Transgenic overexpression of hexose-6-phosphate dehydrogenase in adipose tissue causes local glucocorticoid amplification and lipolysis in male mice

Abstract

The prereceptor activation of glucocorticoid production in adipose tissue by NADPH-dependent 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) has emerged as a potential mechanism in the pathogenesis of visceral obesity and metabolic syndrome. Hexose-6-phosphate dehydrogenase (H6PDH) is an endoplasmic reticulum lumen-resident enzyme that generates cofactor NADPH and thus mediates 11β-HSD1 activity. To determine the role of adipose H6PDH in the prereceptor modulation of 11β-HSD1 and metabolic phenotypes, we generated a transgenic (Tg) mouse model overexpressing H6PDH under the control of the enhancer-promoter region of the adipocyte fatty acid-binding protein (aP2) gene (aP2/H6PDH Tg mice). Transgenic aP2/H6PDH mice exhibited relatively high expression of H6PDH and elevated corticosterone production with induction of 11β-HSD1 activity in adipose tissue. This increase in corticosterone production in aP2-H6PDH Tg mice resulted in mild abdominal fat accumulation with induction of C/EBP mRNA expression and slight weight gain. Transgenic aP2/H6PDH mice also exhibited fasting hyperglycemia and glucose intolerance with insulin resistance. In addition, the aP2/H6PDH Tg mice have elevated circulating free fatty acid levels with a concomitant increased adipose lipolytic action associated with elevated HSL mRNA and Ser⁶⁶⁰ HSL phosphorylation within abdominal fat. These results suggest that increased H6PDH expression specifically in adipose tissue is sufficient to cause intra-adipose glucocorticoid production and adverse metabolic phenotypes. These findings suggest that the aP2/H6PDH Tg mice may provide a favorable model for studying the potential impact of H6PDH in the pathogenesis of human metabolic syndrome.

visceral fat deposition is frequently associated with metabolic syndrome, including visceral obesity, insulin resistance, hypertension, and dyslipidemia (13, 22, 39). However, the underlying mechanisms of these metabolic perturbations are poorly understood. Patients with glucocorticoid (GC) excess (Cushing's syndrome) promote visceral fat deposition and develop metabolic syndrome (1, 21). In adipose tissue, GCs promote lipolysis by stimulating two key enzymes, hormone-sensitive lipase (HSL) and adipose triglyceride lipase (ATGL), to increase hydrolysis of triacylglycerol and release free fatty acids (FFA) in the circulation that is linked to hyperlipidemia and global insulin resistance (8, 24, 46). However, tissue-specific GC availability is regulated by an intracellular endoplasmic reticulum (ER) lumen-resident enzyme, 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), that converts inert cortisone (11-dehydrocorticosterone in rodents) to the active glucocorticoid receptor (GR)-ligand cortisol (corticosterone in rodents) and thereby amplifies intracellular GC reproduction, particularly in adipose tissues (7, 41, 23, 33). Increased adipose GC activation by 11β-HSD1 leads to visceral obesity and features of the metabolic syndrome (29). Prereceptor amplification of GC production in adipose tissue by 11β-HSD1 is thus an emerging etiology for the pathogenesis of metabolic syndrome.

The function of 11β-HSD1 is closely dependent on its ability to activate GCs in vivo, in which it exhibits bidirectional activity. This dual action is controlled by its cofactor (NADPH) that is provided by the enzyme hexose-6-phosphate dehydrogenase (H6PDH) (34, 32). H6PDH is a microsomal enzyme located in the lumen of the ER and principally expressed in insulin target tissues, such as liver and adipose tissues, sites of 11β-HSD1 and GR expression (20, 30). In the ER lumen, H6PDH can metabolize glucose 6-phosphate (G-6-P) and NADP to generate NADPH, which is used by 11β-HSD1. Targeted deletion or mutation of H6PDH results in a switch in the set point of 11β-HSD1 inactivating cortisol to cortisone and demonstrates the physiological relevance of 11β-HSD1 for H6PDH (6, 14). We recently reported that tissue-specific elevation of 11β-HSD1 in liver, visceral, and subcutaneous fat is associated with H6PDH expression in obese db/db diabetic mice (47). In human subcutaneous adipose cells, H6PDH mRNA level is elevated during the differentiation of adipocytes with stimulation of cortisol production by 11β-HSD1 (5). Clinical observations also have shown higher H6PDH expression in patients with type 2 diabetes (T2DM) than normal controls in both visceral and subcutaneous fat, implicating H6PDH as a contributing factor in the pathogenesis of T2DM (45). However, the phenotypic consequence of increasing H6PDH expression exclusively in adipose tissue has not been explored.

To test the hypothesis that increasing the H6PDH expression exclusively in adipose tissue would induce 11β-HSD1 driving local GC production and lead to adverse metabolic phenotypes, we created transgenic mice overexpressing H6PDH under the control of the enhancer-promoter region of the adipocyte fatty acid-binding protein (aP2) gene (aP2/H6PDH mice). Here we show that aP2/H6PDH mice have increased adipose corticosterone production and lipolysis and exhibit the features of modest metabolic syndrome.

MATERIALS AND METHODS

Generation of transgenic aP2-H6PDH mice.

The 5.4-kb αP2 promoter/enhancer was used to drive fat-specific expression of the mouse H6PDH transgene. The transgenic expression vector was constructed by subcloning 4.6 kb EcoRV-NotI fragment of mouse H6PDH cDNA into a eukaryotic expression vector under the αP2 promoter (obtained from Dr. Bruce Spiegelman). The fragment containing αP2 promoter, H6PDH cDNA, and polyadenylation signal was released by restriction enzyme digestion and fractionated through agarose gel electrophoresis. The purified DNA fragment was microinjected in nuclei of pronuclear embryos (C57black/6). Injected embryos were implanted to pseudopregnant fosters to complete the gestation and birth. The transgenic founders were identified by PCR. Two independent lines with H6PDH overexpression in adipose tissue were obtained.

Animal study and metabolic analysis.

This study was approved by the Institutional Animal Care and Use Committee of the Charles Drew University. Animals were weaned at the age of 3 wk and housed in isolated cages under a 12:12-h light-dark cycle. Transgenic αP2-H6PDH mice and their nontransgenic littermates were given a standard laboratory chow diet with water ad libitum, and body weights and food intake were recorded weekly. Glucose tolerance (GTTs) and insulin tolerance (ITTs) tests were performed as previously described (29, 47). For the GTT, 1 g/kg glucose was used in mice after a 12-h fast. For the ITT, animals were fasted for 8 h, and blood samples were drawn at indicated time points following insulin injection (0.75 U/kg ip; Novolin R; Eli Lilly, Indianapolis, IN). Blood samples were collected between 0900 and 1000, and fat pads were weighed and frozen in liquid nitrogen and stored at −80°C until metabolic assays.

Biochemical assays.

Blood glucose levels were determined by the glucose oxidase method. Commercially available kits were used to measure plasma corticosterone, insulin, and FFA levels (Abcam).

Adipocyte morphometric evaluation and culture of primary adipose cells.

Three mice per experimental group were analyzed, with three representative images per section taken for epididymal fat pad sections per mouse. Adipose morphometry was visualized by hematoxylin and eosin staining. Adipocyte size was determined using the NIH ImageJ software. To quantify cell number, adipocytes were isolated from mouse epididymal adipose pad by using 1 mg/ml collagenase type II (Sigma, St. Louis, MO) in Krebs-Ringer-bicarbonate-HEPES (KRBH) buffer (Sigma K4002) with 1% fatty acid-free BSA at a ratio of 1.5 ml buffer for every 100 mg tissue as described previously (12, 11). After digestion, the mature adipocytes were separated from the stromal vascular cells containing preadipocytes by slow-speed centrifugation and washed with KRBH solution by repeated centrifugation. Before all metabolic studies, cell viability was evaluated by trypan or methylene blue exclusion, and cells were counted using an improved Neubauer hemacytometer (American Scientific Products). For FFA assay, 100-μl aliquots of adipocytes were then transferred to 24-well plates and incubated with isoproterenol (1 μM) for 2 h (10). FFA and glycerol levels in the medium were measured with a colorimetric kit by using FFA and a glycerol kit (Abcam), respectively, and normalized to cell density.

Microsomal enzymatic activity assays in adipose tissue.

The adipose microsomal pellet was obtained, and H6PDH enzyme activity was carried out by spectrophotometric measurement of NADPH production in the presence of G-6-P and NADP using absorbance at 340 nm per our previous reports (47, 15). Protein (100 μg) from adipose microsomes was incubated with 0.5–5 mM G-6-P, 1–5 mM NADP, and 100 mM glycine buffer solution at 22°C for 0–10 min. The adipose microsomes were permeabilized with 1% Trition-100 to allow the free access of the cofactor to the intraluminal enzyme. Specific activities were calculated and expressed as micromoles of NADPH production per minute per milligram of protein. The protein concentration was measured by Bradford assay (Bio-Rad Protein Assay Kit). 11β-HSD1 activity was evaluated by addition of 1 mM NADPH and 250 nmol/l 11-dehydrocorticosterone (11-DHC) with 11-[³H]DHC as tracer to microsomes in KRB solution at 37°C for 1–2 h, as done in our previous study (26). Steroids were separated by TLC, and the percentage of conversion of 11-[³H]DHC to [³H]corticosterone was calculated from the radioactivity. 11β-HSD1 activity was also evaluated by immunoassay of the corticosterone produced from 11-DHC using a corticosterone kit (Abcam). Adipose corticosterone concentrations were measured by methanol extraction (37, 42) using a corticosterone ELISA kit (Abcam).

RNA extraction and analysis by quantitative real-time RT-PCR.

Adipose total RNA was extracted using an RNAzol B kit (Invitrogen). Real-time primers for mouse H6PDH (forward: 5′-TGGCTACGGGTTGTTTTTGAA-3′; reverse: 5′-TATACACGGTACATCTCCTCTT CCT-3′), 11β-HSD1 (forward: 5′-CCTTGGCCTCATAGACACAGAAAC-3′; reverse: 5′-GGAGTCAAAGGCGATTTGTCAT-3′), GR (forward: 5′-TGCTATGCTTTGCTCCT GATCTG-3′, reverse: 5′-TGTCAGTTGATAAAACCGCTGC-3′), HSL (forward: 5′-GG GCAAAGAAGGATCGAAGAA-3′; reverse: 5′-GCGTAAATCCATGCTGTGTGA-3′), C/EBPα (forward: 5′- TGGACAAGAACAGCAACGAGTAC-3′; reverse: 5′-CGGT CATTGTCACTGGTCAACT-3′), and 18S (forward: 5′-GGACAGGATTGACAGATTG ATAGC-3′; reverse: 5′-TCGTTA TCGGAATTAACCAGACAA-3′) were designed with Primer express software 2.0 (Applied Biosystems). Real-time RT-PCR analysis was performed using SYBER green detection of amplified products according to the protocols recommended by the manufacturer (Applied Biosystems, Carlsbad, CA). Threshold cycle (C_t) readings for each of the unknown samples were then used to calculate the amount of target genes and were normalized to the signal of 18S rRNA. Data analysis is based on the C_t method.

Immunoblotting analysis.

Adipose tissues proteins were separated on 4–12% acrylamide SDS-PAGE gels (Bio-Rad, Hercules, CA) for analysis of H6PDH, 11β-HSD1, total HSL, phospho-Ser⁶⁶⁰ HSL, ATGL, and GAPDH (Cell Signaling Tech, Danvers, MA). Proteins were transferred to nitrocellulose membranes. Membranes were washed and incubated with the appropriate secondary antibody. Protein bands were visualized using enhanced chemiluminescence (Amersham) and exposure to Hyperfilm ECL X-ray film.

Statistical analysis.

The data shown represent means ± SE for all of the determinations. Data were compared using an unpaired Student's t-test. To compare multiple groups, one-way ANOVA was used. When ANOVA revealed significant differences, then group comparisons were performed using the Newman-Keul's post hoc test. A P value <0.05 was considered statistically significant.

RESULTS

Phenotypic characterization of transgenic aP2/H6PDH mice.

To determine whether overexpression of H6PDH has an effect on energy homeostasis, we first compared body weight, food intake, and white adipose fat pad weight between the aP2-H6PDH transgenic mice (aP2/H6PDH Tg mice) and their nontransgenic littermates controls [wild-type (WT)] in 20-wk-old mice. As shown in Fig. 1A, aP2/H6PDH Tg mice showed no significant difference in body weight compared with WT littermates when animals were maintained on a normal chow diet. Similarly, there was no significant difference in food intake between the aP2/H6PDH Tg mice and WT littermates (data not shown). However, adipose fat pad weight was significantly increased in epididymal fat (P < 0.05) and subcutaneous fat (P < 0.05) pad with a 37% elevation of total fat mass (P < 0.05) in aP2/H6PDH Tg mice compared with age-matched WT control levels (Fig. 1B). Consistent with these observations, morphometric analysis of adipose tissue showed significant increase in adipocyte cell number within the fat tissue in the aP2/H6PDH mice (P < 0.05; Fig. 1C). Histological analysis showed that adipocyte size from the aP2/H6PDH mice was increased by 15% compared with WT controls (P < 0.05, Fig. 1, D and E). These data show that aP2/H6PDH Tg mice develop regional fat accumulation and mild obesity.

Fig. 1.

Body weight gain and fat mass of transgenic (Tg) mice overexpressing hexose-6-phosphate dehydrogenase (H6PDH) under the control of the enhancer-promoter region of the adipocyte fatty acid-binding protein (aP2) gene (aP2-H6PDH Tg). A: body weight gain of male aP2-H6PDH mice on a chow diet (n = 8–10 mice/group). B: increased fat mass in aP2-H6PDH Tg mice. Epi, epididymal fat; Mes, mesenteric fat; Sub, subcutaneous fat. Mice (n = 8–10/group) were killed at the age of 20 wk. C: increased adipocyte numbers in epididymal fat pad and subcutaneous fat pad of aP2-H6PDH Tg mice. D and E: epididymal fats were dissected and used for histology. Average area size of adipocytes on histological slides was measured, using ImageJ. Data are means ± SE. *P < 0.05 vs. wild-type (WT) littermates.

H6PDH expression and enzymatic activity are increased in white adipose tissue of transgenic mice.

To assess the overexpression of the H6PDH transgene in adipose tissue, we measured both endogenous and transgenic H6PDH mRNA and protein expression by real-time RT-PCR and Western blot. Figure 2A illustrates that H6PDH mRNA expression was significantly increased by approximately threefold in white epididymal and mesenteric fat with a more than twofold increase in subcutaneous fat of the aP2-H6PDH Tg mice compared with WT controls (P < 0.01). Similarly, H6PDH protein expression was markedly increased in the adipose tissue of the aP2-H6PDH Tg mice compared with WT controls (Fig. 2, B–D). Parallel to the increase in H6PDH expression, both epididymal and subcutaneous fat microsomal H6PDH activity, as determined by NADPH production, was increased threefold in the aP2-H6PDH Tg mice over WT controls (Fig. 2E), respectively. The endogenous H6PDH mRNA and protein expression in liver and skeletal muscle as well as in brain and kidney were not different between the aP2-H6PDH mice and WT groups (data not shown). Similarly, there was no significant difference in hypothalamic H6PDH mRNA level between the aP2/H6PDH Tg mice and WT controls (Fig. 2F).

Fig. 2.

Comparison of H6PDH mRNA and protein expression, and microsomal H6PDH activity, in epididymal fat (Epid fat) and subcutaneous fat (sub fat) of aP2-H6PDH Tg mice (Tg) and WT littermates (WT). A: relative expression of H6PDH mRNA levels was measured by quantitative real-time RT-PCR and normalized to 18S expression. B-D: the expression of epididymal fat (B) and subcutaneous fat (C) protein was done relative to the amount of GAPDH (D). E: H6PDH activity was measured in epididymal fat and subcutaneous fat microsomes on the basis of NADPH formation using 2 mM glucose 6-phosphate (G-6-P) as substrate in the presence of NADP. F: hypothalamic H6PDH mRNA levels in aP2-H6PDH Tg mice and WT littermates. Data are means ± SE of 6–8 mice/group. *P < 0.01 vs. WT controls; **P < 0.05 vs. WT control mice.

Transgenic aP2-H6PDH mice had enhanced endogenous adipose GC production in adipose tissue.

To determine if overexpression of H6PDH modulates GC action in vivo, we measured adipose 11βHSD1 activity, corticosterone levels, and GR expression in aP2-H6PDH Tg mice. As shown in Fig. 3A, 11β-HSD1 reductase activity was increased by 1.9-fold in epididymal fat, 1.5-fold in mesenteric fat (P < 0.01; Fig. 3A), and 1.4-fold in subcutaneous fat in aP2-H6PDH mice compared with controls. Consistent with these observations, epididymal and mesenteric and subcutaneous adipose 11β-HSD1 mRNA expression was increased by 3.0-, 1.7-, and 1.5-fold in aP2-H6PDH mice compared with their respective controls (P < 0.01; Fig. 3B). Similarly, adipose tissue corticosterone concentrations were significantly elevated by 42% (P < 0.01) in epididymal fat and by 38% (P < 0.01) in subcutaneous fat and by 16% (P < 0.05) in mesenteric fat in aP2-H6PDH mice compared with their respective controls (Fig. 3C), reflecting increased local corticosterone production. However, there was no significant difference in plasma corticosterone levels between aP2-H6PDH mice and WT controls (Fig. 3D). In contrast, real-time RT-PCR analysis showed that GR mRNA expression was elevated by 3.6-fold in epididymal fat (P < 0.01), 1.9-fold in mesenteric fat (P < 0.001), and a 1.7-fold (P < 0.05) in subcutaneous fat in aP2-H6PDH Tg mice compared with WT control levels (Fig. 3E). These data reflect increased adipose GC action in aP2-H6PDH Tg mice despite circulating GC levels. In addition, C/EBPα mRNA levels were significantly increased in epididymal fat (3.3-fold; P < 0.01), subcutaneous fat (2.7-fold; P < 0.01), and mesenteric fat (1.4-fold; P < 0.05) in aP2/H6PDH Tg mice compared with WT controls (Fig. 3F). Moreover, the changes in adipose 11β-HSD1 production and corticosterone concentrations were positively correlated with C/EBPα mRNA expression in aP-H6PDH Tg mice and WT controls (P < 0.01).

Fig. 3.

Adipose 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) and corticosterone levels in epididymal fat, mesenteric fat, and subcutaneous fat of aP2-H6PDH Tg mice and WT littermates. A: adipose 11β-HSD1 reductase activity was measured in adipose microsomes using 11-dehydrocorticosterone as substrate in the presence of NADPH. B, E, and F: expression and relative quantifications of 11β-HSD1, glucocorticoid receptor (GR), and C/EBPα mRNA levels were determined by real-time RT-PCR and normalized to 18S. C: tissue corticosterone concentrations in regional adipose tissue of aP2-H6PDH Tg mice and WT littermates. D: plasma corticosterone levels in aP2-H6PDH Tg mice and WT littermates. Data are means ± SE of 7–8 mice/group. *P < 0.01 and **P < 0.05 vs. WT controls.

aP2-H6PDH Tg mice showed impaired glucose tolerance and insulin sensitivity.

To further explore the impact of overexpression of adipose H6PDH on glucose homeostasis and insulin sensitivity, we performed GTTs and ITTs and compared plasma insulin level between the aP2/H6PDH Tg mice and age-matched WT controls. As shown in Fig. 4A, aP2/H6PDH mice in the fasting state were markedly hyperglycemic (155 ± 14 mg/dl) compared with WT controls (120 ± 7.7 mg/dl; P < 0.001). Consistent with plasma glucose elevation, aP2/H6PDH Tg mice showed higher glucose concentrations at all times during the GTT. Similarly, the area under the curve levels of glucose intolerance in the aP2/H6PDH Tg mice were significantly increased compared with WT controls (Fig. 4B). In the ITT, blood glucose concentration at 60, 90, and 120 min was higher in aP2-H6PDH Tg mice compared with that observed in WT control mice after intraperitoneal insulin challenge (P < 0.05; Fig. 4C). In agreement with these observations, fasting insulin levels were elevated in aP2-H6PDH Tg mice compared with that of WT controls (P = 0.015548; Fig. 4D), indicating that the aP2-H6PDH Tg mice exhibited signs of developing insulin resistance.

Fig. 4.

Glucose homeostasis in aP-H6PDH and WT controls. A: glucose tolerance test (GTT) of chow-fed transgenic and control mice aged 16 wk. B: the glucose area under the curve (AUC) during GTT. C: insulin tolerance test (ITT) in transgenic and WT control mice. ITT was performed on mice after an 8-h fast, and 0.75 U/kg insulin was used. D: the fasting plasma insulin levels in the aP-H6PDH Tg and WT controls. Data are means ± SE of 7–8 mice/group. *P < 0.01 and **P < 0.05 vs. WT control mice.

Transgenic aP2/H6PDH mice had increased circulating FFA levels and lipolysis in adipose tissue.

To investigate if the induction of endogenous adipose GC action may contribute to adipose lipolysis in aP2-H6PDH Tg mice, we measured the plasma FFA levels and the expression of key adipose lipolytic enzymes, HSL mRNA, and protein expression in adipose tissue. Biochemical analyses showed that aP2/H6PDH Tg mice had elevated plasma FFA levels compared with that of WT controls under both fed and fasted conditions (P < 0.01; Fig. 5A). Real-time RT-PCR analysis demonstrated that epididymal fat HSL mRNA level was increased by 2.1-fold (P < 0.01) in the fasting and 1.5-fold (P < 0.05) in fed condition compared with WT controls (Fig. 5B). Parallel to the increase in plasma FFA and adipose HSL mRNA expression, Western blot analysis confirmed that epididymal fat HSL protein levels were increased by 2.3-fold in the fasting state and 1.6-fold in the fed state in aP2-H6PDH Tg mice (P < 0.01 vs. respective WT mice) (Fig. 5, C and D). Consistent with elevated HSL expression, Ser⁶⁶⁰ phosphorylation of epididymal adipose HSL was induced by 3.3- and 2-fold (P < 0.01) in the aP2-H6PDH mice compared with WT animals under both fasted and fed conditions (Fig. 5, C and D), respectively. Similarly, the epididymal fat ATGL protein levels were increased by 3-fold in the fasting state (Fig. 6A) and 1.9-fold in the fed state (Fig. 6B) in aP2-H6PDH Tg mice (P < 0.01 vs. respective WT mice). These data suggest that activation of adipose lipolysis may contribute to elevated circulating FFA levels in the aP2-H6PDH Tg mice.

Fig. 5.

Serum free fatty acid (FFA) levels and adipose lipolytic gene expression in aP2-H6PDH Tg and their WT controls. A: serum FFA levels in WT and aP2-H6PDH Tg mice under both the fed and fasted state. B: quantitative real-time RT-PCR analysis demonstrating the relative alterations of hormone-sensitive lipase (HSL) mRNA expression in the epididymal fat of WT and aP2-H6PDH Tg mice. C and D: Western blot analysis of expression of epididymal fat total HSL and Ser⁶⁶⁰ phosphorylation (p) of HSL in the fasted condition (C) and in the fed state (D) of WT and aP2-H6PDH Tg mice. All protein was standardized to the amount of GAPDH. Data are means ± SE of 8 mice/group. *P < 0.01 and #P < 0.05 vs. WT controls.

Fig. 6.

ATGL expression in aP2-H6PDH Tg mice and WT littermates. Western blot analysis of epididymal fat ATGL protein expression in the fasted state (A) and in the fed condition (B) of WT and Tg mice. All protein was standardized to the amount of GAPDH. Data are means ± SE of 8 mice/group. **P < 0.01 vs. WT controls.

aP2/H6PDH mice increase FFA and glycerol release levels in isolated adipocytes.

To further determine if elevated circulating FFA levels caused by H6PDH overexpression could be observed in an in vitro setting, we measured FFA and glycerol release levels in the media in basal and isoproterenol-treated adipocytes from aP2/H6PDH mice compared with that of WT controls. Trypan blue exclusion assay revealed that the trypan blue uptake of the isolated adipocyte is comparable (2.8 ± 1.2% in aP2/H6PDH mice compared with 2.7 ± 1.1% in WT controls) (P > 0.05; Fig. 7A), and over 97% cells are viable. As shown in Fig. 7, B and C, basal released levels of FFA and glycerol in the media are higher in isolated adipocytes from aP2/H6PDH mice compared with that of WT, respectively. Treatment of cells with isoproterenol induced a significant increase in FFA release levels in isolated adipocytes from WT mice, but even stronger response to isoproterenol stimulation was observed in cells from aP2/H6PDH (Fig. 7B). Similarly, isoproterenol-stimulated glycerol release levels in these cells from WT mice reached only 60% of the levels seen in isolated adipocytes from aP2/H6PDH mice (Fig. 7C). These data indicate that aP2/H6PDH mice directly induced adipose FFA and glycerol release.

Fig. 7.

FFA and glycerol levels in adipocytes isolated from aP2-H6PDH Tg mice and WT littermates. A: trypan blue exclusion assay in isolated adipocytes. B: FFA release in response to isoproterenol stimulation in isolated adipocytes. C: glycerol release in response to isoproterenol stimulation in isolated adipocytes. Adipocytes were isolated from WT and Tg mice and incubated for 2 h without (basal) or with isoproterenol before measurement of FFA and glycerol released in the media. Values are means ± SE of 5–7 animals. †P < 0.05 vs. controls; *P < 0.01 and **P < 0.05 vs. isoproterenol-treated cells.

DISCUSSION

We demonstrated that aP2-H6PDH transgenic mice exhibited increased 11β-HSD1 activity and expression in response to elevated H6PDH expression in adipose tissue. We observed that corticosterone levels were increased in adipose tissue of aP2-H6PDH Tg mice that was accompanied by the induction of adipose microsomal 11β-HSD1 reductase activity, indicating increased conversion of corticosterone from 11-DHC.

Importantly, we found that aP2-H6PDH Tg mice showed fasting hyperglycemia and glucose intolerance along with elevated levels of the fasting plasma insulin and FFA, suggesting that aP2-H6PDH Tg mice exhibit signs of developing insulin resistance and metabolic syndrome. Our present data are consistent with a recent study reporting that increased adipose H6PDH expression may contribute to the development of T2DM in obese diabetic db/db mice (47). Adipose tissue of patients with T2DM have an increased (by ∼2-fold) level of H6PDH mRNA (45), and the conversion of cortisone to cortisol is elevated in these patients with induction of abdominal adipose 11β-HSD1 activity that is implicated to contribute to T2DM. The adipose H6PDH mRNA levels in aP2-H6PDH Tg mice are comparable to or slightly higher than that in adipose tissue of obese rodents and diabetic humans, suggesting that the phenotypic changes found in aP2-H6PDH Tg mice are physiologically and pathologically relevant. However, the aP2 promoter is not exclusively fat-specific; it is also reported to be expressed in macrophages and brains (4, 27), supporting the possibility that nonadipocyte expression of H6PDH may have contributed to their observed phenotype, although H6PDH Tg mice did not have altered hypothalamic H6PDH mRNA levels. Thus it is unlikely that nonadipocyte H6PDH gene expression accounts for the phenotype observed with H6PDH Tg mice.

Induction of H6PDH expression could enhance intracellular NADPH availability that drives 11β-HSD1 activity and consequently intracellular GC production that would lead to insulin resistance and hyperglycemia (25, 45, 47). In accordance with this concept, adipose 11β-HSD1 reductase activity was impaired in H6PDH knockout mice through unavailability of local NADPH, and these mutant mice exhibited an insulin-sensitive phenotype and fasting hypoglycemia (16). These data support the notion that the elevated H6PDH expression in our study is sufficient to increase local GC production in adipose tissues and associated fasting hyperglycemia and adverse metabolic phenotypes, as observed in the aP2-H6PDH Tg mice through induction of 11β-HSD1 activity. The findings in aP2-H6PDH Tg mice endorse transgenic overexpression of fat 11β-HSD1 that generates the metabolic syndrome through elevated adipose GC regeneration (29, 30). Thus, the aP2-H6PDH Tg mouse is an ideal model for studying the potential impact of elevated H6PDH in adipose 11β-HSD1 linked to the pathogenesis of metabolic syndrome.

It is well known that GCs are also capable of promoting adipose tissue differentiation, particularly in the central adipose regions, and their excess in circulation induces visceral obesity, insulin resistance, and T2DM, which are observed in patients with Cushing's syndrome (19, 28). Similarly, the aP2-11β-HSD1 mice show increased activity of adipose GCs and share some phenotypes to that observed in Cushing's syndrome. It is noteworthy that the aP2-H6PDH Tg mice showed mild epididymal fat and subcutaneous fat accumulation with a small amount of weight gain without significant abdominal obesity in response to elevated local GC production in the abdominal region that differs from that seen in aP2–11β-HSD1 mice that develop pronounced visceral obesity (29). Importantly, mesenteric fat GC concentrations are mildly increased but not to the degree that is observed in epididymal and subcutaneous fat in aP2-H6PDH Tg mice. In contrast, aP2–11β-HSD1 mice have more mesenteric fat mass with higher corticosterone production in mesenteric mass than seen in epididymal fat and subcutaneous fat. Histological analysis showed that the average adipose cell size from aP2-H6PDH Tg was relatively large; that is similar but less marked than observed in aP2-11β-HSD1 Tg mice (29). These data could explain the difference in obese phenotype between mice overexpressing 11β-HSD1 and H6PDH in adipocytes. Additionally, C/EBPα, the key lipogenic transcriptional regulator mRNA level for terminal adipogenesis that is stimulated by GCs (43, 50), was elevated in epididymal adipose and subcutaneous adipose tissue and less so in mesenteric fat of the aP2-H6PDH Tg mice compared with WT mice, suggesting that increased C/EBP expression may be crucially involved in local GC production-induced local adipose deposition in the aP2-H6PDH Tg mice. Moreover, C/EBPα is well known to be the key lipogenic gene transcriptor and is sufficient to trigger adipogenesis in adipocytes (49, 36, 35), which is a central feature of obesity and metabolic syndrome. Thus, adipose H6PDH overexpression-mediated activation of C/EBPα itself could promote local lipogenesis that could also contribute to the metabolic consequences observed in aP2-H6PDH Tg mice. In addition, C/EBP is required for the transcriptional activator of 11β-HSD1 in mouse adipocytes and liver (18, 48). Elevated C/EBP expression may thus provide an additional mechanism to maintain adipose 11β-HSD1 activation in the aP2-H6PDH Tg mice.

In adipose tissue, GCs are thought to be responsible, in part, for the elevated levels of circulating FFA during adipose lipolysis, and the prereceptor modulation of 11β-HSD1 is implicated as the mediator of GCs for induction of lipolysis (3, 9). HSL is the crucial enzyme for the hydrolysis of triacylglycerol to regulate plasma FFA levels and is stimulated by GCs. In the present study, aP2-H6PDH Tg mice exhibited a higher level of HSL mRNA and protein expression in response to increased local GC production in the abdominal fat and elevated levels of plasma FFA. This was confirmed by our finding that the aP2-H6PDH Tg mice effectively activated adipose Ser⁶⁶⁰ HSL phosphorylation, which is the key activator of HSL activity (2, 40). Similarly, adipose lipase ATGL protein expression was increased in aP2-H6PDH Tg mice. Moreover, FFA and glycerol released from isolated adipocytes were increased in aP2-H6PDH Tg mice. These data suggest that lipolytic activity is enhanced in adipose tissue. In contrast, lack of adipose GC activity in H6PDH knockout mice decreased HSL mRNA level but did not modulate plasma FFA levels (6). Moreover, increased HSL activity may lead to a lean phenotype (17) and thus would attenuate GC driving fat accumulation-induced obesity and weight gain observed in the aP2-H6PDH Tg mice. In accordance with this concept, inhibition of 11β-HSD1 limits GC exposure to adipose tissue and decreases lipolysis and weight gain (38, 44). These findings support the hypothesis that induction of fat lipolysis most likely occurs via increased adipose GC production-mediated activation of HSL activity in the aP2-H6PDH Tg mice.

In summary, the present study demonstrates that the aP2-H6PDH Tg mouse model has elevated intracellular production of GCs within the adipose tissue and exhibited features of the modest metabolic syndrome. The aP2-H6PDH Tg mice had a profound adipose lipolysis with resultant elevated plasma FFA level, which is thought to contribute to development of insulin resistance and metabolic syndrome. These findings also suggest that manipulation of H6PDH in adipose tissue may be a potential approach and can be translated into novel therapies for human T2DM and metabolic syndrome through prereceptor modulation of H6PDH and 11β-HSD1.

GRANTS

Y. Liu is supported by National Institutes of Health (NIH) Grants KO1-DK-073272 and SC1-DK-087655. T. C. Friedman is supported by NIH grant 2R24-DA-017298.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

YW, LL and YL contributed the study design and concept. YW, LL, HD and WF acquired and analyzed the data. YN and MJ generated aP2- H6PDH transgenic mouse line. YL drafted the manuscript and supervised this project. KL and TCF reviewed this manuscript for important intellectual content. All authors discussed and agreed on the results, and gave final approval.