protein 4 and prediction of incident coronary events in healthy men and women. J Clin Endocrinol Metab 2008;94:255–60. 15. Sun Q, Kiernan UA, Shi L, et al. Plasma retinol-binding protein 4 (RBP4) levels and risk of coronary heart disease: a prospective analysis among women in the nurses’ health study. Circulation 2013;127:1938–47. 16. Vu-Dac N, Gervois P, Torra IP, et al. Retinoids increase human apo C-III expression at the transcriptional level via the retinoid X receptor. Contribution to the hypertriglyceridemic action of retinoids. J Clin Invest 1998;102:625–32.
478
32. Thawnashom K, Tungtrongchitr R, Chanchay S, et al. Association between retinol-binding protein and renal function among Asian subjects with type 2 diabetes mellitus: a cross-sectional study. Southeast Asian J Trop Med Public Health 2011;42:936–45. 33. Polidori MC, Pratico D, Parente B, et al. Elevated lipid peroxidation biomarkers and low antioxidant status in atherosclerotic patients with increased carotid or iliofemoral intima media thickness. J Invest Med 2007;55:163–7. 34. Gey KF, Ducimetiere P, Evans A, et al. Low plasma retinol predicts coronary events in healthy middle-aged men: the PRIME Study. Atherosclerosis 2010;208:270–4.
Volume 348, Number 6, December 2014
RBP4 is a Novel Cardiovascular Disease Risk Factor
35. Al-Daghri NM, Al-Attas OS, Alokail M, et al. Retinol binding protein-4 is associated with TNF-alpha and not insulin resistance in subjects with type 2 diabetes mellitus and coronary heart disease. Dis Markers 2009;26:135–40.
39. Arita Y, Kihara S, Ouchi N, et al. Adipocyte-derived plasma protein adiponectin acts as a platelet-derived growth factor-BB-binding protein and regulates growth factor-induced common postreceptor signal in vascular smooth muscle cell. Circulation 2002;105:2893–8.
36. Lu P, Hu D, Lu J, et al. The association between uric acid and coronary heart disease [in Chinese]. Zhonghua Nei Ke Za Zhi 2002;41:526–9.
40. Okamoto Y, Arita Y, Nishida M, et al. An adipocyte-derived plasma protein, adiponectin, adheres to injured vascular walls. Horm Metab Res 2000;32:47–50.
37. Ouchi N, Kihara S, Arita Y, et al. Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-kappaB signaling through a cAMP-dependent pathway. Circulation 2000;102:1296–301. 38. Ouchi N, Kihara S, Arita Y, et al. Adipocyte-derived plasma protein, adiponectin, suppresses lipid accumulation and class A scavenger receptor expression in human monocyte-derived macrophages. Circulation 2001;103:1057–63.
Ó 2014 Lippincott Williams & Wilkins
41. Ouchi N, Kihara S, Arita Y, et al. Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation 1999;100:2473–6. 42. Wang Y, Lam KS, Xu JY, et al. Adiponectin inhibits cell proliferation by interacting with several growth factors in an oligomerizationdependent manner. J Biol Chem 2005;280:18341–7.
479
Retinol-Binding Protein 4 as a Novel Risk Factor for Cardiovascular Disease in Patients With Coronary Artery Disease and Hyperinsulinemia Fei Li, MD, Ke Xia, MD, Chuanchang Li, MD and Tianlun Yang, MD, PhD
Abstract: Background: Retinol-binding protein 4 (RBP4) is an adipokine associated with insulin resistance (IR) and obesity-related metabolic disorders. To evaluate its association with cardiovascular disease, we compared serum RBP4 concentrations in patients with coronary artery disease (CAD) and in controls. Methods: Sixty patients with coronary angiography-confirmed CAD, including 30 with CAD and 30 with CAD and hyperinsulinemia (CAD/HIns group), and 29 healthy subjects were enrolled. Serum RBP4, IR indexes and cardiovascular risk factors were assessed in these subjects. Results: Serum RBP4 concentrations were significantly higher in the CAD/HIns than in the CAD and control groups (P , 0.01 each). RBP4 concentration was significantly associated with cardiovascular risk factors, including body mass index and concentrations of triglycerides, high-density lipoprotein cholesterol, uric acid, highsensitivity C reactive protein and adiponectin (P , 0.01 each). Two-hour postprandial insulin (b 5 0.224), homeostatic model assessment of IR score (b 5 0.456) and adiponectin concentration (b 5 0.294) were independent factors associated with RBP4 (P , 0.01 each). Conclusions: RBP4 concentration is associated with cardiovascular risk factors related to IR and CAD. Circulating RBP4 could be a marker of metabolic complications and atherosclerosis and could be used to assess CAD. Key Indexing Terms Coronary artery disease; Retinol-binding protein 4; Hyperinsulinemia. [Am J Med Sci 2014;348(6):474–479.]
A
dipose tissue is now viewed as an active endocrine organ, secreting many types of adipokines such as adiponectin (APN), resistin, leptin and tumor necrosis factor a.1–3 Retinol-binding protein 4 (RBP4), first identified in 2005 using gene chips and shown to be a retinol carrier protein in blood,4 is an adipokine primarily secreted by the liver and adipose tissue found to be associated with the development of insulin resistance (IR).5–9 Elevated serum RBP4 concentrations increase IR by inhibiting insulin receptor substrate-1 phosphorylation and phosphatidylinositol 3-kinase activation.4 According to the “common soil” hypothesis,10–13 IR may be the fundamental link connecting many metabolic abnormalities and14 atherosclerotic diseases, including ischemic heart disease, coronary artery disease (CAD) and arteriosclerosis obliterans. However, the relaFrom the Department of Cardiology (FL, KX, TY), Xiangya Hospital, Central South University, Changsha, China; Institute of Molecular Medicine and Surgery (MMK) ( KX), Karolinska Hospital, Stockholm, Sweden; Department of Geriatrics (CL), Xiangya Hospital, Central South University, Changsha, China; and Institute of Hypertension (TY), Central South University, Changsha, China. Submitted September 14, 2013; accepted in revised form June 18, 2014. Supported by Medical Research Council: Chinese Medical Association (Grant number: 08010008); Medical Research Council: Doctoral Fund of Ministry of Education of China (Grant number: 200805331178); Medical Research Council: Health Department, Hunan province, China (Grant number: B2011-010). The authors have no financial or other conflicts of interest to disclose. Correspondence: Tianlun Yang, MD, PhD, Department of Cardiology, Xiangya Hospital, Central South University, Changsha 410078, China (E-mail: [email protected]).
474
tionship of RBP4 to atherosclerosis has not yet been determined. Sun et al15 did a study among 468 women who developed coronary heart disease and 472 matched controls during 16 years of follow-up, then found that in this cohort of women, higher circulating full-length and total RBP4 levels were associated with increased risk of coronary heart disease in a timedependent fashion. Intima-media thickness (IMT) was shown to be directly associated with CAD risk factors.5,16 Assessments of flow-mediated dilatation, IMT and other clinical parameters in 50 subjects with T2DM found that serum RBP4 levels reflect endothelial dysfunction in adults newly diagnosed with T2DM.17 Moreover, several other studies showed that circulating RBP4 levels correlate with IMT,18–22 suggesting that measuring serum RBP4 may be a more useful marker of endothelial dysfunction than IMT. Moreover, RBP4 was found to be independent associated with small, dense, low-density lipoprotein (LDL) and with oxidized LDL,23,24 suggesting that RBP4 plays an important role in atherosclerotic disorders. In elderly individuals, RBP4 concentrations were associated with metabolic syndrome (MetS) and its components in both genders and with prior cerebrovascular disease (CVD) in men,25 suggesting that circulating RBP4 may be a marker of metabolic complications and possibly of atherosclerosis and CAD. In contrast, some other studies suggested that serum RBP4 is not predictive of CAD risk and is not a marker for MetS or IR in patients with CAD. Circulating RBP4 levels were reported to be higher in type 2 diabetic (T2DM) subjects with than without previous clinical atherosclerosis, suggesting that RBP4 may be a biomarker of CAD in T2DM subjects.26 In contrast, a study assessing whether circulating RBP4 was predictive of CVD in 1036 patients who developed CAD over a 6-year period found that CVD risk increased as RBP4 levels increased but that this association was lost after adjustment for other CVD risk factors.14 Moreover, no differences in RBP4 concentrations and RBP4-to-transthyrethin ratio were observed between nondiabetic patients with CAD and controls or between individuals with and without the MetS.27 The role of RBP4 in IR and CAD remains unclear. Classic CVD risk factors, including body mass index (BMI), waist-to-hip ratio, blood pressure (BP) and blood lipid and high sensitivity C-reactive protein (hsCRP) concentrations, represent only part of the risk of CVD. Associations between RBP4 and these classic CVD risk factors in patients with CAD and hyperinsulinemia have not yet been evaluated. To investigate the associations between RBP4 and CVD risk factors, serum RBP4 concentrations were compared in groups of patients with CAD and CAD plus hyperinsulinemia (CAD/HIns) and in healthy controls.
SUBJECTS AND METHODS Subjects Subjects were recruited while undergoing routine clinical procedures between August 2009 and May 2010 at the
The American Journal of the Medical Sciences
Volume 348, Number 6, December 2014
RBP4 is a Novel Cardiovascular Disease Risk Factor
XiangYa Hospital in Hunan Province in China. Three groups were studied: (1) 30 patients with CAD; (2) 30 CAD/HIns patients and (3) 29 healthy subjects (control group). CAD was defined as remote myocardial infarction or luminal diameter stenosis .50% in at least 1 coronary artery or major branch. Hyperinsulinemia was defined according to the criteria of the Chinese Diabetes Society as fasting insulin concentration $15 mU/L and insulin concentration after glucose load $80 mU/L. Subjects previously diagnosed with or treated for diabetes; those with malignant tumors, acute or chronic infection, or who experienced recent trauma or surgery; those classified as New York Heart Association level III or IV and subjects with liver or kidney dysfunction were excluded. The study protocol was approved by our local ethics committee, and all subjects provided written informed consent. Methods Anthropometric measurements were performed while subjects were lightly clothed and without shoes. Height was measured to the nearest 0.1 cm and weight was measured to the nearest 0.1 kg using an automatic height-weight scale. BMI was calculated as the weight (in kilograms) divided by the square of height (in meters). Blood samples were collected into chilled tubes containing EDTA-Na2 and centrifuged at 3000 rpm at 4°C for 15 minutes. Fasting plasma glucose (FPG), serum triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), LDL cholesterol, uric acid (UA), hsCRP and free fatty acid concentrations were measured using an autoanalyzer (Beckman CX-7 Biochemical Autoanalyzer; Beckman Coulter, Inc, Fullerton, CA). Serum concentrations of fasting insulin (FIns) and 2-hour postprandial insulin (2hPInS) were measured using a chemiluminescence method. Serum samples were stored frozen at 280°C until assayed. RBP4 and APN were measured by enzyme-linked immunosorbent assay (R&D Systems, Minneapolis, MN). The intra- and inter-assay coefficients of variation of these tests were ,9% and ,11%, respectively. IR was estimated using homeostatic model assessment of IR (HOMA-IR), calculated as [fasting insulin 3 fasting glucose]/ 22.5.28 Beta cell function was estimated using HOMA-B, calculated as 20 3 FIns/(FPG 2 3.5)29; and insulin sensitivity was estimated using insulin sensitivity index (ISI), calculated as LN [1/(FPG 3 FIns)].30 Statistical Analysis Data are expressed as mean 6 standard deviations (SDs) or as medians and interquartile ranges. Non-normally distributed variables were analyzed using log-transformed values. Differences between pairs of groups were assessed using Student’s t-tests. Spearman’s rank coefficient analysis were performed to determine relationships between RBP4, APN and other variables. Partial correlations analyses were performed to determine associations between RBP4 levels and other factors when controlling some factors such as gender, age or BMI. The Stepwise multiple linear regression analyses was used to select statistically significant variables. P-values ,0.05 were considered significant. Data were analyzed using SPSS 13.0 for Windows (SPSS Inc, Chicago, IL).
RESULTS Participants’ Characteristics There were no significant differences among the CAD, CAD/HIns, and control groups in gender distribution, systolic Ó 2014 Lippincott Williams & Wilkins
BP, diastolic BP (DBP) or waist-to-hip ratio. Controls were significantly younger (P , 0.01). BMI, FIns, 2hPIns, HOMAIR, HOMA-B and ISI (P , 0.01 each), as well as UA and APN (P , 0.05 each) were significantly higher in the CAD/HIns group than in the other 2 groups (Table 1). Serum RBP4 Concentrations Serum RBP4 concentrations were higher in the combined CAD/HIns and CAD groups than in the control group (P , 0.01; Figure 1, Table 1). RBP4 concentrations were similar in the CAD and control groups but were significantly higher in the CAD/HIns than in either of the other 2 groups (P , 0.01). Associations Between Serum RBP4 Levels and Subjects’ Gender and Age Serum RBP4 concentrations were similar in men and women (22.71 6 11.44 mg/mL versus 22.53 6 8.75 mg/mL, P 5 nonsignificant). Age was positively associated with serum RBP4 (r 5 0.361, P 5 0.001), even after adjustment for gender and BMI (r 5 0.339, P 5 0.002). Relationship Between RBP4 and Cardiovascular Risk Factors Serum RBP4 concentrations were positively correlated with BMI, TG, UA and hsCRP concentrations and negatively correlated with HDL-C and APN concentrations (Table 2). HDL-C showed the strongest correlation with RBP4 (r 5 20.290; P , 0.01). Determinants of Serum RBP4 After adjustment for age and gender, serum RBP4 concentrations were modestly correlated with FPG (r 5 0.239), FIns (r 5 0.560), 2hPIns (r 5 0.422) and APN (r 5 20.296) concentrations and with HOMA-IR (r 5 20.567), HOMA-B (r 5 0.393) and ISI (r 5 20.567) scores (P , 0.05 each). Stepwise multiple linear regression analyses showed that 2hPIns (b 5 0.224) and APN (b 5 0.294) concentrations and HOMA-IR score (b 5 0.456) were independent determinants of serum RBP4 concentrations (P , 0.01 each). Clinical Characteristics Relative to Serum RBP4 Quartiles To assess the associations between serum RBP4 concentrations and other CAD risk factors, subjects were divided into serum RBP4 quartiles (Table 3). The proportion of patients with combined CAD/HIns increased as RBP4 levels increased, as did FPG, HDL-C, UA, hsCRP, FIns, 2hPIns, HOMA-IR and ISI.
DISCUSSION Previous studies have suggested that RBP4 concentration is significantly associated with IR,4 T2DM,4 atherosclerosis14,21,22,25,27 and renal dysfunction.26,31,32 The present study examined serum RBP4 levels in patients with CAD and CAD/HIns patients, who have not yet developed glucose metabolism disorders. We confirmed that serum RBP4 levels were elevated in CAD/HIns patients and that the proportion of patients with combined CAD/HIns increased as RBP4 levels increased. In addition, we found that serum RBP4 concentrations were significantly associated with IR, measured as HOMA-IR scores; with beta cell function, measured as HOMA-B score and with insulin sensitivity, measured using ISI score. We also found that HOMA-IR score was an independent determinant of serum RBP4, and that plasma insulin, not
475
Li et al
TABLE 1. Demographic and clinical characteristics of the study subjects Controls N (men/women) Age, yr BMI, kg/m2 SBP, mm Hg DBP, mm Hg WHR TG, mmol/La TC, mmol/L LDL-C, mmol/La HDL-C, mmol/L UA, mmol/L hsCRP, mg/L FFA, mmol/L FPG, mmol/L FIns, mU/La 2hPIns, mU/La HOMA-IRa HOMA-Ba ISI APN, mg/L RBP4, mg/mLa
30 (11/19) 60.9 6 6.2 22.05 6 1.82 138.3 6 17.6 82.4 6 11.0 0.85 6 0.06 0.84 (0.35–1.51)** 4.41 6 0.66 1.90 (1.12–2.37) 2.07 6 0.68** 274.82 6 64.32** 0.32 6 0.15** 0.54 6 0.14 4.73 6 0.42 4.70 (1.7–11.8)** 26.33 (9.80–50.02)** 0.98 (0.38–2.84)** 90.27 (21.66–219.05)** 23.15 6 0.52** 493.60 6 282.73** 16.50 (0.01–31.42)**
CAD
CAD/HIns
30 (19/11) 63.7 6 9.8 22.99 6 3.01 141.5 6 18.6 84.5 6 12.0 0.88 6 0.05 1.34 (0.52–5.10)* 4.32 6 1.36 2.42 (0.72–5.90)* 1.24 6 0.27* 335.04 6 78.51*,*** 3.45 6 3.89** 0.59 6 0.31 4.73 6 0.48 6.06 (1.57–11.70)** 27.19 (1.50–78.6)** 1.26 (0.37–2.28)** 104.2 (13.19–327.46)** 23.2 6 0.54** 225.57 6 108.02*,*** 17.56 (12.80–32.48)**
29 (17/12) 63.7 6 7.2* 24.95 6 3.04* 140.5 6 16.5 84.8 6 11.6 0.89 6 0.05 1.62 (0.44–5.88)* 4.06 6 1.01 2.03 (0.80–3.80) 1.21 6 0.25* 380.10 6 92.94* 6.01 6 7.59** 0.56 6 0.42 5.03 6 0.75 10.43 (5.31–37.40)* 90.8 (19.66–323.90)* 2.22 (1.03–8.43)* 150.00 (71.76–575.79)* 24.08 6 0.55* 271.51 6 93.07* 26.59 (16.49–54.59)*
Data are mean 6 standard deviation (if normally distributed) or median (interquartile range). a These variables were log-transformed. *P , 0.01 versus control group; **P , 0.01 versus CAD/HIns; ***P , 0.05 versus CAD/HIns. 2hPIns, 2-hr postprandial insulin; APN, adiponectin; BMI, body mass index; CAD, coronary artery disease; DBP, diastolic blood pressure; FFA, free fatty acid; FIns, fasting insulin; FPG, fasting plasma glucose; HDL-C, high-density lipoprotein cholesterol; HOMA-B, homeostasis model assessment-B; HOMA-IR, homeostatic model assessment of insulin resistance; hsCRP, high-sensitivity C reactive protein; ISI, insulin sensitivity index; LDL-C, low-density lipoprotein cholesterol; RBP4, retinol-binding protein; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides; UA, uric acid; WHR, waist-to-hip ratio.
plasma glucose, concentrations were independent determinants of circulating RBP4 concentrations. These findings were consistent in showing strong correlations between serum RBP4 and insulin levels, as well as with markers of IR, in agreement with recent results.4,31 We also found that serum RBP4 concentrations are elevated in patients with hyperinsulinemia without glucose metabolism abnormalities and are associated with various clinical parameters associated with IR and CAD. Taken together, these findings indicate that serum RBP4 is an early marker of IR.
FIGURE 1. Serum RBP4 concentrations in CAD and CAD/HIns patients and in controls. Serum RBP4 concentrations were significantly higher in the combined CAD/HIns than the other 2 groups (*P , 0.01 versus Control). CAD, coronary artery disease; RBP4, retinol-binding protein.
476
We also found that traditional CVD risk factors, including BMI and blood lipid (HDL-C, TG), UA and hsCRP concentrations, were significantly associated with serum RBP4 levels, in agreement with previous findings.4 Metabolic lipoprotein abnormalities are important CVD risk factors. During longterm hyperlipidemia, oxidized LDL and TC damage the intima of the arteries, ultimately inducing fibrous plaques within the coronary artery, provoking vascular stenosis or obstruction and leading to myocardial ischemia or necrosis. We have shown here that the concentrations of some lipids are associated with RBP4 concentrations. As we previously suggested,32 retinoids increase the expression of liver apolipoprotein CIII (Apo C-III), an agonist of lipoprotein lipase, through the RAR/RXR pathway. RBP4 may inhibit the catabolism of very low-density lipoprotein (VLDL), increasing VLDL levels and decreasing HDL-C. To maintain VLDL catabolism, cholesterol esterification and transfer protein pathways are activated, increasing TC and TG transport and increasing the concentrations of both in the blood. We also found that UA and RBP4 concentrations were correlated. Hyperuricemia is an independent risk factor for CVD, and being closely related to CAD severity.33 Urate crystals deposit in the walls of the arteries, damaging endothelial cells in the arteries and inducing the formation of oxygen-free radicals. In addition, urate crystals can activate platelets, promoting blood platelet gathering and adhesion and enhancing the likelihood of atherosclerosis. Furthermore, we found that hsCRP and RBP4 concentrations were positively correlated, in agreement with results showing that serum RBP4 levels Volume 348, Number 6, December 2014
RBP4 is a Novel Cardiovascular Disease Risk Factor
TABLE 2. Correlations between RBP4 concentrations and other clinical factors in the included subjects FRG, mmol/L FIns, mU/L 2hPIns, mU/L HOMA-IR HOMA-B BMI, kg/m2 r P
r P
0.259 ,0.05
0.331 ,0.01
0.582 ,0.01
0.452 ,0.01
0.600 ,0.01
ISI 20.600 ,0.01
0.338 ,0.01
TG, mmol/La
TC, mmol/L
LDL-C, mmol/L
HDL-C, mmol/L
UA, mmol/L
hsCRP, mg/L
FFA, mmol/L
APN, mg/L
0.236 ,0.05
0.117 .0.05
0.011 .0.05
20.290 ,0.01
0.243 ,0.05
0.231 ,0.05
0.076 .0.05
20.011 ,0.05
Coefficients (r) and P values were calculated by the Spearman’s rank coefficient test. a These variables were log transformed before analyses. 2hPIns, 2-hour postprandial insulin; APN, adiponectin; FFA, free fatty acid; FIns, fasting insulin; FPG, fasting plasma glucose; HDL-C, highdensity lipoprotein cholesterol; HOMA-B, homeostasis model assessment-B; HOMA-IR, homeostatic model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein; ISI, insulin sensitivity index; LDL-C, low-density lipoprotein cholesterol; RBP4, retinol-binding protein; TC, total cholesterol; TG, triglycerides; UA, uric acid.
are associated with pro-inflammatory factors, not with IR, in patients with T2DM and CAD.5,34 Taken together, our findings suggest that RBP4 might be a useful marker of CVD risk. We therefore hypothesized that RBP4 would be associated with the prevalence of CAD. We found that serum APN concentrations were significantly reduced in subjects with CAD and negatively associated with RBP4 concentrations. APN, the only adipokine with protective effects in humans, has been reported to have antiinflammatory, anti-IR and anti-atherogenic properties.35,36 APN can modulate the biological actions of several growth factors by controlling their bioavailability at a pre-receptor
level.37 APN rapidly accumulates in the subendothelial space of an injured artery38 and inhibits atherogenic processes.38–42 Thus, our findings are consistent with the hypothesis that circulating RBP4 may be a marker of atherosclerosis and CAD. This study had several limitations, including the relatively small sample size and the use of subjects from a single geographic area. Although we found that RBP4 concentrations differed in patients with CAD and controls, the difference was not statistically significant, suggesting the need to measure these serum proteins in larger patient populations. Moreover, our findings, suggesting correlations between serum RBP4
TABLE 3. Subject characteristics according to RBP4 quartiles First quartile Second quartile Number RBP4, mg/mL Control CAD CAD + HIns Age, yr BMI, kg/m2 FPG, mmol/L FIns, mU/La 2hPIns, mU/La TG, mmol/La TC, mmol/L LDL-C, mmol/La HDL, mmol/L UA, mmol/L hsCRP, mg/L FFA, mmol/L HOMA-IRa HOMA-Ba ISI APN, mg/L
22 14.60 (10.01–16.05) 13 9 0 51 6 10 22.71 6 2.54 4.55 6 0.35 4.60 (1.70–11.70) 28.39 (9.80–71.39) 0.99 (0.35–4.74) 4.38 6 0.68 1.90 (0.85–3.28) 1.79 6 0.65 276.01 6 63.89 0.52 6 0.52 0.56 6 0.39 0.95 (0.38–2.28) 95.63 (21.66–296.674.74) 23.03 6 0.53 338.57 6 241.47
23 17.20 (16.33–19.17) 9 13 1 57 6 13 22.79 6 2.59 4.74 6 0.44 5.80 (1.80–11.00) 26.51 (1.50–95.67) 1.13 (0.46–5.88) 4.54 6 1.33 2.21 (1.08–5.90) 1.62 6 0.72 332.74 6 75.05 3.05 6 0.90 0.58 6 0.29 1.12 (0.37–2.18) 80.96 (31.77–333.33) 23.21 6 0.47 352.78 6 257.16
Third quartile
Fourth quartile
22 22.51 (19.59–24.91) 3 6 13 62 6 10 23.36 6 2.39 5.09 6 0.77 9.44 (1.57–18.23) 62.39 (11.36–212.20) 1.38 (0.44–5.10) 4.05 6 1.16 2.08 (0.72–3.68) 1.31 6 0.43 362.10 6 100.17 6.12 6 2.40 0.48 6 0.31 2.03 (0.41–4.69) 132.30 (13.19–575.79) 23.72 6 0.66 251.61 6 160.25
22 32.11 (25.30–54.59) 5 2 15 61 6 11 24.46 6 3.79 4.93 6 0.54 9.65 (3.70–37.40) 83.53 (12.49–323.90) 1.49 (0.49–5.06) 4.09 6 0.88 1.97 (1.01–3.06) 1.30 6 0.41 346.70 6 95.63 3.23 6 1.40 0.64 6 0.39 1.86 (0.84–8.43) 145.63 (46.38–476.43) 23.95 6 0.63 379.58 6 183.47
P ,0.001
0.004 0.17 0.01 ,0.001 ,0.001 0.53 0.35 0.06 0.01 0.007 0.006 0.50 ,0.001 0.03 ,0.001 0.23
Data are presented as mean 6 standard deviation (if normally distributed) or median (interquartile range). a These variables were log transformed before analyses. APN, adiponectin; FFA, free fatty acid; FIns, fasting insulin; 2hPIns, 2-hour postprandial insulin; FPG, fasting plasma glucose; HDL-C, highdensity lipoprotein cholesterol; HOMA-B, homeostasis model assessment-B; HOMA-IR, homeostatic model assessment of insulin resistance; hsCRP, high-sensitivity C-reactive protein; ISI, insulin sensitivity index; LDL-C, low-density lipoprotein cholesterol; RBP4, retinol-binding protein; TC, total cholesterol; TG, triglycerides; UA, uric acid.
Ó 2014 Lippincott Williams & Wilkins
477
Li et al
levels and CAD development, require confirmation in a larger, prospectively recruited patient sample.
17. Oren A, Vos LE, Uiterwaal CS, et al. Cardiovascular risk factors and increased carotid intima-media thickness in healthy young adults: the Atherosclerosis Risk in Young Adults (ARYA) Study. Arch Intern Med 2003;163:1787–92.
CONCLUSIONS
18. Furumoto T, Fujii S, Saito N, et al. Relationships between brachial artery flow mediated dilation and carotid artery intima-media thickness in patients with suspected coronary artery disease. Jpn Heart J 2002;43: 117–25.
Serum RBP4 levels were associated with IR and CVD risk factors in both men and women. These findings agree with the hypothesis that circulating RBP4 may be a marker of the metabolic syndrome, insulin resistant and CAD. Further studies are required to confirm these associations. REFERENCES 1. Fu Y, Luo N, Klein RL, et al. Adiponectin promotes adipocyte differentiation, insulin sensitivity, and lipid accumulation. J Lipid Res 2005; 46:1369–79. 2. Kershaw EE, Flier JS. Adipose tissue as an endocrine organ. J Clin Endocrinol Metab 2004;89:2548–56. 3. Trayhurn P, Wood IS. Adipokines: inflammation and the pleiotropic role of white adipose tissue. Br J Nutr 2004;92:347–55. 4. Yang Q, Graham TE, Mody N, et al. Serum retinol binding protein 4 contributes to insulin resistance in obesity and type 2 diabetes. Nature 2005;436:356–62. 5. Cho YM, Youn B-S, Lee H, et al. Plasma retinol-binding protein-4 concentrations are elevated in human subjects with impaired glucose tolerance and type 2 diabetes. Diabetes Care 2006;29:2457–61.
19. Park SE, Kim DH, Lee JH, et al. Retinol-binding protein-4 is associated with endothelial dysfunction in adults with newly diagnosed type 2 diabetes mellitus. Atherosclerosis 2009;204:23–5. 20. Xiao Y, Xu A, Hui X, et al. Circulating lipocalin-2 and retinol-binding protein 4 are associated with intima-media thickness and subclinical atherosclerosis in patients with type 2 diabetes. PLoS One 2013;8: e66607. 21. Bobbert T, Raila J, Schwarz F, et al. Relation between retinol, retinolbinding protein 4, transthyretin and carotid intima media thickness. Atherosclerosis 2010;213:549–51. 22. Ingelsson E, Lind L. Circulating retinol-binding protein 4 and subclinical cardiovascular disease in the elderly. Diabetes Care 2009;32: 733–5. 23. Wu J, Shi YH, Niu DM, et al. Association among retinol-binding protein 4, small dense LDL cholesterol and oxidized LDL levels in dyslipidemia subjects. Clin Biochem 2012;45:619–22.
6. Graham TE, Yang Q, Bluher M, et al. Retinol-binding protein 4 and insulin resistance in lean, obese, and diabetic subjects. N Engl J Med 2006;354:2552–63.
24. von Eynatten M, Lepper PM, Liu D, et al. Retinol-binding protein 4 is associated with components of the metabolic syndrome, but not with insulin resistance, in men with type 2 diabetes or coronary artery disease. Diabetologia 2007;50:1930–7.
7. Kloeting N, Graham TE, Berndt J, et al. Serum retinol-binding protein is more highly expressed in visceral than in subcutaneous adipose tissue and is a marker of intra-abdominal fat mass. Cell Metab 2007;6:79–87.
25. Ingelsson E, Sundström J, Melhus H, et al. Circulating retinolbinding protein 4, cardiovascular risk factors and prevalent cardiovascular disease in elderly. Atherosclerosis 2009;206:239–44.
8. Ribel-Madsen R, Friedrichsen M, Vaag A, et al. Retinol-binding protein 4 in twins regulatory mechanisms and impact of circulating and tissue expression levels on insulin secretion and action. Diabetes 2009;58:54–60.
26. Cabré A, Lázaro I, Girona J, et al. Retinol-binding protein 4 as a plasma biomarker of renal dysfunction and cardiovascular disease in type 2 diabetes. J Intern Med 2007;262:496–503.
9. Stefan N, Schick F, Hennige AM, et al. High circulating retinolbinding protein 4 is associated with elevated liver fat but not with total, subcutaneous, visceral, or intramyocellular fat in humans. Diabetes Care 2007;30:1173–8.
27. Mahmoudi MJ, Mahmoudi M, Siassi F, et al. Circulating retinolbinding protein 4 concentrations in patients with coronary artery disease and patients with type 2 diabetes mellitus. Int J Diabetes Dev Ctries 2012;32:105–10.
10. Stern MP. Diabetes and cardiovascular disease. The “common soil” hypothesis. Diabetes 1995;44:369–74.
28. Radziuk J. Insulin sensitivity and its measurement: structural commonalities among the methods. J Clin Endocrinol Metab 2000;85: 4426–33.
11. Matsumoto K, Miyake S, Yano M, et al. Insulin resistance and arteriosclerosis obliterans in patients with NIDDM. Diabetes Care 1997;20: 1738–43.
29. Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21:2191–2.
12. Hanley AJ, Williams K, Stern MP, et al. Homeostasis model assessment of insulin resistance in relation to the incidence of cardiovascular disease: the San Antonio Heart Study. Diabetes Care 2002;25:1177–84.
30. Belfiore F, Iannello S, Volpicelli G. Insulin sensitivity indices calculated from basal and OGTT-induced insulin, glucose, and FFA levels. Mol Genet Metab 1998;63:134–41.
13. Bonora E, Formentini G, Calcaterra F, et al. HOMA-estimated insulin resistance is an independent predictor of cardiovascular disease in type 2 diabetic subjects: prospective data from the Verona Diabetes Complications Study. Diabetes Care 2002;25:1135–41.
31. Akbay E, Muslu N, Nayir E, et al. Serum retinol binding protein 4 level is related with renal functions in type 2 diabetes. J Endocrinol Invest 2010;33:725–9.
14. Mallat Z, Simon T, Benessiano J, et al. Retinol-binding protein 4 and prediction of incident coronary events in healthy men and women. J Clin Endocrinol Metab 2008;94:255–60. 15. Sun Q, Kiernan UA, Shi L, et al. Plasma retinol-binding protein 4 (RBP4) levels and risk of coronary heart disease: a prospective analysis among women in the nurses’ health study. Circulation 2013;127:1938–47. 16. Vu-Dac N, Gervois P, Torra IP, et al. Retinoids increase human apo C-III expression at the transcriptional level via the retinoid X receptor. Contribution to the hypertriglyceridemic action of retinoids. J Clin Invest 1998;102:625–32.
478
32. Thawnashom K, Tungtrongchitr R, Chanchay S, et al. Association between retinol-binding protein and renal function among Asian subjects with type 2 diabetes mellitus: a cross-sectional study. Southeast Asian J Trop Med Public Health 2011;42:936–45. 33. Polidori MC, Pratico D, Parente B, et al. Elevated lipid peroxidation biomarkers and low antioxidant status in atherosclerotic patients with increased carotid or iliofemoral intima media thickness. J Invest Med 2007;55:163–7. 34. Gey KF, Ducimetiere P, Evans A, et al. Low plasma retinol predicts coronary events in healthy middle-aged men: the PRIME Study. Atherosclerosis 2010;208:270–4.
Volume 348, Number 6, December 2014
RBP4 is a Novel Cardiovascular Disease Risk Factor
35. Al-Daghri NM, Al-Attas OS, Alokail M, et al. Retinol binding protein-4 is associated with TNF-alpha and not insulin resistance in subjects with type 2 diabetes mellitus and coronary heart disease. Dis Markers 2009;26:135–40.
39. Arita Y, Kihara S, Ouchi N, et al. Adipocyte-derived plasma protein adiponectin acts as a platelet-derived growth factor-BB-binding protein and regulates growth factor-induced common postreceptor signal in vascular smooth muscle cell. Circulation 2002;105:2893–8.
36. Lu P, Hu D, Lu J, et al. The association between uric acid and coronary heart disease [in Chinese]. Zhonghua Nei Ke Za Zhi 2002;41:526–9.
40. Okamoto Y, Arita Y, Nishida M, et al. An adipocyte-derived plasma protein, adiponectin, adheres to injured vascular walls. Horm Metab Res 2000;32:47–50.
37. Ouchi N, Kihara S, Arita Y, et al. Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-kappaB signaling through a cAMP-dependent pathway. Circulation 2000;102:1296–301. 38. Ouchi N, Kihara S, Arita Y, et al. Adipocyte-derived plasma protein, adiponectin, suppresses lipid accumulation and class A scavenger receptor expression in human monocyte-derived macrophages. Circulation 2001;103:1057–63.
Ó 2014 Lippincott Williams & Wilkins
41. Ouchi N, Kihara S, Arita Y, et al. Novel modulator for endothelial adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation 1999;100:2473–6. 42. Wang Y, Lam KS, Xu JY, et al. Adiponectin inhibits cell proliferation by interacting with several growth factors in an oligomerizationdependent manner. J Biol Chem 2005;280:18341–7.
479
