Retinoids and the metabolic syndrome

The metabolic syndrome comprises a number of metabolic disorders, including abdominal obesity, dyslipidemia, hypertension and impaired fasting glucose, that contribute to increased cardiovascular disease and overall mortality. It represents the most rapidly growing health problem in the developed world [101]. A relatively new and growing body of research, which is discussed herein, links these diseases with vitamin A and its metabolites (referred to collectively as retinoids), and to a binding protein that solublizes and transports retinoid within the aqueous environment of the body.

The recent literature raises many fundamental questions regarding whether retinoids play a causal role in metabolic disease development and whether retinoids might have roles in combating disease development. Since retinoids act primarily as transcriptional regulators requiring the actions of their specific nuclear hormone receptors, a major focus of research currently centers on whether retinoid effects are mediated solely via traditional nuclear hormone receptor pathways or, as recent research suggests, via other noncanonical pathways.This provokes the question of what roles retinoid-related proteins and/or retinoid metabolites may have in facilitating or regulating these relationships. Are these interactions uni- or bidirectional, that is, are retinoids playing a role in lipid and carbohydrate metabolism, and is it possible that lipids and sugars are also playing a role in retinoid metabolism? What is the normal role for this metabolic crosstalk in a healthy individual and what goes wrong that can lead to the development of metabolic disease? Ultimately, this new research will provide needed insight into linkages between retinoids and the disorders collectively constituting the metabolic syndrome.

Retinoids are essential for life, from embryonic development through the entirety of the life cycle. Well-established essential retinoid actions include roles in the maintenance of normal cell proliferation and differentiation, normal immune function, normal male and female reproduction and vision. All retinoids enter the diet from one of two sources, either as preformed vitamin A, which is abundant in meat and dairy products, or as provitamin A carotenoids, such as β-carotene, present in fruits and vegetables, which can be metabolized within the body to retinoids [1]. Retinoids act canonically as potent signaling molecules that influence expression of over 500 genes, predominantly via the action of the major transcriptionally active metabolite retinoic acid binding to and activating one of the three retinoic acid receptors (RAR-α, -β and -γ) or three retinoid X receptor (RXR-α, -β and -γ) nuclear hormone receptors. The RXRs, which are activated by 9-cis-retinoic acid and other lipid ligands known collectively as rexinoids, act as heterodimers in concert with a number of other nuclear hormone receptors, such as the peroxisome proliferator-activated receptors (PPARs) and liver X receptors (LXRs). These heterodimerization pairs are involved in the regulation of multiple metabolic pathways and can become activated by retinoids alone or upon binding to ligands derived from cholesterol or fatty acid metabolism or, in some cases both, together [2]. Thus, retinoids acting through RXRs provide an obvious basis for explaining how retinoids influence lipid and carbohydrate homeostasis [2,3]. As the RXRs are almost ubiquitously expressed throughout the body, they are available to heterodimerize with partners in most tissues for inhibiting/activating the transcription of target genes and potentially influencing the development or treatment of metabolic disease [2]. Because of this, there is considerable interest in whether RXR ligands (rexinoids) may have therapeutic value for the treatment or amelioration of metabolic disease. For example, animal model studies have demonstrated that rexinoid treatment of either insulin resistant or diabetic rodents reduced feeding and fat mass, thus suggesting potential therapeutic avenues for treating metabolic syndrome patients with rexinoids [3].

Recent research is highlighting how all-trans-retinoic acid (ATRA) signaling, either in excess or absence, can have marked effects on whole-body metabolism. For example, ATRA is essential for adipocyte differentiation; however, at later stages in this process or at high concentrations, ATRA can inhibit adipogenesis [4]. Mice fed a diet high in ATRA or high in retinol exhibit lower adiposity than controls [4,5]. Other studies in transgenic mice expressing an RAR-α dominant-negative mutation specifically in hepatocytes, thereby ablating ATRA activity and RAR activation, resulted in the development of steatohepatitis (accompanied by a downregulation of hepatic mitochondrial β-oxidation activity) and liver tumors [6]. ATRA has also been shown to have potent effects on whole-body energy homeostasis by increasing the capacity of skeletal muscle for fatty acid oxidation [7]. It is well established at the molecular level that retinoic acid signaling can influence expression of genes involved in fatty acid metabolism, as well as gluconeogenesis [8]. In addition to binding to RARs, ATRA was recently demonstrated to directly activate the PPAR-β/γ isoform [9]. Since the PPAR-β/δ isoform is importantly involved in the regulation of lipid and glucose homeostasis and the development of obesity, diabetes and dyslipidemia, this too may be a potential molecular link between retinoid actions and metabolic disease.

In addition to retinoic acid, a recent report demonstrated that the intermediate to retinoic acid synthesis, retinyaldehyde, can directly inhibit adipogenesis and diet-induced obesity [10]. The mechanism for this inhibition is proposed to involve the suppression of PPAR-γ and RXR responses. However, one might speculate that substrate competition and/or switching on of catabolic genes in response to the excess retinoids could also inadvertently disrupt levels of other lipids required for maintaining normal adipogenesis.

It is well established that retinoids are stored in adipose tissue and that adipose tissue, such as the liver, also secretes serum retinol-binding protein 4 (RBP4) [11,12]. Studies by Kahn and colleagues published in 2005 identified RBP4 as a secreted adipokine, whose secretion is increased in the adipose tissue of the genetically obese ob/ob mice [13]. These investigators further demonstrated that mice lacking RBP4 gene expression had improved insulin tolerance, a response they lost if the RBP4 protein was reinjected into their circulations [14]. Further studies in human subjects demonstrated that RBP4 is preferentially expressed in the visceral fat, which is the fat compartment most associated with the pathogenesis of metabolic syndrome, rather than the subcutaneous fat [15]. Human studies reported by other research groups are in agreement that elevated serum RBP4 levels are associated with insulin resistance, and extended this finding by establishing that the molar ratio of RBP4 to serum retinol in the circulation correlates more strongly with components of the metabolic syndrome than serum RBP4 levels alone [16,17]. However, not all studies are in agreement with these findings regarding RBP4 and insulin responsiveness. Some investigators were unable to detect direct relationships between either adipose mRNA or plasma levels of RBP4 and the degree of adiposity or insulin resistance [18,19]. Much research is presently underway to characterize these associations. Nevertheless, these studies suggest that not only retinoids, but also RBP4, may play direct roles in regulating insulin sensitivity and, therefore, lipid and glucose homeostasis. In addition to transporting retinoids from adipose to peripheral tissue, RBP4’s actions on insulin responsiveness may involve other molecular mechanisms, including direct protein–protein interactions at the cell surface or the involvement of an as yet unidentified ligand that is carried by RBP4 from adipose tissue to the periphery where it modulates insulin signaling.

Provitmain A carotenoids and/or their nonretinoid metabolites have also been suggested to have a role in modulating lipid metabolism. Studies of mutant mice lacking carotenoid-15,15´-monooxygenase (Cmo1^-/- mice), the sole enzyme in the body that converts provitamin A carotenoids to retinoids, indicate that this enzyme plays a previously unsuspected role in modulating hepatic fat metabolism [18]. Cmo1^-/- mice spontaneously develop a fatty liver and display altered serum lipid levels with elevated serum unesterified fatty acids. Feeding high levels of β-carotene to the mice further exasperates the fatty liver phenotype. Additionally, the mice are more susceptible to high fat diet-induced impairments in fatty acid metabolism [20]. Other studies have suggested that nonretinoid cleavage products of carotenoids, the apo-carotenals, which are not converted to retinoids, repress the activation of PPAR and RXR nuclear hormone receptor signaling. Specifically, the apo-14-carotenal cleavage product of β-carotene is proposed to have a role in modulating PPAR and RXR signaling [21]. Collectively, these studies are starting to shed light on the potential crosstalk between carotenoid metabolism and impaired metabolism and storage of lipids.

However, a question remains: why would retinoid and energy metabolism be paired or linked in such a way? Retinoid and lipid metabolism share many common metabolic pathways in the body, although retinoids are present in the body in only micro amounts, whereas lipids are present in macro amounts (at concentrations that differ by 100- to 1000-fold). It is provocative to speculate that the micronutrient vitamin A, in addition to its highly evolved effects on nuclear hormone signaling pathways, may serve as an early sensor for macronutrient metabolism. For our distant ancestors, a varied diet was not the norm. Preformed vitamin A is obtained from a diet of animal food sources, which are rich in lipids and poor in carbohydrates. A provitamin A (carotenoid)-rich diet, however, would be reflective of a more carbohydrate-rich and fat-free diet. It is possible that a metabolic switch evolved to save sugars by inhibiting insulin signaling during times of abundant meat, fat and retinoid consumption, and that, conversely, during times of high vegetable intake, the body was honed to produce more lipids. In the present era, where excess is more frequently the norm, it is likely that these systems, which served life well in the face of limited energy availability, are simply no longer valid in the face of abundance. It now seems likely that a better understanding of the interactions of retinoids and their related proteins with lipid and carbohydrate metabolism will provide new insights that are useful for the development of more focused pharmacological therapies, in addition to a greater understanding of the pathophysiology underlying metabolic disease development.

Financial & competing interests disclosure

SM O’Byrne and WS Blaner are supported by research grants from the NIH (R01 DK068437, R01 DK077221). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.