RBP4: From Retinol Transporter to Biomarker?

Overweight and obesity are major risk factors for type 2 diabetes mellitus and cardiovascular diseases. The coexistence of these 2 conditions may be characterized by the term cardiometabolic disease, the cause of which appears to be multifactorial, involving intricate interactions of environment, behavior, and biology. It is now well recognized that adipose tissue is not only just a site of energy storage, but it also has a crucial endocrine function as a major source of many functionally important signaling proteins (adipokines). Many of these adipokines have been shown to directly or indirectly affect insulin sensitivity through modulation of insulin signaling and the molecules involved in glucose, muscle, and lipid metabolism. Of these adipokines, retinol binding protein (RBP4) has recently attracted much attention because of its newly demonstrated function as a potential determinant of insulin resistance, diabetes, and cardiovascular risks. Although research and our understanding of cardiometabolic disease have advanced, major gaps exist with regard to the obesity-driven mechanisms that increase the progression of cardiometabolic disease, especially in children. One opportunity for elucidating these mechanisms involves validating potential contributing factors resulting from obesity, which independently enhance future susceptibility for cardiometabolic disease. In this context, insulin resistance appears to be a key factor in the development of cardiometabolic disease. Retinol binding protein 4 (RBP4) in its role as a potential signal in mediating obesity-driven insulin resistance seems to be an adipokine of considerable interest.

Four decades ago, when RBP4 was characterized by Kanai and Goodman,1, 2 it was known only as a transport protein for retinol, produced in the hepatocyte, to extrahepatic tissues. In 1992, Tsutsumi et al3 showed that RBP4 was also secreted from the adipocytes. Zovich et al4 demonstrated that differential expression of RBP4 secretion and gene expression were associated with adipogenesis. However, the potential role of RBP4 as a mediator of insulin resistance was recognized only recently through its relationship to glucose transporter 4, the critical insulin-sensitive transporter in insulin-sensitive target tissues. Yang et al5 showed in mice that adipose tissue–specific ablation of glucose transporter 4 increased RBP4 mRNA expression in adipocytes and serum RBP4 levels. Increased circulating levels of RBP4 were also shown to cause insulin resistance in muscle by inhibiting phosphatidylinositol 3-kinase and increased expression of the gluconeogenic enzyme phosphoenolpyruvate carboxykinase in the liver.5 Ever since this article from Kahn et al illuminated the relevance of RBP4 in insulin resistance and type 2 diabetes mellitus, there have been numerous reports related to this molecule and its potential role in insulin resistance. Graham et al6 extended these findings to human beings and showed that RBP4 concentration and the magnitude of insulin resistance were correlated in people with obesity and impaired glucose tolerance, as well as in patients with type 2 diabetes. A genetic predisposition was also suggested: (1) an association of 4 single nucleotide polymorphisms in the RBP4 gene with increased risk of type 2 diabetes in a Mongolian population7; (2) a correlation between RBP4 and the degree of insulin resistance in normoglycemic men with a positive family history of type 2 diabetes in the Caucasian population8; and (3) a promoter polymorphism in the RBP4 gene associated with increased risk of type 2 diabetes in the Rotterdam population.9 Despite these consistent and independent findings on physiology and genetics from diverse groups, a series of negative results regarding the association between RBP4 and insulin resistance have also been reported in clinical studies.10 Therefore, as of now it is unclear whether these discrepancies are solely due to methodologic issues as suggested11 or whether they are apparent uncertainties about the role of RBP4 given the number of reports indicating no association, even with use of the glucose clamp method for measurement of insulin resistance.

In contrast, it appears that the association between serum RBP4 levels and insulin resistance in animals are more consistent.12 This poses the question about whether RBP4 is an adipokine with a differential regulation in animals and human beings and perhaps adipose tissue is a less important source of circulating RBP4 in human beings than in animals. Although one might wonder whether the conflicting results in human beings indicate that RBP4 is not a sensitive marker for insulin resistance in all cases, various studies have also shown that interventions, including physical activity, that improve insulin sensitivity has a favorable impact on RBP4, suggesting that a decrease of serum RBP4 may be one of the mechanisms underlying improvement of insulin resistance.10 The discrepancies in the results also bring up the issue of potential differences in the patient characteristics (sex, race, degree of glucose tolerance or obesity) among the studies and the lack of longitudinal studies in specific populations. It has also been suggested that the regulation of RBP4 may also be influenced by age.13, 14 Recent reports suggesting that functional SNPs in the RBP4 gene may be associated with type 2 diabetes mellitus or insulin resistance, indicating that a potential difference in the effect of RBP4 on insulin resistance in individuals7, 8 underscores the need for a more detailed analysis to clarify these findings.

Although numerous reports on RBP4 in adults appeared in the recent years, there are only a handful of similar studies in the pediatric population,13, 14, 15, 16, 17 and we have been deprived of the much- needed data on the longitudinal changes in RBP4. All studies in children are cross-sectional,13, 15, 16 except for a recent 1-year longitudinal study in Caucasian children by Reinehr et al17 or have involved only short-term interventions in clinic-based populations.13, 17 None of them have studied high-risk populations. Therefore it is difficult to draw conclusions regarding the long-term relationship between adiposity and RBP4 in children. In this context, Goodman et al18 in this issue of The Journal reports longitudinal data over 3 years and addresses the key question of whether RBP4 has a role in modulating insulin resistance and subsequent disease in a very vulnerable population, non-Hispanic black adolescents. In this nested, retrospective study they demonstrated that an increase in RBP4 over 3 years was associated with increased odds of worsening insulin resistance in this population independent of body mass index. Also, it appears from this study that the effect of change in RBP4 on insulin resistance may be dependent on the initial RBP4 level. The use of homeostasis model assessment-insulin resistance (HOMA-IR) as a surrogate measure of insulin resistance in this study may be viewed as a potential limitation. However, even with the use of HOMA-IR, the study demonstrated an association between increases in RBP4 and worsening of insulin resistance over 3 years. It is likely that measurement of insulin sensitivity by more sophisticated, direct methods would have, perhaps, produced better relationships. Indeed assessment of insulin sensitivity in human beings, especially in children, under physiological conditions has been a challenge because of its complex interplay with insulin action and hepatic insulin extraction. The study also showed that both initial RBP4 concentration and change in RBP4 predict change in insulin sensitivity and elevation of triglyceride levels. This is important because of the possibility that baseline RBP4 concentrations may influence subsequent sensitivity to exogenous agents or lifestyle changes. Furthermore, the implication may be that measurement of RBP4 may aid in discriminating overweight children at risk for development of cardiometabolic disease from those may remain resistant to these risks. RBP4, on the other hand, failed to show any relationship with high-density lipoprotein and low-density lipoprotein–cholesterol or fibrinogen concentrations in this study, which is somewhat contrary to few other previous studies.13, 15, 19 The issue of potential relationship between RBP4 with other traditional and nontraditional risk factors for cardiometabolic disease, for example, inflammatory factors, cannot be fully discounted in view of these recent studies.13, 15, 19 Whether RBP4 directly relates to other key components of cardiometabolic disease, such as inflammation or oxidative stress remains to be elucidated.

In conclusion, the causal effects of biomarkers on diseases in human beings are extremely difficult to prove. Research into the effects of RBP4 and perhaps other adipokines on insulin sensitivity is still at an early stage, especially in children. Perhaps, the time is not yet ripe for including RBP4 among the established risk factors for cardiometabolic disease. However, the alarming increase in overweight and obesity in children and the trajectory and reach of related comorbidities such as cardiometabolic disease challenge us to identify potential modifiable pathways that will have sufficient impact to turn these around. In this context, this study is a step in the direction of identifying unique individual risk profiles for obesity related cardiometabolic disease. The direction and the findings of the present study should stimulate further prospective research to address the value of RBP4 and other adipokines as “biomarkers.” Interestingly, although RBP4 was originally characterized by DeWitt Goodman et al, it took more than 4 decades and his daughter, Elizabeth Goodman,18 along with Barbara Kahn's group, to extend the studies on RBP4, to a critically important clinical population, black adolescents. The novel insights from this report by Goodman et al18 are very important, but should not be considered as the end of scientific inquiry in this vast field of adipokines and their potential role in insulin resistance related to cardiometabolic disease in children.