Selective Modulation of Wnt Ligands and Their Receptors in Adipose Tissue by Chronic Hyperadiponectinemia

Background Adiponectin-transgenic mice had many small adipocytes in both subcutaneous and visceral adipose tissues, and showed higher sensitivity to insulin, longer life span, and reduced chronic inflammation. We hypothesized that adiponectin regulates Wnt signaling in adipocytes and thereby modulates adipocyte proliferation and chronic inflammation in adipose tissue. Materials and Methods We examined the expression of all Wnt ligands and their receptors and the activity of Wnt signaling pathways in visceral adipose tissue from wild-type mice and two lines of adiponectin-transgenic mice. The effects of adiponectin were also investigated in cultured 3T3-L1 cells. Results The Wnt5b, Wnt6, Frizzled 6 (Fzd6), and Fzd9 genes were up-regulated in both lines of transgenic mice, whereas Wnt1, Wnt2, Wnt5a, Wnt9b, Wnt10b, Wnt11, Fzd1, Fzd2, Fzd4, Fzd7, and the Fzd coreceptor low-density-lipoprotein receptor-related protein 6 (Lrp6) were reduced. There was no difference in total β-catenin levels in whole-cell extracts, non-phospho-β-catenin levels in nuclear extracts, or mRNA levels of β-catenin target genes, indicating that hyperadiponectinemia did not affect canonical Wnt signaling. In contrast, phosphorylated calcium/calmodulin-dependent kinase II (p-CaMKII) and phosphorylated Jun N-terminal kinase (p-JNK) were markedly reduced in adipose tissue from the transgenic mice. The adipose tissue of the transgenic mice consisted of many small cells and had increased expression of adiponectin, whereas cyclooxygenase-2 expression was reduced. Wnt5b expression was elevated in preadipocytes of the transgenic mice and decreased in diet-induced obese mice, suggesting a role in adipocyte differentiation. Some Wnt genes, Fzd genes, and p-CaMKII protein were down-regulated in 3T3-L1 cells cultured with a high concentration of adiponectin. Conclusion Chronic hyperadiponectinemia selectively modulated the expression of Wnt ligands, Fzd receptors and LRP coreceptors accompanied by the inhibition of the Wnt/Ca2+ and JNK signaling pathways, which may be involved in the altered adipocyte cellularity, endogenous adiponectin production, and anti-inflammatory action induced by hyperadiponectinemia.


Introduction
Visceral adipose tissue in metabolic syndrome is histologically characterized by enlargement of adipocytes due to impaired adipocyte differentiation, accompanied by chronic low-grade inflammation. The enlarged adipocytes release more free fatty acids, glycerol, and proinflammatory cytokines and less adiponectin. Hypoadiponectinemia and chronic inflammation in adipose tissue are closely associated with obesity-linked complications, including type 2 diabetes, coronary heart disease, and nonalcoholic fatty liver disease.
Previously, we established transgenic mouse lines that express full-length human adiponectin in the liver [1]. The hyperadiponectinemic mice show higher sensitivity to insulin, longer life span and resistance to the deleterious effects of a high-fat/high-sugar diet. The high-calorie diet-induced increase in urinary 8-hydroxy-2-deoxyguanosine, a marker of oxidative DNA damage, is markedly suppressed in the transgenic mice. Interestingly, adipocytes of the transgenic mice are reduced in size and increased in number compared with those of wild-type mice in both subcutaneous and visceral adipose tissues, suggesting that adiponectin may play a role in the regulation of adipogenesis. However, the mechanism of adiponectin action in adipose tissue has not been elucidated.
To test the hypothesis that adiponectin regulates Wnt signaling in adipocytes and thereby modulates adipocyte proliferation and chronic inflammation in adipose tissue, we examined the expression of all Wnt ligands and their receptors and the activities of both canonical and non-canonical Wnt signaling in visceral adipose tissue from wild-type mice and adiponectin-transgenic mice.

Animals
We have established three lines of transgenic mice expressing human full-length adiponectin under the control of the serum amyloid P component promoter on the C57BL/6 background [1]. In this study, we used male transgenic mice of line 11 and line 13, which exhibited a high plasma concentration of human adiponectin (284.1644.4 and 183.0632.0 mg/ml, respectively, including the high-molecular-weight isoform) [1]. The high plasma adiponectin enabled us to analyze adiponectin actions in adipose tissue, where adiponectin concentration is assumed to be higher than in peripheral blood. Wild-type C57BL/6 mice originally purchased from Nippon CLEA (Shizuoka, Japan) and maintained in our laboratory served as controls. Plasma adiponectin was 41.965.4 mg/ml in wild-type mice [1]. Mice were fed standard mouse chow (347 kcal/100 g, protein 24.9 g/100 g, fat 4.6 g/ 100 g; Nippon CLEA) and water ad libitum and sacrificed at the age of 20 weeks. To study diet-induced obesity, wild-type mice were maintained on a high-fat/high-sucrose food (592 kcal/100 g, 70% fat, 14% sucrose, 3% other carbohydrates, and 13% protein, Oriental Yeast, Tokyo, Japan) from the age of 8 weeks. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All procedures were approved by the Ethics Review Committee for Animal Experimentation of Kurume University School of Medicine (Permit Number: 2262). Mice were sacrificed under ether anesthesia, and all efforts were made to minimize suffering.

Quantitative Real-time RT-PCR
The mRNA expression of Wnt ligands, Wnt receptors, coreceptors, and components of Wnt signaling pathways was assessed by quantitative real-time RT-PCR. Visceral adipose tissue and quadriceps femoris muscle were obtained from 20-week-old male mice. Preadipocytes and mature adipocytes were isolated by the collagenase digestion method [20]. RNA was isolated using RNA-Bee (Cosmo Bio, Tokyo, Japan), and 5 mg of total RNA was reverse-transcribed to cDNA using a kit from Invitrogen (Carlsbad, CA, USA). SYBR green-based real-time quantitative PCR of cDNA templates was performed using StepOnePlus (Applied Biosystems, Foster City, CA, USA). The PCR cycling conditions were 10 min at 95uC followed by 40 cycles of 30 sec at 95uC, 30 sec at 53-64uC, and 30 sec at 72uC. The results were calculated as the expression of the target gene relative to the expression of the glyceraldehyde-3-phosphate dehydrogenase (Gapdh) gene. Forward and reverse primers are shown in Table 1. The expression of Fabp4 and Pparg was measured by quantitative RT-PCR as markers of mature adipocytes.

Western Blot Analysis
Adipose tissue and 3T3-L1 cells were lysed in ice-cold lysis buffer containing 1 mmol/l dithiothreitol DTT, 0.0025% NP40 and a cocktail of proteinase inhibitors. The lysate was centrifuged at 19,000 g for 15 min at 4uC, and the supernatant was collected as whole-cell extract. To obtain nuclear extract, adipose tissue lysate was centrifuged at 600 g for 15 min at 4uC, and the pellet was solubilized in nuclear lysis buffer containing 0.5 mmol/l DTT, 20% glycerol, 0.2 mmol/l EDTA and protease inhibitors. After centrifugation at 12,000 g for 10 min, the supernatant was collected. The total protein concentrations of the whole-cell and nuclear extracts were measured using the Bradford reagent (Bio Rad, Hercules, CA, USA). After being heated at 100uC for 5 min, 20 mg total protein was loaded into each well, separated by 7.5% SDS-PAGE (Wako, Osaka, Japan) and transferred to a nitrocellulose membrane. The membrane was incubated with rabbit polyclonal antibodies against b-catenin and non-phospho bcatenin, rabbit polyclonal antibody against CaMKII, rabbit polyclonal antibody against phospho-CaMKII Thr286 (p-CaM-KII), rabbit monoclonal antibody against JNK, or rabbit monoclonal antibody against phospho-JNK Thr183/Thr185 (Cell Signaling Technology, Danvers, MA, USA) at 4uC overnight. After being washed, the membrane was incubated with peroxidase-conjugated goat anti-rabbit IgG (Wako) and then visualized using an ECL system (GE Healthcare, Buckinghamshire, UK).

3T3-L1 Cell Culture
3T3-L1 cells were purchased from DS Pharma Biomedical (Osaka, Japan) and used when the cells were nearly confluent. The cells were cultured with 50, 150, or 300 mg/ml of recombinant human adiponectin (Biovendor, Candler, NC, USA) for 24 h. Thereafter, RNA was isolated from the cells, and the mRNA levels of Wnt-related genes were assessed by quantitative real-time RT-PCR as described above. To assess the effect of adiponectin on the non-canonical Wnt signaling pathways, 3T3-L1 cells were cultured with 50, 150, or 300 mg/ml of recombinant human adiponectin, 1 mg/ml of recombinant human Wnt5a (R&D Systems, Minneapolis, MN, USA) or both for 24 h. Wnt5A was shown to be biologically active at the concentration [21], [22]. Total and phosphorylated JNK and CaMKII were analyzed by Western blotting.

Statistical Analysis
Quantitative results are expressed as means and SD. Statistical analysis was performed using the unpaired Student's t test for comparison between two groups after confirming the normality of the data distribution. P values less than 0.05 were considered statistically significant.

Expression of Wnt Receptors and Coreceptors
Next, we analyzed the expression of Wnt receptor genes. Nine out of 10 known Fzd genes were expressed in adipose tissue (Fig. 1B). The expression of Fzd6 and Fzd9 was significantly increased in both lines of adiponectin-transgenic mice, while that of Fzd1, Fzd2, Fzd4 and Fzd7 was reduced compared with wildtype mice. However, the expression of Ror1, Ror2 and Ryk was not different between the transgenic mice and wild-type mice (Fig. 1C). We further examined the expression of Lrp5 and Lrp6 coreceptor genes and found that Lrp6 mRNA was markedly decreased in adipose tissue from the transgenic mice (Fig. 1D).

Analysis of Canonical Wnt Signaling Pathway
To assess canonical Wnt signaling in adipose tissue, b-catenin was measured by Western blot analysis. There was no difference in total b-catenin levels in whole-cell extracts or in non-phospho-bcatenin levels in nuclear extracts between wild-type mice and the transgenic mice ( Fig. 2A, B). Furthermore, wild-type and transgenic mice showed equivalent mRNA levels of b-catenin target genes c-Myc (Myc) and cyclin D1 (Ccnd1) (Fig. 2C).

Analysis of Non-canonical Wnt Signaling Pathways
The expression of RhoA in adipose tissue was not different between the transgenic mice and wild-type mice (Fig. 3A). The protein level of CaMKII, another non-b-catenin-dependent Wnt pathway signaling molecule, was not altered in either line of transgenic mice. However, p-CaMKII, the active form of CaMKII, was significantly reduced in adipose tissue from the transgenic mice (Fig. 3B, C). Similarly, p-JNK was markedly down-regulated in both lines of transgenic mice, while total JNK was comparable between the transgenic mice and wild-type mice (Fig. 3D, E).

Expression of Endogenous Adiponectin and Cyclooxygenase-2
The expression of the endogenous mouse adiponectin gene was increased in the adiponectin-transgenic mice compared with the wild-type mice (Fig. 4A). The hyperadiponectinemic mice of both lines exhibited significantly lower cyclooxygenase-2 mRNA in adipose tissue (Fig. 4B).

Association of Wnt Ligand Gene Expression with Adipocyte Differentiation and Diet-induced Obesity
The expression of adipocyte marker genes, Fabp4 and Pparg, was higher in the mature adipocyte fraction than in the preadipocyte fraction confirming that two cell populations were isolated (Fig. 5A). The elevation of Wnt5b expression was observed in both preadipocyte and mature adipocyte fractions from adiponectin-transgenic mice. However, the reduction of Wnt5a mRNA and the increase of Wnt6 mRNA were observed only in the mature adipocyte fraction (Fig. 5B). The expression of Wnt5b was significantly reduced in adipose tissue from obese mice maintained on the high-fat/high-sucrose food for 12 weeks (Fig. 5C) compared with wild-type mice fed normal mouse chow (body weight 47.860.9 vs. 32.462.0 g, respectively, n = 4 each, p,0.0001). Cyclooxygenase-2 was elevated in the adipose tissue of the mice fed the high-energy food (Fig. 5D).

Effects of Adiponectin on Wnt Signaling Components in 3T3-L1 Cells
The exposure of 3T3-L1 cells to 300 mg/ml of adiponectin for 24 h resulted in the reduction of Wnt1, Wnt2, Wnt4, Wnt5a, Wnt9a, Wnt9b, and Wnt10b expression (Fig. 6A). The mRNA levels of Fzd1, Fzd2, Fzd3, Fzd4, Fzd6, Fzd7, Fzd8, Lrp5, Lrp6, and Ror1 were also decreased after culture in the presence of 300 mg/ml of adiponectin (Fig. 6B). Adiponectin of 150 mg/ml down-regulated the expression of these genes except Wnt9a and Ror1. The expression of Fabp4, a marker of adipocyte maturation, was increased in 3T3-L1 cells cultured with 150 or 300 mg/ml of adiponectin (Fig. 6C). Western blot analysis showed that 24-h incubation of 3T3-L1 cells with adiponectin and/or Wnt5a did not alter the intracellular protein level of total CaMKII or total JNK (Fig. 6D, H). However, p-CaMKII was reduced in 3T3-L1 cells treated with 150 or 300 mg/ml adiponectin but increased in Wnt5a-treated cells compared with non-treated cells. Supplementation with adiponectin did not inhibit the Wnt5a-induced elevation of p-CaMKII (Fig. 6D, E, F, G). In contrast, p-JNK increased not only in Wnt5a-treated cells but also in cells treated with 150 or 300 mg/ml adiponectin (Fig. 6H,I,J,K).

Expression of Wnt Ligand and Wnt Receptor Genes in Skeletal Muscle
The effects of hyperadiponectinemia on Wnt-related gene expression in skeletal muscle were examined for comparison with those in adipose tissue. The mRNA levels of Wnt1, Wnt5a, Wnt10b, Fzd7, Fzd10, and Lrp6 were significantly lower in the skeletal muscle from adiponectin transgenic mice (line 11) than wild-type mice (Fig. 7A, B). The expression of cyclooxygenase-2 was reduced in the skeletal muscle from the transgenic mice (Fig. 7C).

Discussion
In this comprehensive analysis of Wnt ligands and their receptors in adipose tissue, we demonstrated that hyperadiponectinemia elicited profound effects on Wnt pathway components. Of 13 detectable Wnt ligand families, both lines of adiponectintransgenic mice showed increased expression of Wnt5b and Wnt6 and reduced expression of Wnt1, Wnt2, Wnt5a, Wnt9b, Wnt10b, and Wnt11. Several Wnt ligands, including Wnt1, predominantly use the canonical pathway for intracellular signaling, while some other Wnt ligands, including Wnt5a, mainly activate noncanonical signaling [23], [24]. However, a growing body of evidence suggests substantial overlap between these pathways in cell-and context-dependent manners [25], [26]. Wnt ligands can signal through different receptors, including Fzd, Ror1, Ror2 and Ryk. Fzd receptors are seven-pass transmembrane proteins requisite for canonical Wnt signaling. Two members of the Fzd family were up-regulated and four were down-regulated in adipose tissue from the transgenic mice. However, no significant difference was observed in the expression of the non-Fzd Wnt receptors Ror1, Ror2 and Ryk between wildtype mice and the transgenic mice. On the other hand, the hyperadiponectinemic mice showed lower expression of coreceptor Lrp6, whereas Lrp5 expression was not significantly altered.
These observations prompted us to analyze the effects of hyperadiponectinemia on Wnt signaling pathways in adipose tissue. To assess the activity of the canonical signaling pathway, we examined b-catenin. Western blot analysis showed neither total bcatenin in whole-cell extracts nor non-phospho-b-catenin in nuclear extracts was altered in the transgenic mice compared to wild-type mice. Furthermore, no significant difference was observed in the expression of Myc or Ccnd1, classical target genes of the canonical Wnt signaling pathway [27][28][29], between wildtype mice and adiponectin-transgenic mice. Hence, chronic hyperadiponectinemia in the transgenic mice did not alter the downstream signals of the canonical Wnt pathway in adipose tissue. Lrp5 expression might be sufficient to transduce canonical signals in spite of the reduction of Lrp6 expression.
The two non-canonical Wnt signaling pathways are the Wnt/ PCP pathway, involving RhoA GTPase and JNK, and the Wnt/ Ca 2+ pathway, involving CaMKII and protein kinase C. In this study the expression of RhoA was not altered in adipose tissue from the transgenic mice, although RhoA activity was not measured. However, the protein levels of p-JNK and p-CaMKII, but not total JNK and total CaMKII, were markedly reduced in the transgenic mice compared with wild-type mice. Although the physiological roles of individual Fzd members remain to be characterized, Wnt5a binds to Fzd2 and Fzd4 and transduces the Wnt/Ca 2+ signal. The attenuated Wnt/Ca 2+ signaling pathway in the transgenic mice could be attributable to the reduced expression of Wnt5a and its putative receptors Fzd2 and Fzd4.
Insulin resistance and type 2 diabetes are associated with impaired adipogenesis, characterized by increased fat cell size. Previously, we reported that adipocytes are small in size but increase in number in adiponectin-transgenic mice fed a high-fat/ high-sugar diet compared with wild-type mice fed the same diet [1]. The elevated expression of endogenous adiponectin shown in this study may have resulted from the increase of small adipocytes in the transgenic mice. Wnt signaling may play a role in the altered cellularity of hyperadiponectinemic mice because non-canonical Wnt signaling is one of the regulatory mechanisms of osteoblast and adipocyte differentiation from common mesenchymal stem cells [30], [31]. Wnt5a can shift the balance towards osteoblastogenesis through the activation of CaMKII [30][31][32]. Thus, the reduced expression of Wnt5a and lower level of p-CaMKII in the transgenic mice may have enhanced adipogenesis.
Unlike Wnt5a, adipose expression of Wnt5b and Wnt6 was elevated in the transgenic mice. Overexpression of Wnt5b in 3T3-L1 preadipocytes results in the augmentation of adipocyte differentiation, suggesting that Wnt5b is a positive regulator of adipogenesis [33]. Wnt6 may also be involved in the regulation of adipogenesis because Wnt6 mRNA shows biphasic expression during the differentiation period of 3T3-L1 cells [34]. Taken together, our data suggest that the altered expression of Wnt ligands in adipose tissue from the transgenic mice may be involved in the enhanced adipogenesis and thereby the up-regulated production of endogenous adiponectin.
Next, we examined the expression of putative regulators of adipocyte differentiation, Wnt5a, Wnt5b, and Wnt6, in preadipocytes and mature adipocytes. The expression of Wnt5b was elevated in both preadipocytes and mature adipocytes from adiponectin-transgenic mice, while the reduction of Wnt5a mRNA and the increase of Wnt6 mRNA were detected only in mature adipocytes. These observations suggest that Wnt5b may play a major role in the adipocyte differentiation induced by hyperadiponectinemia. This suggestion is supported by the fact that dietinduced obesity resulted in the reduction of Wnt5b expression in adipose tissue. However, it should be noted that the preadipocyte fraction may also contain immune cells and endothelial cells.
Activated Wnt5a signaling has been implicated in the upregulation of proinflammatory cytokines in autoimmune and infectious diseases [35], [36]. Serum levels of Wnt5a were increased in patients with severe obesity [37]. Wnt5a induces cyclooxygenase-2 expression by endothelial cells through the noncanonical Wnt/Ca 2+ signaling pathway, leading to the enhancement of inflammatory cytokine production [7]. As expected, cyclooxygenase-2 expression was significantly reduced in adipose tissue from adiponectin-transgenic mice. Hence, the inhibition of Wnt/Ca 2+ signaling may be a mechanism by which adiponectin -transgenic mice shown as a ratio to that in wild-type mice (n = 6-10, for each group). Western blot analysis of total and phosphorylated CaMKII (B) and the relative protein levels of p-CaMKII (C) in transgenic mice compared with wild-type mice (n = 5, each). Western blot analysis of total and phosphorylated JNK (D) and the relative protein levels of p-JNK (E) in transgenic mice compared with wild-type mice (n = 4, each). Means and SD, *p,0.01 vs. wild-type mice. doi:10.1371/journal.pone.0067712.g003 attenuates the chronic inflammation associated with obesity and type 2 diabetes.
The promoter region of the Wnt5a gene contains a binding element of NF-kB, a transcription factor activated by TNF-a [38]. Down-regulation of NF-kB may mediate the suppressive effect of adiponectin on Wnt5a expression because adiponectin inhibits TNF-a-induced IkB-a phosphorylation and subsequent NF-kB activation [39], [40]. Further studies are needed to define the mechanism of the selective modulation of Wnt ligands and their receptors by adiponectin.
It was of interest that p-JNK was markedly reduced in adipose tissue from the transgenic mice because JNK has been implicated in the development of insulin resistance and diabetes. In obesity, JNK activity is increased in the liver, muscle, and adipose tissues, and an absence of JNK1 results in decreased adiposity and improved insulin sensitivity in obese models of mice [41]. Furthermore, JNK is a crucial regulator of apoptosis and cell proliferation. Activation of JNK might be associated with increased carcinogenesis in obesity [42], and it was hypothesized that decreased adiponectin levels, as observed in obesity, fail to control JNK expression. Although various inflammatory and metabolic pathways may be involved in the activation of JNK, the observations in this study suggest that the attenuation of non-  canonical Wnt signaling is a mechanism by which hyperadiponectinemia suppresses JNK activity in adipose tissue.
Next, we assessed the ex vivo effects of adiponectin on the mRNA levels of Wnt ligands and their receptors using cultured 3T3-L1 cells. The cells were exposed to a high concentration of recombinant adiponectin (300 mg/ml) because adiponectin concentration is likely to be elevated in the micro-environment of adipocytes in vivo. As in adipose tissue from the transgenic mice, down-regulation of Wnt1, Wnt2, Wnt5a, Wnt9b, Wnt10b, Fzd2, Fzd4, Fzd7, and Lrp6 was also detected in 3T3-L1 cells exposed to adiponectin for 24 h. Thus, adiponectin may directly attenuate the expression of these Wnt-related genes in adipose tissue. In contrast, the reduction of Wnt11 expression in adipose tissue may be an indirect effect of hyperadiponectinemia because Wnt11 mRNA was not decreased in 3T3-L1 cells cultured with adiponectin. The expression of some Wnt-related genes was reduced by adiponectin in vitro, but not in adipose tissue from the transgenic mice. Furthermore, up-regulation of Wnt5b, Wnt6, Fzd6 and Fzd9, observed in the adipose tissue from the transgenic mice, was not observed in 3T3-L1 cells exposed to adiponectin. Although these discrepancies might be explained by the distinct responsiveness of 3T3-L1 cells or the relatively short-term of exposure to adiponectin in the culture experiment, the expression of some Wnt-related genes in adipose tissue of the transgenic mice might be modified by endocrinologic and metabolic alterations resulting from hyperadiponectinemia.
In accordance with the in vivo observations of adiponectintransgenic mice, p-CaMKII was decreased in 3T3-L1 cells after 24-h culture with adiponectin. By contrast, p-CaMKII was elevated after exposure to Wnt5a. The supplementation with adiponectin failed to inhibit the activation of CaMKII induced by Wnt5a, suggesting that the suppressive effect of adiponectin on the non-canonical signaling pathway is mainly due to the modulation of Wnt ligand expression rather than alterations in their receptors.
As expected, phosphorylation of JNK, another non-canonical signaling molecule, was augmented in cells incubated with Wnt5a. However, exposure of 3T3-L1 cells to adiponectin also resulted in activation of JNK. This finding is not surprising because it was reported that adiponectin stimulates JNK ex vivo through the activation of phospholipase C [43], [44]. The suppression of JNK Figure 7. Analysis of the Wnt ligand family, their receptors and cyclooxygenase-2 in skeletal muscle from wild-type mice and adiponectin-transgenic mice. mRNA levels of the Wnt ligand family (A), Frizzled receptors (B), and the cyclooxygenase-2 gene (C) in skeletal muscle from adiponectin-transgenic mice shown as ratios to those of wild-type C57BL/6 mice at the age of 20 weeks (n = 4, each). Means and SD. *p,0.05, **p,0.01 vs. wild-type mice. doi:10.1371/journal.pone.0067712.g007 activity observed in adiponectin-transgenic mice may have been a more chronic effect of adiponectin.
Finally, we analyzed the alterations of Wnt-related gene expression in skeletal muscle from the transgenic mice and found that the expression of Wnt1, Wnt5a, Wnt10b, Fzd7, Fzd10, and Lrp6, as well as the cyclooxygenase-2 gene, was reduced. All of the genes except Fzd10 were also down-regulated in adipose tissue from the transgenic mice, although there was considerable organspecificity in the response of the Wnt system to hyperadiponectinemia. The physiological relevance of the changes in Wnt-related gene expression in skeletal muscle found in this study requires further investigation.
In conclusion, using adiponectin-transgenic mice, we showed that chronic hyperadiponectinemia altered the expression pattern of Wnt ligands, Fzd receptors and the Lrp6 coreceptor in adipose tissue, suggesting that the inhibition of the Wnt/Ca 2+ and JNK signaling pathways may be involved in the enhanced adipocyte differentiation and attenuated inflammation in adipose tissue induced by hyperadiponectinemia. Thus, adiponectin might exert these effects when adiponectin production is augmented by body weight loss or the administration of thiazolidine derivatives. However, further studies are required to elucidate the doseresponse relationship and to determine whether adiponectin exhibits similar action in other tissues.