**4. Proinflammatory cytokines production**

Obesity is a low grade systemic inflammatory state; several studies have addressed adipocyte-related mechanisms involved in regulating proinflammatory [15, 40] and adipocyte-specific cytokines. Systemic chronic inflammation is due to an inflammatory response in adipose tissues that are under quick expansion, and in which macrophages and adipocytes are activated and stimulate the production of proinflammatory cytokines and adipokines [41]. Visceral WAT is considered a main source of inflammation related to obesity. Obese subjects with higher visceral fat exhibit monocyte-chemotactic protein (MCP)-1 expression and infiltration of macrophages in visceral fat compared to subcutaneous fat [42]. Also, several *in vivo* studies have shown higher levels of plasminogen activator inhibitor (PAI)-1, IL-6, and TNF-α in visceral obesity [43, 44].

Recent studies have shown the inhibitory role of estrogen to the inflammatory response in adipose tissue, and cardiovascular and nervous systems [35]. ERα is located in cytokineproducing cells, such as macrophages and microglia, and several *in vitro* studies have reported that E2/ ERα decreases the number of pro-inflammatory cytokines [35, 45, 46]. Another *in vivo* study has demonstrated that both ERα and ERβ regulate proinflammatory cytokine and chemokine production through E2-dependent and independent mechanisms [47].

Because energy expenditure is related to inflammation, a role for nuclear receptor kappa B (NF-κB) has been suggested. NF-κB is a protein complex that controls transcription factor, and is known to have a crucial role in regulating immune responses to infection or stress. NF-κB induces the transcription of inflammatory cytokines, such as TNF-α, IL-1, IL-6, and MCP-1 [41]. In the classical pathway, NF-κB is mediated through IKKβ-induced phosphorylation and proteasome-mediated degradation of IkBα [48]. The other pathway is the activation by hypoxia in the absence of IkBα degradation, a pathway in adipocytes and macrophages which contributes to chronic inflammation in the adipose tissues of obese subjects [41, 49]. Limited evidence has proposed an inhibitory effect of E2 via ERα on NF-κB [50, 51], which partially explains the anti-inflammatory properties of estrogen.

#### **5. Adipokine expression/secretion**

Adipokines are a family of cytokines which include leptin, resistin, adiponectin, and TNF-α and are primarily secreted from adipocytes. These adipokines are released from different tissues and organs, and are not exclusively produced by WAT [14, 52].

Leptin, the product of the *ob* gene, has been shown to be a key metabolic hormone in regulating appetite body weight and energy homeostasis [53, 54]. In humans, circulating leptin levels have a parallel correlation with the amount of body fat. In addition, serum levels of leptin have been shown to be higher in women than men [55]. This finding has been hypothesized to be due to a different pattern in fat deposition between the genders and/or the effects of a different hormonal milieu [56]. With respect to the change in leptin levels according to menopausal status, data are inconsistent; some data have shown no change in leptin levels between the pre- and post-menopause [57], while other studies have demonstrated a decrease in leptin levels during the menopausal transition [58, 59]. Some investigators have suggested that the low amount of visceral fat compared with subcutaneous fat in the genders [60, 61] makes it unlikely that the leptin secretion rates from these two depots differ, and may therefore be the cause for the sexual dimorphism in leptin concentrations, suggesting an important role for sex steroid hormones [53]. According to

Obesity is a low grade systemic inflammatory state; several studies have addressed adipocyte-related mechanisms involved in regulating proinflammatory [15, 40] and adipocyte-specific cytokines. Systemic chronic inflammation is due to an inflammatory response in adipose tissues that are under quick expansion, and in which macrophages and adipocytes are activated and stimulate the production of proinflammatory cytokines and adipokines [41]. Visceral WAT is considered a main source of inflammation related to obesity. Obese subjects with higher visceral fat exhibit monocyte-chemotactic protein (MCP)-1 expression and infiltration of macrophages in visceral fat compared to subcutaneous fat [42]. Also, several *in vivo* studies have shown higher levels of plasminogen

Recent studies have shown the inhibitory role of estrogen to the inflammatory response in adipose tissue, and cardiovascular and nervous systems [35]. ERα is located in cytokineproducing cells, such as macrophages and microglia, and several *in vitro* studies have reported that E2/ ERα decreases the number of pro-inflammatory cytokines [35, 45, 46]. Another *in vivo* study has demonstrated that both ERα and ERβ regulate proinflammatory cytokine and

Because energy expenditure is related to inflammation, a role for nuclear receptor kappa B (NF-κB) has been suggested. NF-κB is a protein complex that controls transcription factor, and is known to have a crucial role in regulating immune responses to infection or stress. NF-κB induces the transcription of inflammatory cytokines, such as TNF-α, IL-1, IL-6, and MCP-1 [41]. In the classical pathway, NF-κB is mediated through IKKβ-induced phosphorylation and proteasome-mediated degradation of IkBα [48]. The other pathway is the activation by hypoxia in the absence of IkBα degradation, a pathway in adipocytes and macrophages which contributes to chronic inflammation in the adipose tissues of obese subjects [41, 49]. Limited evidence has proposed an inhibitory effect of E2 via ERα on NF-κB

Adipokines are a family of cytokines which include leptin, resistin, adiponectin, and TNF-α and are primarily secreted from adipocytes. These adipokines are released from different

Leptin, the product of the *ob* gene, has been shown to be a key metabolic hormone in regulating appetite body weight and energy homeostasis [53, 54]. In humans, circulating leptin levels have a parallel correlation with the amount of body fat. In addition, serum levels of leptin have been shown to be higher in women than men [55]. This finding has been hypothesized to be due to a different pattern in fat deposition between the genders and/or the effects of a different hormonal milieu [56]. With respect to the change in leptin levels according to menopausal status, data are inconsistent; some data have shown no change in leptin levels between the pre- and post-menopause [57], while other studies have demonstrated a decrease in leptin levels during the menopausal transition [58, 59]. Some investigators have suggested that the low amount of visceral fat compared with subcutaneous fat in the genders [60, 61] makes it unlikely that the leptin secretion rates from these two depots differ, and may therefore be the cause for the sexual dimorphism in leptin concentrations, suggesting an important role for sex steroid hormones [53]. According to

activator inhibitor (PAI)-1, IL-6, and TNF-α in visceral obesity [43, 44].

chemokine production through E2-dependent and independent mechanisms [47].

[50, 51], which partially explains the anti-inflammatory properties of estrogen.

tissues and organs, and are not exclusively produced by WAT [14, 52].

**5. Adipokine expression/secretion** 

**4. Proinflammatory cytokines production** 

recent data derived from healthy pre- and post-menopausal women, postmenopausal women had increased levels of tissue plasminogen activator antigen (tPA), MCP-1, and adiponectin [24]. Furthermore, an increase in intraabdominal fat was correlated with Creactive protein, tPA, and leptin, and negatively with adiponectin levels. The results imply that during the menopausal transition, women have adverse changes in inflammatory markers and adipokines which correlate with increased visceral obesity.

Several animal studies have indicated a stimulatory effect of estrogen on leptin expression and secretion in rat adipose tissues [59, 62]. Machinal et al. reported that there are regional differences in leptin expression between subcutaneous and deep fat tissues in rats, and that leptin secretion increased in the deep fat tissue tissues [63]. Based on one study involving human adipocytes, estrogen is likely to stimulate leptin expression [53]. In a previous study, the association between ER and adipokine expression in 3T3-L1 adipocytes was investigated [64]. The results showed that ERα has a stimulatory effect on leptin expression, while the expression of ERβ is inversely correlated with leptin expression. Therefore, discordant findings regarding the estrogen effect on leptin expression/secretion from other studies can be explained by the different expression of the two ER subtypes, which have opposite actions on the expression of leptin. In addition, if there are regulating factors (genetic or environmental) for ERα or ERβ expression in adipocytes, this may explain how the different expression of ER in adipose tissue can affect the diverse obesity phenotypes.

Adiponectin is a 30 kDa protein secreted abundantly from adipocytes [65, 66], and functions to exert anti-diabetic and anti-atherogenic properties. The serum adiponectin levels are decreased in obese individuals, metabolic syndrome, and type 2 DM [67, 68]. Similar to leptin, circulating adiponectin levels show a sexual dimorphism with higher levels in women than men [69]. Although the estrogen effects for adiponectin expression are limited, some *in vitro* studies have reported no direct regulatory effect in humans and mouse adipocytes [64, 70].

Resistin is a cysteine-rich protein that was originally described as an adipose-derived protein in rodents that links obesity to insulin resistance [71]. However, in human data, the relationship of resistin with obesity and insulin metabolism is still debated. A widely reported biological role of resistin is the regulatory effect involved in inflammatory processes. In addition to adipocytes, resistin is induced by lipopolysaccharides and TNF-α in macrophages [72]. The other data showed that resistin induces or is induced by IL-6 and TNF-α via NF-κB in human monocytes [73, 74]. Data on estrogen effects for resistin expression are limited. In a mouse adipocyte model, the regulatory effect of 17β-E2 on resistin expression is discordant [64, 75].

Recently described novel adipokines, such as visfatin, retinol binding protein-4, and omentin, have been shown to exert some metabolic properties, but their biological actions linked to obesity and interactions with estrogen need to be elucidated.
