Reduced Slc2a4/GLUT4 expression in subcutaneous adipose tissue of monosodium glutamate obese mice is recovered after atorvastatin treatment

Background Decreased expression of glucose transporter protein GLUT4, encoded by the solute carrier 2A4 (Slc2a4) gene, is involved in obesity-induced insulin resistance. Local tissue inflammation, by nuclear factor-κB (NFκB)-mediated pathway, has been related to Slc2a4 repression; a mechanism that could be modulated by statins. Using a model of obesity with insulin resistance, this study investigated whether (1) inflammatory markers and Slc2a4 expression are altered; (2) atorvastatin has beneficial effects on inflammation and Slc2a4 expression; and (3) inhibitor of NFκB (IKK)/NFκB pathway is involved in subcutaneous adipose tissue (SAT). Findings Obese mice showed insulin resistance, decreased expression of Slc2a4 mRNA (66%, P < 0.01) and GLUT4 protein (30%, P < 0.05), and increased expression of interleukin 6 (Il6) mRNA (44%, P < 0.05) in SAT. Obese mice treated with atorvastatin had enhanced in vivo insulin sensitivity, besides increased Slc2a4/GLUT4 expression and reduced Il6 expression in SAT. No alterations of tumor necrosis factor-α, interleukin 1β and adiponectin expression or IKKα/β activity in SAT of obese mice or obese mice treated with atorvastatin were observed. Conclusions Atorvastatin has beneficial effect upon glycemic homeostasis, which may be related to its positive impact on Il6 and Slc2a4/GLUT4 expression in SAT.


Introduction
The glucose transporter protein GLUT4 is responsible for insulin-mediated glucose uptake in adipose tissue and skeletal muscle, and plays an important role in glycemic homeostasis [1]. In adipocytes, several transcriptional factors regulate Slc2a4 gene, which encodes for GLUT4 protein, including the nuclear factor-κB (NFκB) [2][3][4].
Obesity is associated with insulin resistance and reduced Slc2a4/GLUT4 expression in both muscle and fat [2,5]. On the other hand, studies on transgenic mice have pointed out that overexpression of GLUT4 in fat enhances in vivo glucose tolerance and insulin sensitivity [6]. Furthermore, since the identification of proinflammatory cytokine TNF-α in fat [7] and its relationship to insulin resistance [8], obesity has been closely related to a low grade chronic inflammatory state.
Several studies have demonstrated that statins exert pleiotropic actions besides cholesterol lowering. Recently, some studies have reported evidences for antiinflammatory and insulin sensitizing effects of statin in visceral adipose tissue (VAT) of glutamate monosodiuminduced obese mice [2,9]. In addition, there are evidences that IKK/NFκB pathway is involved in the Slc2a4 gene expression in VAT [2].
Not only VAT but also subcutaneous adipose tissue (SAT) have been associated to insulin resistance [10]. Considering that SAT is responsible for most of systemic free fatty acids, which are known to induce peripheral insulin resistance, SAT could have more impact on insulin resistance than VAT [11]. Additionally, it is important to understand molecular differences between VAT and SAT, because these fat depots have different biological properties, regarding glucose and fat cell metabolism [12,13]. The present study addresses the effects of atorvastatin on glucose disposal and inflammation focusing on SAT. For that purpose, we used a previous tested model of obesity and insulin resistance, and an anti-inflammatory and insulin sensitizing treatment [2,5].

Animals treatments
Obesity induction in male offspring mice (CD1) was carried out by subcutaneous injections of MSG (2 mg/g body weight) from days one to five, and on day seven after birth [2]. Control mice were injected with saline solution. Animals were weaned and allowed free access to standard rodent chow and water ad libitum until 19 weeks of age, when atorvastatin treatment of obese mice started. Atorvastatin was given in chow (0.1% w/w) for four weeks. At the end of treatment (23-week-old mice) the obesity degree was estimated by Lee obesity index [body weight (g) 1/3 /nasoanal length (cm)], and subcutaneous adipose tissue (SAT) from abdominal, lateral and dorsal regions, as well as blood samples were collected under anesthesia (50 mg/kg b.w. sodium pentobarbital, i.p.). Blood samples for glucose, insulin, cholesterol, AST and ALT quantification were collected from retro-ocular bleeding. For ITT experiments, another group of animals was used and blood samples were collected from the tail vein. Mice were euthanized with an overdose of sodium pentobarbital. All procedures were approved by the Ethical Committee for Animal Research of the Institute of Biomedical Sciences, University of São Paulo (123/2005).

Plasma analysis
Levels of plasma glucose, insulin, cholesterol, activity of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured as previously described [14,15] after 4-hour restricted feeding.
Insulin sensitivity was analyzed by measuring the glucose disappearance constant (kITT) during the intravenous insulin tolerance test (IVITT) [14].
GLUT4 protein and Slc2a4 mRNA analysis GLUT4 was analyzed in SAT by Western blotting and Slc2a4 mRNA by real-time PCR as previously described [4]. For Western blotting, protein-loaded control was checked by analyzing post-transferring Coomassie-stained gels [16]. For real-time PCR, several housekeeping genes, such as mouse Rplp0 (36B4), Gapdh and Actb were tested. Rplp0 was used for normalization. The primer sequences of the following genes, mouse Slc2a4 [GenBank: SAT, subcutanous adipose tissue, VAT, visceral adipose tissue, kITT, glucose disappearance constant obtained in the intravenous insulin tolerance test. Data from plasma cholesterol, glucose and insulin concentration and kITT were obtained from mice subjected to 4-hour food deprivation. Data are means S.E. of 4 to 6 (morphological parameters), and 3 to 7 (metabolic hormonal parameters) animals. *P < 0.05, **P < 0.01 and ***P < 0.001 vs. CTL; #P < 0.05 and ##P < 0.01 vs. OB, One-way analysis of variance and Student-Newman-Keuls post hoc test.

Statistical analysis
All data are expressed as means ± S.E. Comparison of the means were performed by one-way analysis of variance (ANOVA), with Student-Newman-Keuls as a post hoc test.

Results
Although untreated obese mice (OB) and atorvastatintreated obese mice (OBA) showed similar body weight to that observed in control mice (CTL), the significant higher Lee obesity index indicated that both groups were obese (Table 1). Besides, both OB and OBA showed augmentation of absolute (293% and 241% vs CTL, P < 0.001, respectively) and relative (171% and 132% vs CTL, P < 0.01, respectively) SAT weight when compared to CTL (Table 1). OB presented hyperinsulinemia and decreased glucose disappearance constant (38% vs CTL, P <0.05) in IVITT, depicting the whole body insulin-resistant condition (Table 1). Moreover, as evident in Figure 1 reduced (66% and 30% vs CTL, P < 0.01 and P < 0.05, respectively), indicating the participation of this territory in the whole body glycemic homeostasis. Interestingly, atorvastatin treatment did not only reduce plasma cholesterol (Table 1), but also had a positive impact on glucose metabolism, reducing the level of insulinemia and increasing the insulin sensitivity of OBA as measured by IVTT (Table 1). Atorvastatin also restored the expression of Slc2a4 mRNA and GLUT4 protein in SAT of obese mice (Figure 1). Additionally, the drug did not alter the plasma activity of AST and ALT, indicating no apparent hepatotoxicity (Table 1).
In order to verify the effect of obesity and atorvastatin treatment upon inflammation in SAT, the expression of some cytokines were investigated. OB showed increased Il6 mRNA expression in SAT (44% vs CTL, P < 0.05), but unchanged Tnf, Il1b or Adipoq gene expression (Figure 2). Moreover, atorvastatin treatment was able to drastically reduce Il6 mRNA expression (44% vs OB, P < 0.05) in SAT (Figure 2).
The IKK/NFκB pathway was accessed in SAT. Total IKK-β content and phosphorylated IKK-α/β in SAT of OB and OBA were unaltered when compared to CTL (Figure 3).

Discussion
The current study demonstrated that atorvastatin treatment restores GLUT4 protein and mRNA expression in SAT of OB, contributing to the amelioration of wholebody insulin resistance.
It has been reported that atorvastatin has beneficial impacts on inflammation and glucose metabolism. Our laboratory recently demonstrated that OBA had reduced circulating cytokines and reduced cytokines expression in VAT, which resulted in insulin sensitizing effects [2]. SAT and VAT show distinct physiological characteristics such as cytokine expression [17] and insulin signaling [18]. Therefore, by using an established model of obesity and insulin resistance, and an anti-inflammatory treatment (atorvastatin), this study analyzed the SAT of obese mice.
Many reports have shown that reduced GLUT4 expression is related to insulin resistance, and improvement of GLUT4 content is related to increased insulin sensitivity [2, 5,19]. Moreover, mice with adipose-selective reduction of GLUT4 have a striking reduction in glucose uptake by adipocytes [20]. In contrast, overexpression of GLUT4 in adipose tissue increases glucose influx into adipocytes [6]. The present data clearly show that obese mice developed insulin resistance which was accompanied by reduced Slc2a4 mRNA and protein content in SAT. It has been reported that Slc2a4 mRNA [21] and protein [22] expression is reduced in SAT of obese women. Regarding SAT of obese mice, this is the first report investigating Slc2a4 expression. It has been reported that OB mice present reduced Slc2a4 mRNA and GLUT4 protein expression in VAT [2,5]. Therefore, it can be assumed that the reduction of Slc2a4/GLUT4 expression in both SAT and VAT contributes to the whole-body insulin resistance in this animal model of obesity. On the other hand, atorvastatin was able to restore Slc2a4 mRNA and protein content not only in VAT [2], but also in SAT of OB, contributing for amelioration of insulin resistance. Obesity is considered as a low grade chronic inflammatory state [2,23]. OB showed increased Il6 expression in SAT, but showed no alteration in Tnf, Il1b and Adipoq expression. The literature concerning the impact of obesity on Tnf expression in mice SAT is contentious [24,25]. As to Il1b expression in SAT, a single study shows reduction in obese mice [25]. Furthermore, few reports have shown Il6 increase in VAT of obese mice [2,17], but to date there has been no report about obesity impact on Il6 expression in SAT of mice. Taking into account that IL-6 can reduce Slc2a4/GLUT4 expression in adipocytes [26], the present findings point out that IL-6 may be an important repressor of the Slc2a4 gene in SAT but the involved mechanisms still needs clarification.
Finally, we are, for the first time, reporting that IKK content and phosphorylation are not altered in SAT of OB. IKK/NFκB negatively modulates the expression of Slc2a4 gene [2][3][4]27], and we have demonstrated that the reduced expression of Slc2a4/GLUT4 in VAT of OB correlates to enhanced IKK phosphorylation [2]. However, in SAT, the present findings could not confirm the involvement of this inflammatory pathway in the modulation of Slc2a4 gene.
In conclusion, Slc2a4 expression in SAT is reduced in OB, which contributes to the impairment of glycemic homeostasis. Atorvastatin treatment improves insulin resistance which may be related to its positive impact on Il6 and Slc2a4/GLUT4 expression in SAT.

Competing interests
The authors declare that they have no competing interests.