Peroxisome proliferator-activated receptors (PPARs) are a group of nuclear receptor superfamily which functions as transcription factors regulating gene expression and play important roles in the regulation of cellular differentiation, development, and energy metabolism (Schoonjans et al., 1996). The three types of PPAR (α, γ, and β/δ) have been identified (Schoonjans et al., 1996). The PPAR-γ agonists, thiazolidinediones (TZDs) are widely used for anti-diabetic drugs that improve insulin sensitivity through enhancement of glucose disposal as well as reduction of gluconeogenesis in the target tissues of the body including skeletal muscle, liver, and adipose tissue (Furnsinn and Waldhausl, 2002; Kintscher and Law, 2005). A number of studies have shown that TZDs such as rosiglitazone and pioglitazone increased circulating levels of adiponectin in both human and experimental rodent models (Choi et al., 2005; Iwaki et al., 2003; Kubota et al., 2006; Pajvani et al., 2004). In addition to PPAR-γ, the PPAR-α agonist induced the increase in the circulating level of adiponectin associated with improvement in insulin sensitivity. For example, fenofibrate an agonist of nuclear receptor PPAR-α, increased serum levels of adiponectin in patients with primary hypertriglyceridemia (Koh et al., 2005). Previous studies have implicated that hypoadiponectinemia is associated with hypertension (Adamczak et al., 2003; Papadopoulos et al., 2009). Therefore, it is tempting to speculate whether anti-hypertensive drugs such as candesartan and losartan (angiotensin II receptor antagonists) increase adiponectin. These drugs, indeed, elevated circulating adiponectin without altering adiposity (Celik et al., 2006; Furuhashi et al., 2003; Koh et al., 2004; Koh et al., 2006). In addition, several other drugs for anti-diabetic (glimepiride) and anti-hypertension (nebivolol, β₁ receptor blocker) have been shown to enhance plasma adiponectin concentrations in human subjects (Celik et al., 2006; Nagasaka et al., 2003). However, it is unclear whether elevated adiponectin is associated with improved cardiovascular outcomes.

In addition to pharmacological agents, dietary fish oils (FO) and polyunsaturated fatty acids (PUFA) have been shown to increase mRNA expression of adiponectin in adipose tissue and circulating levels of adiponectin in several experimental models and human (Mostowik et al., 2013; Neschen et al., 2006; Rossi et al., 2005). Furthermore, Oolong tea, green tea extract and (-)-catechin increased plasma adiponectin in humans and rodent models (Cho et al., 2007; Li et al., 2006; Shimada et al., 2004). Table 1 summarized the effects of pharmacological agents and dietary intervention on the expression of adiponectin.

It is well documented that exercise or regular physical activity has beneficial effects on metabolic and cardiovascular disease. Considering previous literatures, it is unclear whether exercise training (physical activity) increases adiponectin in circulation and its receptors in insulin-sensitive tissues such as adipose tissue, liver, and skeletal muscle. Complicating interpretation of the existing data is dependent on multiple factors including species, the pathological condition, types (endurance vs resistance exercise), intensity (low, moderate, and intense), and duration of exercise (acute vs chronic, short-term vs long-term), and sex. For example, in healthy, young subjects, it seemed that both acute and chronic aerobic exercise did not alter plasma level of adiponectin (Ferguson et al., 2004; Hulver et al., 2002; Jurimae et al., 2006; Punyadeera et al., 2005). However, chronic endurance training increased plasma adiponectin in obese adolescents (Balagopal et al., 2005), obese adults (Kondo et al., 2006), Caucasian subjects with impaired glucose tolerance (IGT) and T2D (Bluher et al., 2006; Oberbach et al., 2006). Furthermore, endurance training increased mRNA expression of adiponectin receptor (AdipoR) 1 and 2 in adipose tissue and skeletal muscle in normal glucose tolerance (NGT), IGT, and type 2 diabetic patients (Bluher et al., 2006; Bluher et al., 2007; Oberbach et al., 2006). On the other hand, some studies by several other groups have shown that aerobic exercise did not change adiponectin expression in obese subject (Polak et al., 2006), insulin resistant female subjects (Marcell et al., 2005), and patients with T2D (Boudou et al., 2003; Yokoyama et al., 2004). Interestingly, Fatouros et al. have demonstrated that only moderate-high intensity resistance training, not low intensity, increased plasma adiponectin in inactive subjects, suggesting that the intensity of exercise may be an important factor in the expression of adiponectin (Fatouros et al., 2005). Table 2 shows a summary of studies examining effects of exercise training on adiponectin and AdipoRs in both human and experimental models.

Calorie restriction (CR) refers to a dietary regimen low in calories without malnutrition and is known as an efficient lifestyle modification that delays the onset of metabolic and cardiovascular disease (Cava and Fontana, 2013). Weight loss and/or CR have been shown to improve insulin resistance, T2D, and cardiovascular dysfunction in both human and rodent models (Weiss and Fontana, 2011). In addition, sustained weight loss by gastric bypass surgery ameliorated cardiovascular dysfunction (Brethauer et al., 2011; Zhang et al., 2011). Although it is not clear whether beneficial effects of these interventions are mediated by adiponectin signaling pathways, a number of studies have shown that CR and/or weight loss by gastric bypass surgery increased circulating levels of adiponectin. For example, CR increased circulating adiponectin in normal (Fontana et al., 2010; Schulte et al., 2013) and obese subjects (Kobayashi et al., 2004; Oberhauser et al., 2012; Varady et al., 2010). However, some studies have shown that CR did not change plasma levels of adiponectin in patients with metabolic syndrome (Xydakis et al., 2004) and T2D (Plum et al., 2011). Interestingly, Varady et al. (2009) have implicated that circulating adiponectin concentration increased 20% in the 5–10% weight loss group, not less than 5% weight loss group, suggesting a minimum degree of weight loss are required to improve adipokine profile in severely obese women. On the other hand, Plum et al. suggested that Roux-en-Y gastric bypass surgery, not low calorie diet group, increased plasma adiponectin concentration in patients with T2D, although weight loss was comparable in both groups (Plum et al., 2011). This may suggest that Roux-en-Y gastric bypass surgery is more effective method than low calorie diet regimen in some kinds of obese diabetic patients. Because metabolic and cardiovascular diseases are multi-factorial phenomena, we need to consider the effect of other metabolic disorders such as dyslipidemia, hypercholesterolemia, and hypertension on the expression of adiponectin. Table 3 summarizes the effects of CR, weight loss and gastric bypass surgery on adiponectin.