Adiponectin Is an Important Determinant of ApoA-I Catabolism

【摘要】  Objective- Plasma concentration of adiponectin is positively correlated with high-density lipoprotein (HDL) cholesterol level. However, the role of adiponectin on HDL metabolism remains unknown. This prompted us to perform an in vivo kinetic study of apoA-I, the main apolipoprotein of HDL, using stable isotopes, in 22 subjects with a wide range of plasma adiponectin, including 11 patients with metabolic syndrome (8 with type 2 diabetes, 3 without type 2 diabetes) and 11 normal individuals.

Methods and Results- In the 22 studied subjects, plasma adiponectin levels ranged from 2.57 to 14.44 µg/mL and apoA-I fractional catabolic rate (FCR) values ranged from 0.142 to 0.340 day -1. A strong negative correlation was found between adiponectin and apoA-I FCR ( r =-0.66, P <0.001) in the whole studied population and, to a similar extent, in patients with metabolic syndrome ( r =-0.73, P =0.010) and normal subjects ( r =-0.68, P =0.020), separately. In multivariable analysis, apoA-I FCR was associated negatively with adiponectin ( P =0.005) and positively with HDL triglycerides/cholesterol ratio ( P =0.006), but not with age, sex, body mass index (BMI), waist circumference, plasma triglycerides, HDL cholesterol, fasting glycemia, and QUICKI. Both adiponectin and HDL triglycerides/cholesterol ratio explained 62% of the variance of apoA-I FCR and adiponectin on its own explained 43%.

Conclusions- Our kinetic study shows a strong negative correlation between adiponectin and apoA-I FCR, which can explain the positive link between HDL cholesterol and adiponectin. This association is independent of obesity, insulin resistance, and the content of triglycerides within HDL particles. These data suggest that adiponectin may have a direct role on HDL catabolism.

We show a very significant association between adiponectin and apoA-I FCR independent of other factors, such as HDL triglyceride content or clinical and biological indexes of insulin sensitivity. Plasma adiponectin, on its own, explains 43% of apoA-I FCR variance, suggesting that adiponectin may have a direct role on HDL catabolism.

【关键词】  adiponectin apoAI HDL cholesterol insulinresistance kinetic

Introduction

Adiponectin is a peptide predominantly synthesized in the adipose tissue that plays an important role in carbohydrate and lipid metabolism and vascular biology. 1,2 It has been suggested to be a link between obesity, insulin resistance, and cardiovascular disease. Plasma adiponectin levels are reduced in individuals with abdominal obesity, metabolic syndrome, and/or type 2 diabetes. 1,3-5 Adiponectin concentration has been found negatively correlated with abdominal obesity and insulin resistance in humans 6,7 and has been shown to predict the development of type 2 diabetes. 8,9 Furthermore, several studies indicate that plasma adiponectin levels are significantly decreased in patients with coronary heart disease 10,11 and high plasma adiponectin predict a lower risk of future myocardial infarction in nondiabetic 12 and diabetic individuals. 13 However, the link between adiponectin and cardiovascular disease could be partly mediated by its effects on lipids, because several studies indicate that the inverse association between adiponectin and coronary disease is importantly attenuated or no more significant after adjustment for lipids, particularly high-density lipoprotein (HDL) cholesterol. 12,13 Adiponectin is related to lipid metabolism, principally higher levels of HDL cholesterol and lower levels of triglycerides. 3 The positive association between plasma adiponectin and HDL cholesterol has been found in nondiabetic 6,14,15 and diabetic individuals. 16,17 Furthermore, the positive correlation between adiponectin and HDL cholesterol has been shown to be independent of body mass index (BMI), body fat distribution and insulin sensitivity, 6,7,14,16,17 suggesting a potential direct link between adiponectin and HDL metabolism. The mechanism for the association between plasma adiponectin and HDL cholesterol is still unknown and has not been previously investigated. In order, to get further insight into the effect of adiponectin on HDL metabolism, we performed an in vivo kinetic study of apoA-I, the major apolipoprotein of HDL particles, using stable isotopes, in a group of individuals with a wide range of plasma adiponectin, including patients with metabolic syndrome and normal subjects.

Materials and Methods

Subjects

We studied 22 white individuals, including 11 patients with metabolic syndrome and 11 normal subjects. All the patients with metabolic syndrome were abdominally obese (waist circumference 102 cm) and were classified as having metabolic syndrome according to the National Cholesterol Education Program Adult Treatment Panel III criteria. 18 All the patients with metabolic syndrome had biological indexes of insulin resistance: increased HOMA-IR (homeostasis model assessment) 19 and decreased QUICKI (quantitative insulin-sensitivity check index). 20,21 Among the 11 patients with metabolic syndrome, 8 had type 2 diabetes according to the criteria of the American Diabetes Association. Patients with type 2 diabetes were treated with diet alone (2 patients) or with glinides (6 patients). All patients with metabolic syndrome had normal renal function and were not using any medication known to affect lipid metabolism.

All normal subjects were in good health, with normal glucose tolerance and normal plasma lipid levels. They were not using any medication. All the women included in the study were not using oral contraceptives. The protocol was approved by the Dijon University Hospital ethics committee and written informed consent was obtained before the study was started.

Experimental Protocol

The kinetic study was performed in the fed state. Food intake, with a leucine-poor diet (1700 kcal/day-1, 55% carbohydrates, 39% fats, and 7% proteins), was fractionated in small portions, which were provided every 2 hours starting 6 hours before the tracer infusion up to the end of the study to avoid important variations in apolipoprotein plasma concentration, as previously performed by our group 22,23 and others. 24 The endogenous labeling of apoA-I was performed by administration of L-[1-1 3 C] leucine (99 atom%; Eurisotop, Saint Aubin, France), dissolved in 0.9% NaCl solution. At 08:00 AM, each subject received intravenously a primed infusion of 0.7 mg/kg -1 of tracer, immediately followed by a 16-hour constant infusion of 0.7 mg/kg -1 per hour -1. Blood samples were collected at 0, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 12, 14, 15, and 16 hours after the primed infusion. Serum was separated by centrifugation for 10 minutes at 4°C and 3000 g.

To avoid the influence of acute exercise on lipid metabolism, all studied subjects were instructed to refrain from strenuous exercise 3 days before the kinetic study.

Analytical Procedure

Analytical procedures were performed as previously described in details. 22,25

Isolation of Apolipoproteins

Very-low-density lipoprotein (VLDL) and HDL were isolated from plasma by a gradient ultracentrifugation using a SW41 rotor in a L90 apparatus (Beckman Instruments, Palo Alto, Calif). VLDL and HDL fractions were then dialyzed against a 10 mmol/L -1 ammonium bicarbonate buffer pH 8.2 containing 0.01% EDTA and 0.013% sodium azide. VLDL and HDL fractions were delipidated 1 hour at -20°C using 10 volumes of diethylether-ethanol 3:1. Apolipoproteins from each lipoprotein fraction were isolated by preparative discontinuous sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 3% gel and 15% gel. After staining with Coomassie blue R-250, apolipoprotein bands were excised from polyacrylamide gels and hydrolyzed in 6 mol/L HCl at 110°C for 16 hours under nitrogen. Samples were then centrifuged to remove polyacrylamide. Supernatants were lyophilized in a Speed Vac (Savant Instrument, Farmingdale, NY). Lyophilized samples were dissolved in 50% acetic acid and applied to an AG-50W-X8 200 to 400 mesh cation exchange resin (Bio-Rad, Richmond, Va) and amino acids were recovered by elution with 4N NH 4 OH.

Determination of Leucine Enrichment by Gas Chromatography/Combustion/Isotope Ratio Mass Spectrometry

Amino acids were converted to N-acetyl O-propyl esters and were analyzed on a Finnigan Mat Delta Plus Advantage isotope ratio mass spectrometer (Finnigan Mat, Bremen, Germany). 13 C leucine enrichment was initially expressed in delta and converted in tracer/tracee ratio before modeling. 26

Modeling

Kinetic data were analyzed with the simulation analysis and modeling SAAM II program 27 (SAAM Institute, Inc, Seattle, Wash). ApoA-I and B-100 data were analyzed using the following monoexponential function: A(t)=Ap(1-exp[-k(t-d)]), where A(t) is the apolipoprotein enrichment at time t, Ap the enrichment at the plateau of the VLDL apo B100 curve, d the delay between the beginning of the experiment and the appearance of tracer in the apolipoprotein, and k the fractional synthetic rate (FSR) of the apolipoprotein. 26,28 It was assumed that the VLDL apoB100 tracer/tracee ratio at the plateau corresponds to the tracer/tracee ratio of the leucine precursor pool. 28 This estimation is made on the assumption that apoB-100 and the great majority of apoA-I are synthesized by the liver, as previously demonstrated by Ikewaki et al. 28 In the steady state, the FSR equals the fractional catabolic rate (FCR). 26

The apoA-I production rate (PR) was calculated as the product of the apoA-I pool size and its FSR divided by body weight. Pool size was estimated by the product of apoA-I plasma concentration and plasma volume, calculated as 4.5% of body weight. 25 In obese subjects (BMI 30), a correction of plasma volume was performed as previously reported by many authors. 29,30 The plasma volume was modified by multiplying by a correction factor to take into account the decrease in relative plasma volume associated with an increase in body weight. 31 The correction factor obtained from specific tables 31 increases with body weight, ranging from 0.99 up to 0.90 for the more obese patients.

Analytical Methods

Laboratory Measurements

Concentrations of apoA-I were determined by immunoturbidimetry with anti-apoA-I antibodies from Boehringer Mannheim. All chemical lipid assays were performed on a Cobas-Fara Centrifugal Analyzer (Hoffmann-La Roche). Total cholesterol and triglyceride concentrations were measured by enzymatic method using Boehringer Mannheim and Roche reagents, respectively.

Plasma glucose concentration was measured by enzymatic method (glucose oxidase) on a Vitros 950 analyzer (Ortho Clinical Diagnostics, Rochester, NY). Plasma insulin was measured by radioimmunoassay (CIS Bio International, Gif sur Yvette, France).

Plasma adiponectin concentration was determined using an enzyme immunoassay kit (Quantikine R&D, Systems, Minneapolis, Minn). The kit used measures the monomeric, the dimeric, and the trimeric forms of adiponectin. The intra-assay and inter-assay coefficients of variation for this method were less than 5% and 7%, respectively.

Insulin Resistance Evaluation

The insulin resistance level was estimated in all individuals by using both the homeostasis model assessment (HOMA) method 19 and the quantitative insulin-sensitivity check index (QUICKI). 20,21 HOMA index (HOMA-IR) was calculated as [fasting glycemia (mmol/L) x fasting insulin (µU/mL)]/22.5. 19 QUICKI, which has been shown to be among the most accurate indexes for determining insulin sensitivity in humans, was calculated as 1/[log(fasting insulin expressed in (µU/mL)+log (fasting glycemia expressed in mg/dL)]. 20,21

Statistical Analysis

Data are reported as mean±SD. Statistical calculations were performed using the SPSS software package. For continuous variables, a Kolmogorov-Smirnov analysis was performed to test for normality. Comparisons of continuous variables between patients with metabolic syndrome and normal subjects were performed either by unpaired Student t test for normally distributed data or by nonparametric Mann-Whitney U test for non-normally distributed data (plasma triglycerides). The Pearson correlation coefficients (r) were determined by linear regression analysis. Statistical significance of the correlation coefficients was determined by the method of Fisher and Yates. Because apoA-I FCR is one of the term in the equation to calculate apoA-I PR, the correlation between plasma adiponectin and apoA-I PR has been performed using partial correlation, after controlling for apoA-I FCR. A multivariable linear regression analysis was performed to analyze the influence of different factors on apoA-I FCR. A 2-tailed P =0.05 was accepted as statistically significant.

Results

Clinical and Biological Characteristics

Clinical and biological characteristics are shown in Table 1 for the entire studied population, including both patients with metabolic syndrome and normal subjects and for each subgroup (patients with metabolic syndrome, normal subjects). Patients with metabolic syndrome compared with normal individuals had higher values for BMI, waist circumference, fasting glycemia, fasting insulin, HOMA-IR, triglycerides, and HDL triglycerides/cholesterol ratio, and lower HDL cholesterol concentrations and QUICKI values. Plasma adiponectin level was significantly lower in patients with metabolic syndrome than in normal subjects.

TABLE 1. Clinical and Biological Characteristics of the Entire Studied Population (Patients With Metabolic Syndrome and Normal Subjects) of Patients With Metabolic Syndrome and of Normal Subjects

Kinetic Data

The kinetic curves of apoA-I in patients with metabolic syndrome and controls are shown in Figure 1. ApoA-I kinetic parameters are shown in Table 2. ApoA-I pool was significantly lower in patients with metabolic syndrome due to significantly increased apoA-I FCR. ApoA-I PR was not different between patients with metabolic syndrome and normal subjects.

Figure 1. Kinetic curves of HDL apoA-I obtained during a primed constant infusion of L-[1 to 13C]leucine. [13C]leucine enrichment values, expressed as percentage of VLDL apo B100 plateau, for control subjects (white circles) and patients with metabolic syndrome (black circles) are shown. The curves were obtained by monoexponential modeling. Data are shown as mean±SEM.

TABLE 2. ApoA-I Kinetic Parameters in Patients With Metabolic Syndrome and in Normal Subjects

Correlation

In univariate analysis, plasma adiponectin level was significantly and negatively correlated with age ( r =-0.50, P =0.019), BMI ( r =-0.45, P =0.038), waist circumference ( r =-0.56, P =0.006), plasma triglycerides ( r =-0.46, P =0.032), HDL triglycerides/cholesterol ratio ( r =-0.49, P =0.021), and HOMA-IR ( r =-0.43, P =0.046). Plasma adiponectin was significantly and positively correlated with HDL cholesterol ( r =0.43, P =0.044) and QUICKI ( r =0.48, P =0.024). The correlation coefficient between plasma adiponectin and plasma apoA-I was r =0.40 ( P =0.063).

As shown in Figure 2, we found a strong negative correlation between plasma adiponectin and apoA-I FCR ( r =-0.66, P <0.001). This negative correlation between plasma adiponectin and apoA-I FCR was also found in each group of subjects to a similar extent, the patients with metabolic syndrome ( r =-0.73, P =0.010) and the normal subjects ( r =-0.68, P =0.020). The negative correlation between plasma adiponectin and apoA-I FCR was similar for both men ( r =-0.68, P <0.05) and women ( r =-0.68, P <0.05). Plasma adiponectin was not correlated with apoA-I PR (after controlling for apoA-I FCR).

Figure 2. Correlation between plasma adiponectin and apoA-I FCR in the 22 studied subjects.

In univariate analysis, apoA-I FCR was significantly and positively correlated with BMI ( r =0.52, P =0.012), plasma triglycerides ( r =0.62, P =0.002), HDL triglycerides/cholesterol ratio ( r =0.65, P =0.001), fasting glycemia ( r =0.45, P =0.036), and HOMA-IR ( r =0.56, P =0.007). ApoA-I FCR was negatively correlated with QUICKI ( r =-0.57, P =0.005), HDL cholesterol ( r =-0.55, P =0.008), and plasma adiponectin ( r =-0.66, P <0.001). ApoA-I FCR was not correlated with age.

Multivariable Analysis

A multivariable linear regression analysis was performed to analyze the association between apoA-I FCR and several variables. The variables introduced into the model were those that were significantly associated with apoA-I FCR in the univariate analysis (BMI, waist circumference, plasma triglycerides, HDL cholesterol, HDL triglycerides/cholesterol ratio, fasting glycemia, QUICKI, plasma adiponectin), age, and sex, because of potential apoA-I FCR differences between men and women. The multivariable analysis showed that apoA-I FCR was negatively associated with plasma adiponectin ( P =0.005) and positively associated with HDL triglycerides/cholesterol ratio ( P =0.006) ( Table 3 ). These 2 variables explained 62% of apoA-I FCR variance ( r 2 =0.62). The association between plasma adiponectin and apoA-I FCR was shown to be important, with plasma adiponectin explaining 43% of apoA-I FCR variance.

TABLE 3. Relationship Between apoA-I FCR and Several Variables Assessed by Multivariable Linear Regression in the Entire Studied Population (r 2 =0.62)

Discussion

Several studies have clearly shown a significant correlation between plasma adiponectin and HDL cholesterol level in nondiabetic 6,14,15,32 and diabetic individuals. 16,17 This association between adiponectin and HDL cholesterol has been shown to be independent of BMI, body fat distribution, and insulin sensitivity in cross-sectional studies. 6,7,14,16,17,32 Moreover, in a recent intervention study, it has been shown that the augmentation of plasma adiponectin after weight loss in obese patients was positively correlated with the increase in plasma HDL cholesterol independently of changes in adiposity and insulin sensitivity. 33 All these data strongly suggest a possible direct link between adiponectin and HDL cholesterol. To get further insight into the role of adiponectin on HDL metabolism, we report the first study of the relationship between plasma adiponectin and apoA-I kinetics in humans. To have a good estimate of the relationship between adiponectin and apoA-I metabolism, we have performed a study in individuals with a wide range of plasma adiponectin and a wide range of apoA-I FCR including patients with metabolic syndrome and normal subjects. Age was different between the 2 groups. However, this age difference is not likely to affect our results because it has been shown that apoA-I kinetic is independent of age. 34 Moreover in our study, data from the univariate and multivariate analyses clearly show that apoA-I FCR is not influenced by age. Our study shows that adiponectin is significantly and negatively associated with apoA-I FCR and that this association is independent of other factors usually associated with apoA-I FCR (sex, abdominal obesity, plasma triglycerides, HDL triglycerides, insulin sensitivity). Because 8 of our patients had type 2 diabetes, we screened for a possible association between plasma adiponectin and fasting glycemia. But fasting glycemia was not shown to be associated with plasma adiponectin level neither in the univariate analysis nor in the multivariate analysis. In our study, plasma adiponectin can explain 43% of the variance of apoA-I FCR, indicating a tight link between both variables. Furthermore, the strong negative correlation between plasma adiponectin and apoA-I FCR is found not only in the whole studied population (including patients with metabolic syndrome and normal subjects) but also in each group (the patients with metabolic syndrome and the normal subjects). This confirms the results of the multivariable analysis showing that the association between adiponectin and apoA-I FCR is independent of insulin sensitivity.

The positive correlation between HDL cholesterol and plasma adiponectin, previously reported, 6,14,15,32 is confirmed in the present study. HDL cholesterol is also significantly negatively correlated with apoA-I FCR and it is now admitted that increased apoA-I catabolism, indicating accelerated HDL catabolism, is the kinetic factor responsible for decreased HDL cholesterol level in insulin-resistant subjects. 30 In our study, HDL cholesterol, which is negatively correlated with apoA-I FCR in the univariate analysis, is no more associated with apoA-I FCR in the multivariable analysis when plasma adiponectin is present into the model. This indicates that the association between adiponectin and HDL cholesterol is explained by the relationship between adiponectin and apoA-I FCR.

The multivariable analysis indicates that apoA-I FCR is independently associated with plasma adiponectin (negatively) and HDL triglycerides/cholesterol ratio (positively) and that the 2 variables explain 62% of its variance. HDL particles of patients with metabolic syndrome are enriched in triglycerides. 35 As previously reported, 30,36 we found that triglyceride enrichment of HDLs is a factor for increased apoA-I catabolism. It has been shown that HDLs with a higher triglyceride content are better substrates for hepatic lipase leading to their faster catabolism. 36 An association between adiponectin and triglycerides has been reported in several studies 3,14,16,17,32,33 and is confirmed in the present one. Because plasma triglycerides (or HDL triglyceride content) is usually correlated with both apoA-I catabolism and plasma adiponectin, we could have suspected that the association between adiponectin and apoA-I FCR might be explained by the known link between plasma adiponectin and triglycerides. Our data exclude this hypothesis and clearly show that plasma adiponectin is an important determinant of apoA-I catabolism independently of triglycerides, explaining by its own 43% of apoA-I FCR variance.

Interestingly, we show that clinical (waist circumference) and biological (HOMA-IR, QUICKI) indexes for insulin resistance are significantly associated with apoA-I FCR in the univariate analysis, but no more in the multivariable analysis when plasma adiponectin is present in the statistical model. These results suggest that adiponectin could be an important link between insulin resistance and increased apoA-I catabolism.

Our data suggest that adiponectin may have a direct role on HDL catabolism. However, the mechanisms linking plasma adiponectin and HDL metabolism remain unknown and further studies are needed to explore them.

In conclusion, our in vivo kinetic study shows a significant negative correlation between plasma adiponectin and apoA-I FCR in individuals with a wide range of plasma adiponectin, including patients with metabolic syndrome and normal subjects. The significant association between plasma adiponectin and apoA-I FCR is independent of other factors, such as HDL triglyceride content or clinical and biological indexes of insulin sensitivity. Plasma adiponectin, on its own, explains 43% of the variance of apoA-I FCR. These data suggest that adiponectin may have a direct role on HDL catabolism.

Acknowledgments

We are indebted to Véronique Jost for the preparation of 13 C leucine, Cécile Gibassier for dietary assistance and Dominique Battaut, Elisabeth Niot, and Liliane Princep for invaluable technical assistance.

【参考文献】
  Arita Y, Kihara S, Ouchi N, Takahashi M, Maeda K, Miyagawa J, Hotta K, Shimomura I, Nakamura T, Miyaoka K, Kuriyama H, Nishida M, Yamashita S, Okubo K, Matsubara K, Muraguchi M, Ohmoto Y, Funahashi T, Matsuzawa Y. Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun. 1999; 257: 79-83.

Goldstein BJ, Scalia R. Adiponectin: a novel adipokine linking adipocytes and vascular function. J Clin Endocrinol Metab. 2004; 89: 2563-2568.

Hotta K, Funahashi T, Arita Y, Takahashi M, Matsuda M, Okamoto Y, Iwahashi H, Kuriyama H, Ouchi N, Maeda K, Nishida M, Kihara S, Sakai N, Nakajima T, Hasegawa K, Muraguchi M, Ohmoto Y, Nakamura T, Yamashita S, Hanafusa T, Matsuzawa Y. Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol. 2000; 20: 1595-1599.

Weyer C, Funahashi T, Tanaka S, Hotta K, Matsuzawa Y, Pratley RE, Tataranni PA. Hypoadiponectinemia in obesity and type 2 diabetes: close association with insulin resistance and hyperinsulinemia. J Clin Endocrinol Metab. 2001; 86: 1930-1935.

Matsuzawa Y, Funahashi T, Kihara S, Shimomura I. Adiponectin and metabolic syndrome. Arterioscler Thromb Vasc Biol. 2004; 24: 29-33.

Cnop M, Havel PJ, Utzschneider KM, Carr DB, Sinha MK, Boyko EJ, Retzlaff BM, Knopp RH, Brunzell JD, Kahn SE. Relationship of adiponectin to body fat distribution, insulin sensitivity and plasma lipoproteins: evidence for independent roles of age and sex. Diabetologia. 2003; 46: 459-469.

Tschritter O, Fritsche A, Thamer C, Haap M, Shirkavand F, Rahe S, Staiger H, Maerker E, Haring H, Stumvoll M. Plasma adiponectin concentrations predict insulin sensitivity of both glucose and lipid metabolism. Diabetes. 2003; 52: 239-243.

Lindsay RS, Funahashi T, Hanson RL, Matsuzawa Y, Tanaka S, Tataranni PA, Knowler WC, Krakoff J. Adiponectin and development of type 2 diabetes in the Pima Indian population. Lancet. 2002; 360: 57-58.

Spranger J, Kroke A, Mohlig M, Bergmann MM, Ristow M, Boeing H, Pfeiffer AF. Adiponectin and protection against type 2 diabetes mellitus. Lancet. 2003; 361: 226-228.

Dzielinska Z, Januszewicz A, Wiecek A, Demkow M, Makowiecka-Ciesla M, Prejbisz A, Kadziela J, Mielniczuk R, Florczak E, Janas J, Januszewicz M, Ruzyllo W. Decreased plasma concentration of a novel anti-inflammatory protein-adiponectin-in hypertensive men with coronary artery disease. Thromb Res. 2003; 110: 365-369.

Dunajska K, Milewicz A, Jedrzejuk D, Szymczak J, Kuliczkowski W, Salomon P, Bialy D, Poczatek K, Nowicki P. Plasma adiponectin concentration in relation to severity of coronary atherosclerosis and cardiovascular risk factors in middle-aged men. Endocrine. 2004; 25: 215-221.

Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB. Plasma adiponectin levels and risk of myocardial infarction in men. JAMA. 2004; 291: 1730-1737.

Schulze MB, Shai I, Rimm EB, Li T, Rifai N, Hu FB. Adiponectin and future coronary heart disease events among men with type 2 diabetes. Diabetes. 2005; 54: 534-539.

Yamamoto Y, Hirose H, Saito I, Tomita M, Taniyama M, Matsubara K, Okazaki Y, Ishii T, Nishikai K, Saruta T. Correlation of the adipocyte-derived protein adiponectin with insulin resistance index and serum high-density lipoprotein-cholesterol, independent of body mass index, in the Japanese population. Clin Sci (Lond). 2002; 103: 137-142.

Martin LJ, Woo JG, Daniels SR, Goodman E, Dolan LM. The relationships of adiponectin with insulin and lipids are strengthened with increasing adiposity. J Clin Endocrinol Metab. 2005; 90: 4255-4259.

Shetty GK, Economides PA, Horton ES, Mantzoros CS, Veves A. Circulating adiponectin and resistin levels in relation to metabolic factors, inflammatory markers, and vascular reactivity in diabetic patients and subjects at risk for diabetes. Diabetes Care. 2004; 27: 2450-2457.

Schulze MB, Rimm EB, Shai I, Rifai N, Hu FB. Relationship between adiponectin and glycemic control, blood lipids, and inflammatory markers in men with type 2 diabetes. Diabetes Care. 2004; 27: 1680- 1687.

Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001; 285: 2486-2497.

Matthews DR, Hosker JP, Rudenski BA, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and B-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia. 1985; 28: 412-419.

Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G, Quon MJ. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab. 2000; 85: 2402-2410.

Chen H, Sullivan G, Quon MJ. Assessing the predictive accuracy of QUICKI as a surrogate index for insulin sensitivity using a calibration model. Diabetes. 2005; 54: 1914-1925.

Duvillard L, Pont F, Florentin E, Gambert P, Vergès B. Significant improvement of apoB-containing lipoprotein metabolism by insulin treatment in NIDDM patients. Diabetologia. 2000; 43: 27-35.

Pont F, Duvillard L, Florentin E, Gambert P, Vergès B. Early kinetic abnormalities of apoB-containing lipoproteins in insulin-resistant women with abdominal obesity. Arterioscl Thromb Vasc Biol. 2002; 22: 1726-1732.

Taskinen MR, Packard CJ, Sheperd J. Effect of insulin therapy on metabolic fate of apolipoprotein B-containing lipoproteins in NIDDM. Diabetes. 1990; 39: 1017-1027.

Pont F, Duvillard L, Maugeais C, Athias A, Persegol L, Gambert P, Vergès B. Isotope ratio mass spectrometry, compared with conventional mass spectrometry in kinetic studies at low and high enrichment levels: application to lipoprotein kinetics. Anal Biochem. 1997; 248: 277-287.

Pont F, Duvillard L, Vergès B, Gambert P. Development of compartmental models in stable isotope experiments: application to lipoprotein metabolism. Arterioscl Thromb Vasc Biol. 1998; 18: 853-860.

Barrett PHR, Bell BM, Cobelli C, et al. SAAM II: Simulation, analysis and modeling software for tracer and pharmacokinetic studies. Metabolism. 1998; 47: 484-492.

Ikewaki K, Rader DJ, Schaefer JR, Fairwell T, Zech LA, Brewer HB Jr. Evaluation of apoA-I kinetics in humans using simultaneous endogenous stable isotope and exogenous radiotracer methods. J Lipid Res. 1993; 34: 2207-2215.

Cummings MH, Watts GF, Pal C, Umpleby M, Hennessy TR, Naoumova R, Sönksen PH. Increased hepatic secretion of very low density lipoprotein apolipoprotein B100 in obesity: a stable isotope study. Clin Science. 1995; 88: 225-233.

Pont F, Duvillard L, Florentin E, Gambert P, Vergès B. High-density lipoprotein apolipoprotein A-I kinetics in obese insulin resistant patients. An in vivo stable isotope study. Int J Obes Relat Metab Disord. 2002; 26: 1151-1158.

Dagher FJ, Lyons JH, Finlayson DC, Shamsai J, Moore FD. Blood volume measurement: a critical study. Advanc Surg. 1965; 1: 69-109.

Coté M, Mauriege P, Bergeron J, Almeras N, Tremblay A, Lemieux I, Després JP. Adiponectinemia in visceral obesity: impact on glucose tolerance and plasma lipoprotein and lipid levels in men. J Clin Endocrinol Metab. 2005; 90: 1434-1439.

Baratta R, Amato S, Degano C, Farina MG, Patane G, Vigneri R, Frittitta L. Adiponectin relationship with lipid metabolism is independent of body fat mass: evidence from both cross-sectional and intervention studies. J Clin Endocrinol Metab. 2004; 89: 2665-2671.

Matthan NR, Jalbert SM, Lamon-Fava S, Dolnikowski GG, Welty FK, Barrett HR, Schaefer EJ, Lichtenstein AH. TRL, IDL, and LDL apolipoprotein B-100 and HDL apolipoprotein A-I kinetics as a function of age and menopausal status. Arterioscler Thromb Vasc Biol. 2005; 25: 1691-1696.

Hansel B, Giral P, Nobecourt E, Chantepie S, Bruckert E, Chapman MJ, Kontush A. Metabolic syndrome is associated with elevated oxidative stress and dysfunctional dense high-density lipoprotein particles displaying impaired antioxidative activity. J Clin Endocrinol Metab. 2004; 89: 4963-4971.

Pietzsch J, Julius U, Nitzsche S, Hanefeld M. In vivo evidence for increased apolipoprotein A-I catabolism in subjects with impaired glucose tolerance. Diabetes. 1998; 47: 1928-1934.

作者单位：From Service Endocrinologie (B.V., J.M.P., F.G.), Diabétologie et Maladies Métaboliques, Centre Hospitalier Universitaire de Dijon, France; INSERM U 498 (B.V., J.M.P., L.D., G.D., E.F., P.G.), Faculté de Médecine, Dijon, France.