The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Department of Endocrinology and Metabolism C, Aarhus University Hospital, Aarhus Sygehus, Tage Hansensgade 2, DK-8000 Aarhus C, Denmark. E-mail: firstname.lastname@example.org
Objective: To investigate the presence and regulatory properties of the adiponectin receptors, AdipoR1 and AdipoR2, in human adipose tissue (AT) and in isolated human adipocytes.
Research Methods and Procedures: The effect of obesity, weight loss, and gender on expression of AdipoR1 and AdipoR2 was investigated in subcutaneous AT. The influence of fat distribution on these receptors was investigated in paired samples of subcutaneous and omental AT. Gene expression of these receptors was quantified by reverse transcriptase-polymerase chain reaction.
Results: AdipoR1 mRNA levels were ∼10-fold higher than adipoR2 in both AT fragments and in isolated adipocytes. AdipoR1 expression was lower in AT from obese subjects (p < 0.05) compared with that from normal-weight subjects, and AdipoR1 displayed a negative correlation with BMI (r = −0.53, p < 0.01). In obese subjects, weight loss (∼12 kg) increased AdipoR1 expression by 80% in AT (p < 0.01). Concerning regional differences, AdipoR1 showed significantly lower expression in omental AT than in subcutaneous AT (p < 0.01). No gender difference was observed in the expression of these receptors. In human preadipocyte cultures, AdipoR1 expression was not induced during the differentiation process, whereas AdipoR2 was induced by 5-fold (p < 0.05).
Discussion: AdipoR1 is highly expressed in human AT, indicating that adiponectin may have biological effects in AT in an autocrine/paracrine manner. AdipoR1 expression in AT is reduced in obese subjects and is increased after weight loss. Thus, it can be suggested that adiponectin might have reduced biological effects in AT due to low levels of adiponectin receptors in obese subjects and in omental adipocytes, which may further aggravate the negative metabolic effect of low levels of adiponectin characterizing the obese state.
Adipose tissue (AT)1 secretes a number of proteins and hormones collectively referred to as adipokines, which have multifaceted biological actions (1). Adiponectin is one of the most recently discovered adipokines, and it is produced exclusively by adipocytes (2,3,4). Adiponectin has been found to ameliorate insulin resistance by increasing fatty acid oxidation (5,6) and by suppressing hepatic glucose production (7,8). The fatty acid oxidation and glucose use seem to occur through the activation of 5′-adenosine monophosphate (AMP)-activated protein kinase (AMPK) (9). Decreased levels of plasma adiponectin have been found in relation to obesity, type 2 diabetes, and cardiovascular disease (10,11).
Two complementary DNA-encoding adiponectin receptors (named AdipoR1 and AdipoR2) have been cloned recently (12). These proteins are related to G protein-coupled seven-transmembrane domain receptors. However, the sequence homology of both AdipoR1 and AdipoR2 with these receptors is low. Furthermore, the N terminus is intracellular, and the C terminus is extracellular, which is opposite to the topology of classical G protein-coupled receptors (12). AdipoR1 seems to be a high-affinity receptor for globular adiponectin and a low-affinity receptor for full-length adiponectin, whereas AdipoR2 was found to be an intermediate-affinity receptor for both forms of adiponectin (12). AdipoR1 is found abundantly expressed in skeletal muscle, and AdipoR2 is expressed predominantly in the liver. These receptors are suggested to mediate the effects of adiponectin on AMPK (9,13), PPAR-α ligand activities (14), fatty acid oxidation, and glucose uptake (12)
Only little is known about the existence and regulation of these adiponectin receptors, and Fasshauer et al. (15) have recently demonstrated the presence of both receptors in 3T3-L1 adipocytes. They found that growth hormone up-regulated the expression of AdipoR2 in these cells. In contrast, neither insulin nor cytokines had any effect on AdipoR1 and AdipoR2 gene expression in 3T3-L1 cells. Recent studies indicate that adiponectin receptors are also expressed in macrophages (16) and in pancreatic cells (17).
The present study was undertaken to determine the presence and regulatory properties of AdipoR1 and AdipoR2 gene expression in human AT and in isolated human adipocytes. Furthermore, gene expression of these receptors was investigated during the differentiation of preadipocytes to mature adipocytes. Because adiponectin has been suggested to be a link between obesity and insulin resistance (18,19,20), we investigated the impact of obesity on the gene expression of these receptors, in addition to the effect of weight loss. Finally, we investigated the expression of these receptors in AT from different depots within the same individual, as well as the influence of gender.
Research Methods and Procedures
Obesity and Regional Differences
The study on regional differences in the gene expression of AdipoR1 and AdipoR2 was performed using AT biopsies from subcutaneous abdominal and omental AT depots, obtained from patients undergoing surgery. The study group consisted of 17 lean and 12 obese individuals. The lean subjects (17 women) (age, 48.5 ± 1.5 years; BMI, 24.0 ± 0.5 kg/m2) underwent laparoscopic gynecological surgery, such as hysterectomy, due to benign diseases. The obese subjects (12 women) (age, 36 ± 2.5 years; BMI, 44.4 ± 1.38 kg/m2) underwent laparoscopic surgery for obesity (gastric banding). After an overnight fast, the AT biopsies were taken as paired samples from the two AT depots in the beginning of the surgical procedure.
Furthermore, for the correlation analysis between BMI and AdipoR1/R2 gene expression, an additional five moderately obese women (age, 33.4 ± 4.1 years; BMI, 34.2 ± 1.8 kg/m2) were included in these analyses. All of the subjects were white and apparently healthy except for one of the obese patients from the surgery group who suffered from diet-treated type 2 diabetes. None of the participants was taking any medication known to affect AT metabolism.
Weight Loss Study
The effect of weight loss included 32 obese individuals (age, 44 ± 1.4 years; BMI, 38 ± 0.7 kg/m2). These subjects received a very low-calorie diet (3.4 MJ/d) for 8 weeks. Subcutaneous abdominal AT biopsies were taken by needle aspiration before and at the end of the study.
Study of the Influence of Gender
For the study of gender differences, subcutaneous AT biopsies were obtained from 35 subjects (18 men, age, 35.6 ± 2 years; BMI, 25.9 ± 0.6 kg/m2; and 17 women, age, 37 ± 2.2 years; BMI, 25.2 ± 0.6 kg/m2). All subjects were healthy and not taking any regular medication.
Study of Preadipocyte Differentiation
Subcutaneous AT samples were obtained from eight individuals undergoing liposuction from the abdominal region for cosmetic reasons. The tissue samples were immediately transferred to the laboratory for preadipocyte isolation. All protocols were approved by the local Ethics Committee, and the subjects provided written informed consent.
The biopsies were taken from the subcutaneous abdominal region by needle aspiration as previously described (21). All biopsies were transported in sterile containers to the laboratory within 30 minutes after removal. The AT was washed repeatedly with isotonic saline and was either immediately frozen in liquid nitrogen and stored at −80 °C for later RNA extraction or was used for subsequent culture. The actual number of subjects used in each study is indicated in the legends to the figures.
Isolation Procedures and Culture of Adipocyte Precursor Cells
Adipocytes were isolated by collagenase digestion (0.15 mg/g AT) of AT fragments in 10 mM HEPES buffer for 60 minutes at 37 °C (22). The isolated adipocytes were washed three times in buffer containing 5% albumin and then resuspended in medium 199 containing 1% bovine serum albumin and 25 mM HEPES. Finally, the 200-μL cell suspension containing 10% adipose cells, corresponding to ∼100, 000 adipocytes in each tube, was snap-frozen in liquid nitrogen and kept at −80 °C for later RNA extraction. After the initial collagenase digestion, the stroma-vascular (S-V) fraction remaining was centrifuged for 15 minutes at 6300g, resuspended in 9 mL of buffer, and filtered through a nylon mesh. This procedure was repeated three times, after which the supernatant was removed, and the S-V fraction was snap-frozen in liquid nitrogen and kept at −80 °C for later RNA extraction.
The isolation and differentiation (preadipocytes) were performed according to the method described by Hauner et al. (23) with minor modifications. Isolated adipocytes were prepared as described above. After centrifugation at 200g for 10 minutes to separate adipocytes and non-adipocyte cells from the AT, the pellet (preadipocytes and erythrocytes) was incubated with an erythrocyte lysing buffer consisting of 0.154 M NH4Cl, 10 mM KHCO3, and 0.1 mM EDTA-Na (pH 7.3) for 10 minutes at room temperature. The preadipocyte fraction was filtered through a nylon mesh with a pore size of 150 μm and centrifuged as mentioned above. The fraction was resuspended in 10 mL of Dulbecco modified Eagle (DME)-Ham's F-12 medium (1:1 vol/vol) and filtered through a nylon mesh with a pore size of 70 μm. After a thorough mix, the cells were incubated in DME/HAM's F-12 medium (1:1 v/v), supplemented with 10% fetal calf serum, 15 mM NaHCO3, 15 mM HEPES, and 1% penicillin/streptomycin and kept at 37 °C in 5% CO2. To determine the number of cells, 50 μL of cell suspension mixed with 50 μL of trypan blue was counted with a Thomacytometer. The cells were incubated in six-well plates at a density of 500, 000 cells/well. After 16 to 20 hours of incubation, the medium was removed, and the cells were washed. Hereafter (stated as Day 0), the cells were incubated in serum-free medium, consisting of DME/HAM's F-12 medium with 33 μM biotin, 17 μM pantothenate, 10 μg/mL human transferrin, 100 nM insulin, 200 pM triiodothyronine, and 1% penicillin/streptomycin. Incubation was continued until 80% of the cells were fully differentiated into mature adipocytes. Cells were regarded as fully differentiated when displaying a round shape and a cytoplasma completely filled with multiple lipid droplets. Total incubation period was 14 to 16 days. For the first 3 days, the medium was supplemented with 0.2 mM isobutylmethylxanthine. Culture medium was changed every 2 to 3 days.
Isolation of RNA
Total RNA was isolated using the TriZol reagent (Invitrogen, Carlsbad, CA). RNA was quantified by measuring absorbency at 260 and 280 nm. The integrity of the RNA was checked by visual inspection of the two ribosomal RNAs, 18S and 28S, on an agarose gel.
RNA was reversely transcribed with RT and random hexamer primers at 23 °C for 10 minutes, 42 °C for 60 minutes, and 95 °C for 10 minutes according to the manufacturer's instructions (GeneAmp RNA PCR Kit; PerkinElmer Life and Analytical Sciences, Boston, MA). Then, 2 μL of each RT reaction was amplified in a 20-μL PCR-mastermix containing the specific primers, Hot Star Taq DNA polymerase, and SYBR-Green PCR-buffer. All samples were determined as duplicates.
The primer pairs used were: AdipoR1 (110 bp), 5′-CCGGTTTGCCACTCCTAAGC and 5′-TGACAAAGCCCTCAGCGATAG; AdipoR2 (222 bp), 5′-AGGCCGCCACCATAGGG and 5′-CGCCGATCAAACGAAACT; adiponectin (301 bp), 5′-CATGACCAGGAAACCACGACT and 5′-TGAATGCTGAGCGGTAT; and as housekeeping gene, β-actin (658 bp), 5′-ACGGGGTCACCCACACTGT and 5′-CTAGAAGCATTTGCGGTGGACGATG; S-18 (484 bp), 5′-CCTACACGCCGCCGCTTGTGC and 5′-TTTCTTCTTGGACACACCCAC. Real-time quantization of AdipoR1/R2 and adiponectin mRNA relative to β-actin or S-18 mRNA was performed by using a SYBR-Green real-time PCR assay and an iCycler PCR machine (Bio-Rad, Hercules, CA) as previously described (19). In brief, samples were incubated in separate tubes for an initial denaturation at 95 °C for 10 minutes, followed by 40 PCR cycles, where AdipoR1/R2 mRNA and β-actin or S-18 mRNA were amplified. Each cycle consisted of 30 seconds at 95 °C, 30 seconds at 57 °C, and extension for 60 seconds at 72 °C. During the extension step, increase in fluorescence was measured in real time. Data were obtained as threshold cycle (CT) values. CT was defined as the cycle number at which the fluorescence reached 10 times the SD of the baseline fluorescence. Relative gene expression was calculated using the formula: 1/2(CT AdipoR1 -CT −β-actin), essentially as described in the User Bulletin No. 2, 1997 from PerkinElmer Life and Analytical Sciences.
Normality of distribution of the data was tested by the Kolmogorov-Smirnov test. The weight loss data were not normally distributed and were, therefore, logarithmically transformed. A two-tailed paired Student's t test was used for comparison of mean differences between depots in the same individual and for data before and after weight loss. A non-paired Student's t test was used to test the statistical significance when comparing two independent samples, i.e., differences between lean and obese and between men and women. The significance of correlations was examined using a Pearson bivariate correlation coefficient. To analyze changes over time, a one-way repeated ANOVA (one-way ANOVA) was used. Data are presented as means ± SE or as the coefficient of variation. p < 0.05 was considered statistically significant. For analyses, the SPSS statistical package was used (SPSS Inc., Chicago, IL).
AdipoR1 and AdipoR2 in AT
Both AdipoR1 and AdipoR2 are expressed in whole AT and in isolated adipocytes (Figure 1). These receptors were, however, also expressed in the S-V fraction of AT in humans (Figure 1). Compared with the AdipoR expression in whole AT, adipoR1 showed ∼2-fold higher expression in the S-V fraction (p < 0.01), and the adipoR2 expression was reduced by 75% (p < 0.001) (Figure 1). We used the adipocyte-specific adiponectin gene as a control, and in contrast to AdipoR1 and AdipoR2, the expression of adiponectin was higher in the isolated adipocytes as compared with the AT fragments, with really no expression in the S-V fraction (Figure 1). AdipoR1 gene expression is ∼10-fold higher than the expression of AdipoR2 in human AT and in isolated adipocytes (Figure 2, A and B). No gender difference was observed in the expression of the two receptors in subcutaneous AT (data not shown).
Influence of Obesity and Fat Depots on AdipoR1 and AdipoR2 Gene Expression
AdipoR1 gene expression in subcutaneous AT was 60% lower in obese women with a BMI > 35 kg/m2 compared with normal-weight women (BMI < 26 kg/m2) (14.8 ± 2.5 vs. 38 ± 6.2; p < 0.01; Figure 2A). Similar differences were observed in the omental AT, where the AdipoR1 expression was 62% lower in the obese as compared with normal-weight women (10.1 ± 1.8 vs. 26.7 ± 3.6; p < 0.01; Figure 2A).
AdipoR2 gene expression was also lower (30%) in subcutaneous AT in the obese women compared with the lean women (1.7 ± 0.3 vs. 2.4 ± 0.4; p < 0.05). No difference in AdipoR2 mRNA between obese and lean women in omental AT was observed (Figure 2B).
AdipoR1 expression was significantly lower (34%) in omental AT as compared with subcutaneous AT in obese subjects (BMI > 35 kg/m2) (10.1 ± 1.8 vs. 15.4 ± 2.6; p < 0.05, N = 11, Figure 2A). In normal-weight subjects (BMI < 26 kg/m2), a similar reduction in AdipoR1 expression was found in omental AT compared with subcutaneous AT (26.7 ± 3.6 vs. 38.0 ± 6.2; p = 0.08, N = 17). The expression of adipoR2 was not significantly different in subcutaneous AT compared with omental AT in the obese subjects. In lean subjects, however, the AdipoR2 gene expression was significantly lower in omental AT as compared with subcutaneous AT (reduced by 37%; 1.5 ± 0.2 vs. 2.4 ± 0.4; p < 0.05, N = 17, Figure 2B).
Correlation with BMI
In the correlation study, AdipoR1 and AdipoR2 expression were measured in subjects covering a wide range of BMI. A significant inverse association was found between BMI and AdipoR1 mRNA levels in subcutaneous AT (r = −0.53; p < 0.01, N = 39) and in omental AT (r = −0.52; p < 0.01, N = 32) (Figure 3). A slight inverse correlation was found between BMI and AdipoR2 mRNA levels in subcutaneous AT (r = −0.27; p = 0.09; N = 40), whereas no correlation with AdipoR2 mRNA in omental AT was observed (r = 0.02; not significant, N = 32) (data not shown).
Effect of Weight Loss
Weight loss was induced by a very low calorie diet (VLCD) for 8 weeks in 32 obese subjects, with a mean weight loss of 12.1 ± 0.6 kg. This weight loss was associated with an increased expression of AdipoR1 in subcutaneous AT by 81% (p < 0.01) (Figure 4). The expression of AdipoR2 was not affected by the weight loss (Figure 4). As a control, we also measured the expression of adiponectin in AT, and this expression was significantly increased by 65% (p < 0.01) in association with the weight loss (Figure 4).
We also investigated the effect of acute weight loss obtained by total fasting for 6 days. In five very obese subjects, 6 days of total fasting had no effect on the expression of AdipoR1 or AdipoR2 in AT (data not shown).
AdiopR Expression in Primary Preadipocyte Cultures
The expression of AdipoR1 was rather constant during the differentiation period (from Days 0 to 15) (Figure 5A). In contrast, the gene expression of AdipoR2 was very low in preadipocytes but was clearly induced during the adipocyte differentiation, with a 5-fold increment in expression observed between Days 0 and 15 (p < 0.05; Figure 5B).
In this study, we demonstrated the presence of AdipoR1 and AdipoR2 mRNA in human AT and in isolated human adipocytes. The expression of AdipoR1 is 10- to 15-fold higher than that of AdipoR2, indicating that AdipoR1 is the predominant adiponectin receptor in human adipocytes. This low level of expression of AdipoR2 in human AT is compatible with the suggestion that AdipoR1 is ubiquitously expressed, whereas AdipoR2 is more restricted to liver and skeletal muscle (12). AdipoR1 was expressed in nearly similar quantities in AT fragments and isolated adipocytes but was expressed ∼2-fold higher in the S-V fraction of the AT. These findings indicate that this receptor is expressed in adipocytes and in other cell types in whole AT, with a higher expression in some non-fat cells in AT such as monocytes and possibly endothelial cells (16,24). These cell type differences in the expression of AdipoR1 are in contrast to the adipocyte-specific gene, adiponectin, which is expressed more in adipocytes than in whole AT, with nearly no expression in the S-V fraction. Compared with AdipoR1, the cell-dependent expression of AdipoR2 was more similar to the expression of adiponectin.
During adipocyte differentiation the expression of AdipoR2 was found to be increased, whereas AdipoR1 mRNA levels were relatively stable during the differentiation process. These latter findings are in accordance with findings in 3T3-L1 adipocytes where AdipoR2 was also found to be induced during differentiation to mature adipocytes (15).
Gender differences in the level of adiponectin in the circulation have been shown previously, with women displaying higher levels of adiponectin than men (3). However, with regard to expression of adiponectin receptors in AT, we found no gender differences, indicating that there might not be gender differences in the effect of adiponectin in AT.
Interestingly, we found AdipoR1 and AdipoR2 levels to be considerably lower (30% to 60% lower) in AT from obese subjects as compared with AT from lean subjects. Moreover, we found that AdipoR1 gene expression was negatively correlated with BMI in both omental AT and subcutaneous AT. These findings are in accordance with data obtained in obese (ob/ob) mice, where Tsuchida et al. (25) have observed reduced expression of these adiponectin receptors in both AT (70% reduction) and skeletal muscle (50% reduction) compared with lean mice. AdipoR1 gene expression was lower in the omental AT as compared with subcutaneous AT. As is shown for AdipoR1, the gene expression of adiponectin in AT is also found to be low in relation to obesity (20,26). Moreover, the present fat depot differences, with reduced expression of AdipoR1 in omental AT as compared with subcutaneous AT, have also been found in relation to the expression of adiponectin (20,27). These data indicate that the biological effect of adiponectin is reduced in obese subjects and is reduced in omental AT as compared with subcutaneous AT. Together with the fact that plasma adiponectin is low in obese subjects, these data might suggest a severely reduced action of the adiponectin system in obese subjects.
The down-regulation of adiponectin receptors in AT from obese subjects appears to be reversible, because weight loss (12 kg) was found to be followed by a significant increment in the expression of AdipoR1, in particular, by ∼80%. In mice, fasting for 48 hours up-regulates both AdipoR1 and AdipoR2 expression in liver and in skeletal muscle (25). This is in contrast to our findings in AT, because we did not find changes in AdipoR1 and AdipoR2 expression in very obese subjects after 6 days of total fasting. Taken together, these findings suggest differential effects of acute fasting compared with weight loss on the adiponectin receptors in different tissues.
Because both obesity and weight loss are associated with changes in insulin sensitivity and levels of plasma insulin, it can be hypothesized that insulin may be involved in the regulation of expression of AdipoR1/AdipoR2. Tsuchida et al. (25) found both in vivo (mice) and in vitro that insulin may directly decrease the expression of AdipoR1/AdipoR2 in liver and in skeletal muscle. Moreover, in an investigation of subjects with and without a family history of type 2 diabetes, a positive correlation was found between insulin sensitivity and the expression level of these receptors in human skeletal muscle as well as an inverse correlation with fasting plasma insulin (28). In contrast, a very recent study by Staiger et al. (29) demonstrated that, in healthy subjects, plasma insulin was positively correlated and insulin sensitivity negatively correlated with the expression of adiponectin receptors in skeletal muscle. In addition, insulin had no direct effects on receptor expression in vitro using murine C2Cl2 myotubes (29). Finally, Debard et al. (30) were not able to find alterations in AdipoR1 and AdipoR2 in skeletal muscle of type 2 diabetic patients. Thus, the data on the effects of insulin on these receptors and how they are regulated in skeletal muscle are still very scanty and warrant further investigations. With respect to adipocytes, no effects of insulin have been found on AdipoR1 and AdipoR2 in 3T3-L1 adipocytes (15). AdipoR1/AdipoR2 expression in other cell types such as macrophages has been found to be regulated by PPAR-agonists (16), and, in pancreas, these receptors were up-regulated by the monounsaturated fatty acid, oleate (17).
The expression of AdipoR1 and AdipoR2 both in AT and in isolated adipocytes indicates that adiponectin may have autocrine and paracrine effects in AT. The action of adiponectin has, to some degree, been characterized in the liver and in skeletal muscle cells, but its effect on adipocytes is still largely unknown. In liver and muscle cells, adiponectin has been found to stimulate AMPK and enhance lipid oxidation and to improve the insulin action on glucose metabolism (9,13,14,15). Because we and others have found AMPK activity in adipocytes (31), it is possible that adiponectin may have similar effects in AT. Moreover, AT produces a large amount of cytokines and cytokine-like molecules, and due to its anti-inflammatory effects in other cell systems (32), adiponectin might also be suggested to exert anti-inflammatory effects in AT. These possible effects in AT, however, await further studies.
The limitation of the present study is related to the fact that we have measured gene expression levels; therefore, further studies are necessary to reveal whether these receptor expressions and changes in expression are also reflected in changes in the level of receptor protein and in biological function.
In summary, this study demonstrates the presence of AdipoR1 and AdipoR2 in human AT. We demonstrate for the first time, to our knowledge, that obesity is associated with lower levels of AdipoR1 in AT and that diet-induced weight loss in obese subjects is followed by an increase in the level of AdipoR1 mRNA. Furthermore, lower levels of AdipoR1 mRNA were found in omental AT as compared with subcutaneous AT in both obese and lean subjects. These findings are compatible with the suggestion that AT expression of AdipoR1 and AdipoR2 might be involved in the pathogenesis of obesity-induced health complications such as insulin resistance and cardiovascular disease.
We appreciate the technical assistance of Lenette Pedersen and Pia Hornbek. This study was supported by grants from the Danish Diabetes Association, the Novo Nordic Foundation, the Clinical Institute at Aarhus University, and the Danish Medical Research Council.