Regulation of energy metabolism and maintenance of metabolic homeostasis: the adiponectin story after 20 years

The current obesity epidemic and its major sequelae, including type 2 diabetes, atherosclerosis, and cardiovascular diseases, threaten human health, productivity, and life quality. Recent studies, which continue to further clarify the molecular basis of these diseases, aim to ultimately provide a path to better diagnoses and therapies for disease prevention and reversal. Insulin resistance is a defining feature of obesity-linked type 2 diabetes and nearly always coexists with reduced plasma adiponectin levels and a constellation of other abnormalities that increase the risks for cardio- and cerebro-vascular diseases.

Adiponectin, identified independently by four groups using different approaches, was originally cloned as an adipocyte-enriched protein highly induced during 3T3-L1 adipocyte differentiation (as reviewed by Wang and Scherer in this issue). Over the past two decades, adiponectin has been widely recognized as an insulin sensitizer implicated in the regulation of energy metabolism and insulin response in major insulin target tissues, and as an adipokine that modulates inflammatory responses. While the salutary effects of adiponectin in metabolism and cardiovascular protection are well established, the relatively high plasma concentration and the presence of multimeric forms of this adipokine have, at least partially, limited its use as a therapeutic agent. Thus, a more comprehensive understanding of the molecular details of adiponectin signaling would provide a guide to therapies that target the root cause(s) of obesity-linked type 2 diabetes and insulin resistance.

First, Wang and Scherer provide a historic perspective on adiponectin research completed during the past 20 years, focusing on major findings as well as key questions on adiponectin and adiponectin-related therapeutic implications. Since the initial report of adiponectin in 1995 (Scherer et al., 1995), a large body of work continues to reveal the physiological actions and underlying molecular mechanisms of this adipokine. Adiponectin suppresses hepatic gluconeogenesis by downregulating genes involved in hepatic glucose production, and promotes fatty acid oxidation in skeletal muscle via AMPK activation, thereby improving the overall in vivo insulin sensitivity. In addition to the liver, adipose tissue, and skeletal muscle, adiponectin also targets a variety of organs, tissues, and cell types, including pancreatic β cells, immune cells, the kidney, heart, and central nervous system.

Identification of the adiponectin receptors (Yamauchi et al., 2003), as well as a recent crystal structure study of the human AdpoR1 and AdipoR2 (Tanabe et al., 2015), provided mechanistic insights toward adiponectin function. Multiple intracellular signaling pathways including the intracellular binding partners APPL1 and APPL2 (Mao et al., 2006) and other effector molecules downstream of AdipoR1 and AdipoR2 have been identified. In the context of whole animal studies, the regulation of adiponectin secretion and multimerization also comes into play. Several molecules, including DsbA-L, Ero1-Lα, and ERp44 (Liu and Liu, 2012), regulate adiponectin multimerization and secretion, thereby modulating the plasma levels of high-molecular weight (HMW) adiponectin, a more active form and potent insulin sensitizer.

As Wang and Scherer pointed out, as a fat-derived hormone, adiponectin fulfills a critical role: adiponectin is a messenger that connects adipose tissue with other organs. The signaling pathways downstream of the adiponectin receptors have emerged as potential therapeutic targets, and the growing understanding of the molecular details of adiponectin action will likely guide the design of adiponectin-based therapies.

The second review by Ruan and Dong discusses tissue- and cell-specific functions of adiponectin, focusing on regulation of adiponectin signaling, and potential crosstalk and connections between the adiponectin and other signaling pathways involved in metabolic regulation. Insulin resistance refers to a state in which physiological concentrations of insulin fail to elicit normal responses in insulin target tissues to meet the metabolic demands of the body. Two series of independent, intriguing and seminal studies have advanced our understanding of the adiponectin–insulin axis and established the liver and skeletal muscle (Berg et al., 2001; Yamauchi et al., 2002) as two major targets of adiponectin.

Adiponectin also elicits other signaling events downstream of its receptors. While no intrinsic protein kinase activity has ever been associated with either AdipoR1 or AdipoR2, APPL1 and APPL2 are the intracellular binding partners of AdipoR1 and AdipoR2 (Mao et al., 2006). APPL1 mediates adiponectin-induced AMPK and p38 MAPK activation, and whole-body knockout of APPL1 impairs adiponectin signaling, resulting in insulin resistance (Ryu et al., 2014). APPL2 negatively regulates adiponectin signaling by inhibiting the interaction of APPL1 with AdipoRs.

As reviewed by Ruan and Dong, multiple cellular pathways, including the AMPK, PPAR, p38 MAPK, CaMKK-β, and ceramide signaling pathways, likely mediate the plethora of adiponectin effects in various tissue sites, depending on cellular contexts. A better understanding of the cellular circuitry mediating adiponectin activity, together with empirical trials, will provide druggable targets for the treatment of obesity-related metabolic diseases.

Next, Hui and colleagues comprehensively review the control of energy metabolism and vascular homeostasis by FGF21–adiponectin axis. While FGF21 is mainly produced by the liver and adiponectin nearly exclusively by adipose tissue, the two hormones share biological activities. Several observations support the role of adiponectin as an obligatory mediator of FGF21 in vivo. These observations support the hypothesis that FGF21 activity depends on adiponectin. Interestingly, FGF21 preferentially targets adipose tissue for its functions, but acts indirectly on the liver likely through circulating factors in vivo. This review summarizes the effects of FGF21 on gene induction, protein expression, post-translational modification, multimerization, and secretion of adiponectin, as well as the relevance of PPAR-γ in the process, and discusses recent research endeavors in the FGF21–adiponectin field, aiming to address several remained uncertainties: (i) what are the mechanism(s) for adiponectin to mediate the biological functions of FGF21? (ii) does FGF21 directly affect on metabolic tissues through its receptors? (iii) does FGF21 have any adiponectin-independent actions? The investigations in the FGF21–adiponectin axis under physiological and pathological conditions provide clinical perspectives on a potential disruption of the FGF21–adiponectin axis in obesity. Thus, the discovery of the FGF21–adiponectin axis offers promising insights toward pharmacological interventions of obesity-related cardio-metabolic disorders.

Finally, Luo and Liu discuss recent findings on the regulation of innate immunity and metabolic stress sensing by adiponectin. A large amount of evidence implicates that innate immune responses are key mediators in the regulation of glucose homeostasis and whole-body energy metabolism. Adiponectin influences the innate immune system, including macrophage proliferation, plasticity, and polarization, innate-like lymphocyte activity, and other innate immune cell functions, and plays a major role in the modulation of inflammation and maintenance of metabolic homeostasis. While adiponectin exerts pro-inflammatory activities in some contexts, it potently suppresses inflammation and promotes macrophage polarization toward the anti-inflammatory M2 phenotype. Perhaps the most nebulous aspect of adiponectin action is its effect on thermogenesis. However, whether this adipokine promotes or inhibits adaptive thermogenesis remains to be further determined. While one study demonstrated that chronic cold exposure induces adiponectin production in subcutaneous white adipose tissue, which is indispensable for subcutaneous adipose beigeing via promoting M2 macrophage proliferation (Hui et al., 2015), another study showed that adiponectin inhibits thermogenesis independently of AdipoR1 and AdipoR2 in adipocytes (Qiao et al., 2014). Further investigations, including clinical studies, are needed to identify the roles of adiponectin in modulating the beigeing process, either through macrophages or through other types of innate immune cells.

The technologies enabling generation of genetically modified mice have revolutionized mechanistic studies of many human diseases including obesity-linked type 2 diabetes. Many observations about the biological functions and mechanisms of action of adiponectin are based on cell and animal models of type 2 diabetes. However, the transfer of models and mechanisms based on animal studies are not axiomatic, and whether the results from experimental animals have direct relevance to human physiology remain unproven; this has slowed the bench to clinic transition. Future efforts, aided by new research tools for rapid and efficient gene manipulation, as well as more appropriate and sophisticated animal models with improved predictive value in the context of human diseases, will improve mechanistic models and identify and validate viable therapeutic targets.

[This work was supported by grants from the National Institutes of Health (R01 grant DK100697 and R01 DK076902), the National Nature Science Foundation of China (81130015), and the National Basic Research Program of China (2014CB910501) to F.L.]