Relationship Between Retinol-Binding Protein-4/Adiponectin and Leptin/Adiponectin Ratios with Insulin Resistance and Inflammation

Introduction

C-reactive protein (CRP) is now accepted as the prototypic marker of inflammation. In fact high-sensitivity (hs) CRP levels have been shown to predict cardiovascular events in numerous studies. ¹ Also the homeostasis model assessment of insulin resistance (HOMA-IR) has been a widely accepted measure of in vivo insulin resistance, and it has been shown to correlate with the euglycemic clamp and other measures. ² Recently there has been much interest in the dysregulation of adipokines in obesity and metabolic syndrome. ^{3 –7} Also it has been previously suggested that the leptin/adiponectin ratio correlates with insulin resistance and appears to predict cardiovascular disease (CVD) in some, but not all, studies. ^{8 –15} However, the role of leptin in humans with regard to insulin resistance and CVD is far from settled. ^3,6,16 Retinol-binding protein-4 (RBP4) is an adipokine that is produced by the liver and adipose tissue and has been shown to impair insulin action in animal models and cell biological experiments. ^{17 –19} However, studies in humans examining RBP4 in metabolic syndrome and diabetes, while confirming a relationship with insulin resistance, are not uniform. ^{19 –21} There is greater scientific rationale supporting a role for RBP4 rather than leptin in inducing insulin resistance. Thus, in this study of a cohort that comprises individuals without metabolic syndrome and patients with metabolic syndrome, we determined whether the ratio of leptin/adiponectin and RBP4/adiponectin correlated with measures of insulin resistance (HOMA-IR) and inflammation (hsCRP).

Materials and Methods

All subjects were recruited from Sacramento County, California, through fliers and advertisements in the newspaper. The study included subjects (aged 27–69 years) with nascent metabolic syndrome (n=40) and control subjects (n=32). Metabolic syndrome was defined using the criteria of the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III), as described previously. ^22,23 Briefly, subjects classified as having nascent metabolic syndrome ^22,23 (no diabetes or clinical CVD) had to have at least three risk features of metabolic syndrome to sustain the diagnosis, including central obesity, hypertension, fasting glucose levels between 100 and 125 mg/dL, dyslipidemia [low high-density lipoprotein (HDL), high triglycerides (TGs)], and/or hypertension. None of the subjects were on statins, angiotensin receptor blockers, fibrates, aspirin, or peroxisome proliferator-activated receptor-γ (PPARγ agonists. Metabolic syndrome patients were on diuretics, β-blockers, or angiotensin-converting enzyme (ACE) inhibitors for their hypertension. None of the subjects had any reported chronic inflammatory, renal, or hepatic diseases. Control subjects needed to have at least two features of metabolic syndrome and not be on blood pressure medications. Other exclusion criteria for control subjects were fasting plasma glucose (>100 mg/dL) and TGs (>200 mg/dL). ^22,23 All subjects had a hsCRP levels <10 mg/L and normal white cell counts.

Informed consent was obtained from participants in the study, which was approved by the institutional review board at the University of California Davis. All human investigation was conducted according to the principles expressed in the Declaration of Helsinki. After history and physical examination, fasting blood was obtained for basic chemistries and insulin to calculate HOMA-IR and hsCRP levels as described previously. ^22,23 In addition adiponectin, leptin, and RBP4 were assayed by enzyme-linked immunosorbent assay (ELISA) using kits from Linco as described previously with an intraassay coefficient of variation of <9%. ²³

Statistical analysis

Data are expressed as mean±standard deviation (SD) or, for skewed variables, as median and interquartile range. Log transformations were applied to skewed data prior to parametric analyses. Spearman rank correlation coefficients were computed to assess the association between the adiponectin, leptin, and RBP4 variables and metabolic measurements. Comparisons between the control and metabolic syndrome groups were made with two-sample t-tests and analysis of covariance (ANCOVA) to adjust for body mass index (BMI) and age. Data were analyzed using SAS version 9.3 (SAS Institute, Cary, NC).

Results

The salient characteristics of the study group are shown in Table 1. In this cohort of 72 patients, 58 female and 14 male, comprising individuals without (n=32) and with the metabolic syndrome (n=40), there was a significant correlation of the leptin/adiponectin ratio with BMI (r=0.55, P<0.0001) and waist circumference (r=0.40, P=0.0006). However, the RBP4/adiponectin ratio did not correlate significantly with BMI or waist circumference. As a measure of insulin resistance, HOMA-IR was calculated. In this cohort, neither the leptin/adiponectin nor the RBP4/adiponectin ratios correlated with HOMA-IR (r=0.21, r=0.22, respectively, P≥0.12). Because CRP is a prototypic marker of inflammation, the ratios were also correlated with this biomarker. Both the leptin/adiponectin ratio (r=0.48, P<0.0001) and the RBP4/adiponectin ratio (r=0.28, P=0.02) correlated significantly with hsCRP levels. Because this cohort comprised patients with metabolic syndrome and controls, we compared the ratios using BMI as a co-variant. The RBP4/adiponectin ratio was significantly increased in patients with metabolic syndrome compared to controls, even after adjusting for BMI and age. Also the RBP4/adiponectin ratio was significantly increased following adjustment for waist circumference (P=0.01). However, there was no significant increase in the co-variate-adjusted ratio of leptin/adiponectin (Fig. 1).

OPEN IN VIEWER

FIG. 1.

Retinol-binding protein-4 (RBP4)/adiponectin and leptin/adiponectin ratios in controls and metabolic syndrome subjects. The lower and upper limits of the boxes indicate the 25th and 75th percentiles, the line within the boxes depicts the median, and the whiskers (error bars) below and above the boxes indicate the 10th and 90th percentiles. Controls versus metabolic syndrome: RBP4/adiponectin [body mass index (BMI) adjusted P=0.004]; leptin/adiponectin (BMI adjusted P=0.11). Asterisk (*) indicates significance.

Discussion

In this short report, we tested the validity of two relevant ratios—leptin/adiponectin and RBP4/adiponectin—as measures of insulin resistance and inflammation. Because there is scanty data on the RBP4/adiponectin ratio and there is greater scientific plausibility for a role in insulin resistance based on the work from the Kahn group, ^17,18 we included this ratio also. In this cohort, which was not diabetic or elderly with a mean age of 53 years, we failed to show a correlation with HOMA-IR, an accepted surrogate of insulin resistance. Although it can be argued that our sample size was inadequate, we need to emphasize that these ratios were tested in a wide spectrum of individuals spanning no features of metabolic syndrome to all five features. In fact, the range of HOMA-IR was 0.9–9.1 and fasting insulin ranged from 1.4 to 40.8 μU/mL. Hence both insulin-sensitive and insulin-resistant persons made up our study group. Also, von Eynatten in their study as well as others failed to show a correlation between RBP4 and insulin resistance. ^19,20

Most interestingly, we showed a significant correlation of both ratios with hsCRP, the prototypic downstream biomarker of inflammation. With respect to the leptin/adiponectin ratio, some studies support a role in atherosclerosis, ^{11 –13} including the prospective study of Kappelle et al., ¹⁵ which showed that the ratio predicted incident CVD in an adjusted analyses over a 3-year follow-up in males. However, it needs to be pointed out that in the MONICA/KORA Augsberg study, which comprised both males and females, the leptin/adiponectin ratio did not predict incident CVD. ¹⁴ Nonetheless, the majority of studies have shown a relationship with CVD, and this could be explained on the basis of this ratio defining a proinflammatory phenotype. In fact, the significant correlation with hsCRP supports this notion. Furthermore, numerous studies support a role for leptin promoting inflammation and adiponectin as an anti-inflammatory protein. ^{3,5 –7,24}

Our novel observation is the significant correlation between the RBP4/adiponectin ratio and hsCRP. Once again, some data support a role for RBP4 in inducing inflammation. ^25,26 Most interestingly, studies also support a role of RBP4 in CVD. ^{27 –30} Hence, RBP4, by inducing inflammation and endothelial dysfunction, can contribute to CVD; however, more prospective studies are required to confirm these findings.

Finally, although numerous groups have reported increases in the leptin/adiponectin ratio in metabolic syndrome, they failed to correct for adiposity, and leptin is clearly a marker of adiposity. ^31,32 In this report, we show that this ratio is not significantly increased in metabolic syndrome following adjustment for adiposity, and we once again caution researchers to adjust for obesity before ascribing a biomarker/biomediator to be dysregulated in metabolic syndrome per se. However, we make the novel observation that following adjustment for both BMI and waist circumference, the RBP4/adiponectin ratio is significantly increased in nascent metabolic syndrome without diabetes, clinical CVD, microalbuminuria, or renal disease and thus could be a better predictor of CVD in metabolic syndrome. However, this can only be settled in large prospective studies in patients with nascent metabolic syndrome.

There are certain limitations to our study. It could be argued that our sample size was not large enough. Furthermore, like most investigators, we assayed total adiponectin and not the more biologically active high-molecular-weight forms. Be that as it may, we clearly show that both ratios are valid measures of inflammation and could have a relevant role in predicting CVD. Previously, we showed substantial dysregulation of subcutaneous adipose tissue biology in nascent metabolic syndrome, contributing to the increased inflammation. ²³ In this report, we suggest that the RBP4/adiponectin ratio might provide an additional measure of the role of inflammation, especially emanating from adipose tissue in metabolic syndrome.

In conclusion, although we failed to show a significant correlation of either biomarker with insulin resistance in a middle-aged nondiabetic population, we have shown that both are measures of inflammation and, in fact, the RBP4/adiponectin ratio is superior in metabolic syndrome.