Epoxyeicosatrienoic Acid Agonist Regulates Human Mesenchymal Stem Cell–Derived Adipocytes Through Activation of HO-1-pAKT Signaling and a Decrease in PPARγ

Abstract

Human mesenchymal stem cells (MSCs) expressed substantial levels of CYP2J2, a major CYP450 involved in epoxyeicosatrienoic acid (EET) formation. MSCs synthesized significant levels of EETs (65.8 ± 5.8 pg/mg protein) and dihydroxyeicosatrienoic acids (DHETs) (15.83 ± 1.62 pg/mg protein), suggesting the presence of soluble epoxide hydrolase (sEH). The addition of an sEH inhibitor to MSC culture decreased adipogenesis. EETs decreased MSC-derived adipocytes in a concentration-dependent manner, 8,9- and 14,15-EET having the maximum reductive effect on adipogenesis. We examined the effect of 12-(3-hexylureido)dodec-8(Z)-enoic acid, an EET agonist, on MSC-derived adipocytes and demonstrated an increased number of healthy small adipocytes, attenuated fatty acid synthase (FAS) levels (P < 0.01), and reduced PPARγ, C/EBPα, FAS, and lipid accumulation (P < 0.05). These effects were accompanied by increased levels of heme oxygenase (HO)-1 and adiponectin (P < 0.05), and increased glucose uptake (P < 0.05). Inhibition of HO activity or AKT by tin mesoporphyrin (SnMP) and LY2940002, respectively, reversed EET-induced inhibition of adipogenesis, suggesting that activation of the HO-1-adiponectin axis underlies EET effect in MSCs. These findings indicate that EETs decrease MSC-derived adipocyte stem cell differentiation by upregulation of HO-1-adiponectin-AKT signaling and play essential roles in the regulation of adipocyte differentiation by inhibiting PPARγ, C/EBPα, and FAS and in stem cell development. These novel observations highlight the seminal role of arachidonic acid metabolism in MSCs and suggest that an EET agonist may have potential therapeutic use in the treatment of dyslipidemia, diabetes, and the metabolic syndrome.

Introduction

M esenchymal stem cells (MSCs) are self-renewable multipotent stromal cells that have the ability to give rise to cells of diverse lineages and also modulate the inflammatory response through downregulation of proinflammatory molecules and upregulation of prosurvival and antiinflammatory molecules simultaneously [1]. Under conditions of increased levels of oxidative stress and increased levels of reactive oxygen species (ROS), MSCs exhibit increased senescence, poor differentiation, and a reduced capacity for tissue repair [2,3]. An increase in ROS occurs, resulting in a progressive deterioration in adipocyte and vascular function with a reduced number and function of circulating endothelial progenitor cells and elevation of inflammatory cytokines from adipose tissue [4,5]. Oxidative stress has been considered a major factor impairing MSC function leading to a decreased osteogenesis in favor of adipogenesis [6,7]. Due to these characteristics, MSCs have been studied for the use as potential cell-based transplantation therapy for treatment of diabetes and obesity [8,9].

Heme oxygenase (HO)-1 has been identified as a primary antioxidative defense system for hematopoietic stem cells [10]. We have shown that upregulation of HO-1 gene expression decreases adipokines such as IL-1 and IL-6 and increases levels of adiponectin, which is produced solely by adipocyte [11]. Further, an increase in HO-1 protein levels is associated with a parallel increase in the AMP-activated kinase (AMPK) and AKT signaling pathway [12 –14]. Increased HO-1 expression in obesity and type 2 diabetes results in a decrease in visceral and subcutaneous fat content, improved insulin sensitivity, and increased insulin receptor phosphorylation [15 –18]. Further, MRI studies showed that upregulation of HO-1 decreased adiposity and adipocytes hypertrophy [16,17,19]. A recent study from our lab also suggests that decreasing oxidative stress via induction of HO-1 shifts MSC differentiation in favor of osteogenic rather than adipogenic lineage [7].

The cytochrome P450 (P450)-derived epoxyeicosatrienoic acids (EETs) represent a class of lipid mediators with cytoprotective properties. Recently, we have shown that EETs decrease adiposity and insulin resistance in an animal model of obesity and diabetes via an increase in HO-1 gene expression and signaling cascade [20] including activation of AMPK and phosphorylated AKT (pAKT). Sacerdoti and coworkers [21,22] studied the interactions between HO-1 gene expression and EETs in vitro and showed that EETs act as inducers of HO-1 protein and HO activity. EETs are derived from arachidonic acid by the action of different CYP450 enzyme isoforms. These isoforms demonstrate tissue-specific expression and exhibit relative regioselectivity and stereospecificity. Members of the CYP2C and CYP2J families are the predominant epoxygenases in liver, kidney, brain, and blood vessels of rodents and humans [23,24]. Human stromal MSCs express CYP450 monooxygenase and form 20-HETE [25], but their ability to form EET has not been investigated. Moreover, nothing is known about the effect of EET on MSC-derived adipocyte HO-1, adiponectin, and downstream signaling cascade, including AMPK and pAKT. In this study, we report that the effect of EETs on MSCs suggests that MSC-derived adipocytes express CYP-epoxygenase 2J2 and its product activity, EETs. Treatment with EETs decreased adipocyte differentiation via an increase in HO-1 and decreased in PPARγ, C/EBPα, and FAS, suggesting that EETs are inhibitors of MSC-derived adipocyte stem cell progenitors and lipid homeostasis.

Materials and Methods

Human bone-marrow-derived MSC differentiation into adipocytes with treatment of EET and EET agonist and HO activity

Frozen bone marrow mononuclear cells were purchased from Allcells. After thawing, mononuclear cells were resuspended in an α-minimal essential medium (α-MEM; Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Invitrogen) and 1% antibiotic–antimycotic solution (Invitrogen). The cells were plated at a density of 1–5 × 10⁶ cells per 100 cm² dish. The cultures were maintained at 37°C in a 5% CO₂ incubator, and the medium was changed after 48 h and every 3–4 days thereafter. When the MSCs were confluent, the cells were recovered by the addition of 0.25% trypsin–ethylenediaminetetraacetic acid (Invitrogen). MSCs (Passage 2–3) were plated in a 75-cm² flask at a density of 1–2 × 10⁴ cells and cultured in α-MEM with 10% FBS for 7 days. The medium was replaced with the adipogenic medium, and the cells were cultured for an additional 21 days. The adipogenic medium consisted of a complete culture medium supplemented with Dulbecco's modified Eagle medium (DMEM)–high glucose, 10% (v/v) FBS, 10 μg/mL insulin, 0.5 mM dexamethasone (Sigma-Aldrich), 0.5 mM isobutylmethylxanthine (Sigma-Aldrich), and 0.1 mM indomethacin (Sigma-Aldrich).

Human MSCs were cultured to characterize the functional relevance of a numbers of EET agonists to HO-1-adiponectin and lipid content in adipocytes in the presence and absence of a soluble epoxide hydrolase (sEH) inhibitor, 12-(3-adamantan-1-yl-ureido)-dodecanoic acid (AUDA). During the adipogenesis, 5,6-, 8,9-, 11,12-, and 14,15-EET were added to human MSC cultures with or without an inhibitor of sEH to measure the efficiency of 11,12-EET and 14,15-EET to inhibit adipogenesis. EETs and AUDA were added every 3 days at a dose of 1 μM. Media were changed every 2 days. MSC-derived adipocytes were cultured in the adipogenic differentiation medium, and EET agonist was added every 3 days at a dose of 1 μM. After treatment with EET agonist, HO activity was determined by measuring the amount of CO generated in the culture medium during adipogenesis at day 14.

Detection of MSC cell markers by fluorescence-activated cell sorting analysis

Human MSCs are defined by an array of positive and negative markers. MSCs are normally plastic-adherent under standard culture conditions, expressing CD105, CD73, and CD90. MSCs must lack expression of CD45, CD34, CD14 or CD11b, CD79 or CD19, and HLA-DR. Also, MSCs must be able to differentiate into osteoblasts, adipocytes, and chondroblasts in vitro [26].

Human MSC phenotype was confirmed by flow cytometry (Elite ESP 2358; Beckman-Coulter) using several known MSC markers. The negative markers used were anti-CD34 and anti-CD45 (BD-Pharmingen), also known to be expressed as hematopoietic stem cell marker and common lymphocyte antigen. CD90, CD105, and CD166 were used as positive markers for MSCs. The data were analyzed using WinMDI 2.8 software.

Oil Red O staining

For Oil Red O staining, 0.21% Oil Red O in 100% isopropanol (Sigma-Aldrich) was used. Briefly, adipocytes were fixed in 10% formaldehyde, washed in Oil Red O for 10 min, and rinsed with 60% isopropanol (Sigma-Aldrich), and the Oil Red O was eluted by adding 100% isopropanol for 10 min and OD measured at 490 nm, for 0.5-s reading. MSC-derived adipocytes were measured by Oil Red O staining (OD = 490 nm) after day 14. Each values of Oil Red O staining were normalized by cell numbers (values at OD = 490 nm/10⁶ cells ratio). Harris and coworkers reported that cytochrome P450 arachidonic acid metabolites are potent mitogens for epithelial cells [27], so we normalized adipogenesis level by cell number.

Effect of EETs on lipid droplet size

After induction of adipogenesis, lipid droplets were stained with 2 μM BODIPY 493/503 (Molecular Probes), a specific stain for cellular lipid droplets [28]. Cell size was measured using an ImagePro Analyzer (Media Cybernetics). The classification of the size of lipid droplets was based on size by area (pixels).

Western blot analysis

Western blot analysis of adipocyte cell lysate was carried out as described previously [16,17,19]. Levels of the P450 epoxygenase CYP2J2, HO-1, and HO-2 were determined as described previously [20,29]. The phosphorylation of AKT and AMPK was analyzed by immunoblots with antibodies against phospho Ser473 AKT and phospho-Thr172 AMPK. Total AKT, AMPKα2 or -1, and β-actin were used as loading controls. Phosphorylation levels were quantified by scanning densitometry using an imaging densitometer, normalized to the levels of total protein. The relative phosphorylation in each signaling molecule was calculated relative to basal and/or control levels. FAS was measured by immunoblotting with the corresponding antibody and the level was normalized to loading β-actin and the results are presented as relative to the basal or to the control levels.

Glucose uptake

MSCs were cultured in the adipogenic medium containing 10 μg/mL insulin, 0.1 mM dexamethasone, 0.2 mM indomethacin, 10% FBS, and 1% antibiotic–antimycotic solution. At 50% confluence, EET agonist or vehicle solutions were added every 2 days for 14 days and glucose uptake was determined using ³H-[2]-deoxyglucose by the methods of Zen Bio. ³H-2-deoxyglucose was used as the substrate for glucose transporter 4 (GLUT4). The assays were performed in a 96-well plate with 100 nM insulin and 10 μM cytochalasin B used as the positive and negative controls, respectively. Samples were examined in triplicate.

Statistical analyses

Statistical significance (P < 0.05) between experimental groups was determined by the Fisher method of analysis of multiple comparisons. For comparison between treatment groups, the null hypothesis was tested by either a single-factor analysis of variance for multiple groups or unpaired t-test for 2 groups, and the data are presented as mean ± standard error.

Results

Determination of MSC phenotypes

MSCs were examined for both positive and negative markers by flow cytometry to characterize the MSCs. Confirmation of MSC phenotype was made by the presence of the positive markers, CD105 (86.7%), CD90 (99.8%), and CD166 (98.8%). The absence of CD3 (0%), a hematopoietic stem cell marker, and CD45 (0.02%), a lymphocytic marker, confirmed that MSC was not contaminated (Fig. 1). Our population of MSCs was found to be 86.7% positive for CD105, 99.8% positive for CD90, and 98.8% positive for CD166. There was <0.02% contamination by the negative markers, CD45 and CD34 (Fig. 1).

FIG. 1.

(A) Surface expression of CD90, CD105, and CD166 on human mesenchymal stem cells was analyzed by fluorescence-activated cell sorting analysis. Percent of cells expressing positive markers CD90 (Thy-1), CD105 (endoglin and SH2), and CD166 (activated leukocyte cell adhesion molecule) were 99.8%, 86.7%, and 98.8%, respectively. (B) CD45 and CD34 were shown to be expressed in <0.2% of cells, the same as unstained isotypes. CD45 (common lymphocytes antigen) and CD34 (hematopoietic stem cell marker) were used as negative markers.

Ability of MSC-derived adipocyte on manufacture EET, dihydroxyeicosatrienoic acid, and CYP2J2 expression

MSCs express higher levels of EETs compared to MSC-derived adipocyte. The basal levels of all 4 EET region isomers, including 14,15-EET, 11,12-EET, 8,9-EET, and 5,6-EET, in human MSCs were decreased in MSC-derived adipocyte. As shown in Fig. 2A, levels of EETs were 2 times lower in adipocytes (21.26 ± 2.109 pg/mg protein) compared with MSCs (65.8 ± 5.8 pg/mg protein). The dihydroxyeicosatrienoic acid (DHET) levels (Fig. 2B), which are a determinant for sEH activity, are higher in adipocytes (41.8 ± 3.24 pg/mg protein) than in MSCs (15.83 ± 1.62 pg/mg protein). Western blot analysis showed that MSCs display a substantial level of epoxygenase CYP2J2. MSC-derived adipocytes decreased epoxygenase CYP2J2 expression (P < 0.05), suggesting that epoxygenase activity decreased during MSC-derived adipocyte differentiation (Fig. 2C).

FIG. 2.

(A) Levels of EETs were 2 times lower in adipocytes (21.26 ± 2.109 pg/mg protein) compared with MSCs (65.8 ± 5.8). (B) The dihydroxyeicosatrienoic acid (DHET) levels are higher in adipocytes (41.8 ± 3.24) compared to MSCs (15.83 ± 1.62). (C) Representative western blot and densitometry analysis of CYP2J2 protein in MSCs and adipocytes. Values are means ± standard error (n = 3; *P < 0.05). EET, epoxyeicosatrienoic acids; MSCs, mesenchymal stem cells.

Effect of 5,6-, 8,9-, 11,12-,14,15-EETs and AUDA on adipogenesis

Four different epoxide regioisomers—5,6-, 8,9-, 11,12-, and 14,15-EET—are generated by epoxygenase activity. Daily supplementation of 5,6-, 8,9-, 11,12-, and 14,15-EET was effective in adipogenesis suppression at 7 or 14 days. Addition of the sEH inhibitor AUDA increased the inhibitory effect of 11,12-EET and 14,15-EET on adipogenesis (Fig. 3A). These results are quantified (Fig. 3B) and show that 8,9-EET and 14,15-EET (P < 0.01) are more effective in suppressing adipogenesis than 5,6-EET and 11,12-EET, and that AUDA had a synergistic effect on adipogenesis suppression.

FIG. 3.

(A, B) Mesenchymal stem cells were cultured to characterize the functional relevance of a number of EETs to lipid content in adipocytes in the presence and absence of a soluble epoxide hydrolase inhibitor (AUDA). n = 4; *P < 0.05 versus 5,6- and 11,12-EET; **P < 0.001 versus 8,9-, 14,15-, 11,12-EET+AUDA, and 14,15-EET+AUDA. AUDA, 12-(3-adamantan-1-yl-ureido)-dodecanoic acid; EET, epoxyeicosatrienoic acids.

Effect of the 12-(3-hexylureido) dodec-8(Z)-enoic acid on adipogenesis

We investigated the effect of the EET agonist 12-(3-hexylureido)dodec-8(Z)-enoic acid on adipogenesis using standard culture conditions. The effect of the EET agonist on adipogenesis was determined by counting cells with lipid droplets in the cytoplasm and cells positive for the lipid-specific dye, BODIPY (Fig. 4A). The percentage of cells with morphological lipid droplets was decreased as EET concentration increased. In untreated MSC–adipocytes grown in the adipogenesis differentiation medium, formation of lipid droplets was detected: 16.9% ± 3.1% of the cells by either morphological or BODIPY analysis (n = 6). EET agonist decreased adipocyte differentiation in a concentration-dependent manner (Fig. 4A). BODIPY staining was barely detectable at 10 μM, although the cells became loaded with lipid droplets in untreated MSC-derived adipocyte progenitor cells. Quantification of BODIPY-stained cells showed an increase in adipocytes with the absence of the EET agonist (250 ± 2) compared with the presence of the EET agonist (100 ± 2.1 and 50 ± 1.8 pixels) at 1 and 10 μM, respectively. In agreement with this observation, the average size of lipid droplet in the presence of EET agonist decreased in a dose-dependent manner (Fig. 4A).

FIG. 4.

Pharmacological effect of EET agonist on MSC-derived adipocyte cell differentiation. (A) Dose–response effect of EET agonist on phase-contrast image under the fluorescence microscopy. Lipid droplets were stained with BODIPY and size of droplets were measured using ImagePro Analyzer (version 6.2, Media Cybernetics; *P < 0.05 vs. 1 and 2 μM; **P < 0.01 vs. 10 μM). (B) Effect of EET agonist on PPARγ and C/EBPα expression in MSC-derived adipocytes. (*P < 0.005 vs. day 5; **P < 0.01 vs. day 10). EET, epoxyeicosatrienoic acids; MSCs, mesenchymal stem cells.

Effect of EET agonist on HO-1 protein, HO activity, and lipid content in MSC-derived adipocytes

The induction of adipocytes by culturing cells in the adipogenic medium resulted in a significant (P < 0.01) decrease in HO-1 protein levels (Fig. 5A). This decrease was partially restored by culturing the cells in the presence of the EET agonist 12-(3-hexylureido)dodec-8(Z)-enoic acid [30]. HO-1 protein levels were significantly (P < 0.05) increased in MSC-derived adipocytes by treatment with the EET agonist and displayed a 2-fold increase in HO-1 protein levels (Fig. 5A). In addition, HO activity, as measured by CO release, was increased (P < 0.05) in the presence of the EET agonist. The basal levels of HO activity were 152 ± 11 (pmol CO formed/mg/h) compared to 256 ± 21 (pmol CO formed/mg/h) (Fig. 5B). We further examined the effect of SnMP on adipogenesis and lipid content. SnMP caused an increase in adipogenesis (P < 0.01) that was reversed by the presence of the EET agonist. SnMP and EET agonist together caused an increase in adipogenesis (P < 0.01), indicating that the effect of SnMP superseded that of the EET agonist (Fig. 5C). Similar results were seen with lipid content (Fig. 5D).

FIG. 5.

(A) Effect of EET agonist on HO-1 expression in MSC-derived adipocytes (**P < 0.001 vs. MSC *P < 0.05 vs. EET agonist). (B) HO activity was measured by CO formation (*P < 0.05 vs. EET agonist). (C, D) Effect of SnMP or EET agonist on lipid droplets. Data are expressed as mean ± standard error (n = 4; *P < 0.0001 vs. SnMP, EET agonist+SnMP). EET, epoxyeicosatrienoic acids; HO, heme oxygenase; MSCs, mesenchymal stem cells; SnMP, tin mesoporphyrin.

Effect of EET agonist on the adipogenesis markers PPARγ and C/EBPα in MSC-derived adipocytes

We examined the effect of an EET agonist on PPARγ and C/EBPα expression as adipogenic differentiation markers at days 5 and 10. Densitometry analysis showed that the levels of PPARγ and C/EBPα were increased on both day 5 (P < 0.05) and day 10 (P < 0.01) compared to MSCs. EET-agonist-treated adipocytes displayed a decrease in PPARγ and C/EBPα levels, whereas PPARγ and C/EBPα levels were increased during adipogenesis (Fig. 4A, B).

Effect of EET agonist on adiponectin, pAKT, pAMPK, FAS, and glucose uptake

As shown in Fig. 6A adiponectin levels were increased (P < 0.05) by the EET agonist and this effect was reversed by SnMP (Fig. 6A). Phosphorylation of AKT was increased by culturing adipocytes with the EET agonist, while no effect on AKT was seen (Fig. 6B). The presence of EET reversed the effect of the adipogenic medium and increased activation of pAKT. The changes in protein expression of pAKT mirrored those seen with HO-1 protein expression (Fig. 5A). Thus, EET increased pAKT and HO-1 [22,31 –33]. Densitometry analysis showed that pAMPK levels were no different between MSC-derived adipocytes and those treated with EET agonist (Fig. 6B). To further examine if EET agonist-dependent HO-1-pAKT regulates glucose-induced lipid accumulation and decreased fatty acid synthesis in adipocytes, the effect of the EET agonist on protein levels was determined in MSC-derived adipocytes. As seen in Fig. 6B, untreated adipocytes displayed a marked increase in FAS levels, while HO-1 levels were decreased during adipogenesis. The increase in FAS was prevented by the EET agonist at concentrations ranging from 1 to 2 μM, reaching a level comparable to that in either MSC. Glucose uptake in MSCs treated with 1 μM EET agonist was significantly (P < 0.05) increased compared to untreated MSCs after adipogenesis (day 14), indicating an increase in the function of GLUT4. This was reserved by the HO inhibitor, SnMP, in a significant (P < 0.01) manner (Fig. 6C).

FIG. 6.

(A) Effect of SnMP and EET agonist on adiponectin levels (n = 3; *P < 0.0001 vs. control and SnMP). (B) Representative immunoblotting analysis with antibodies against phosphorylated AKT (pAKT) and pAMPK (*P < 0.05 vs. EET agonist). (C) Effect of EET agonist on glucose uptake in MSC-derived adipocytes (n = 3; *P < 0.05 vs. EET agonist; ^# P < 0.01 vs. EET+SnMP). EET, epoxyeicosatrienoic acids; FAS, fatty acid synthase; MSCs, mesenchymal stem cells; SnMP, tin mesoporphyrin. pAMPK, phosphorylated AMP-activated kinase.

Pharmacological inhibition of EET-mediated adipogenesis by LY294002

We examined the effect of LY2940002 on adipogenesis, which was measured as the relative absorbance of Oil Red O at day 14. EET agonist decreased the levels of Oil Red O staining compared with control (P < 0.02), whereas LY294002 increased Oil Red O staining in MSC-derived adipocytes (P < 0.05). Inhibition of AKT by treatment with LY2940002 (5 μM) strongly induced adipogenesis on day 14 (P < 0.05) (Fig. 7). Additionally, as seen in Fig. 7A, LY294002 prevented the decrease of adipogenesis in EET-treated cells. EET agonist treatment during adipogenesis decreases lipid droplets compared to control.

FIG. 7.

(A) Effect of EET agonist and LY2940002 on adipogenesis. (B) Adipogenesis was measured as the relative absorbance of Oil Red O at day 14 after inducing adipogenesis as described in the Materials and Methods section. Bars represent the mean ± standard error of the mean of 4 independent experiments (*P < 0.02 vs. EET, LY2940002+EET; **P < 0.05 vs. LY2940002). EET, epoxyeicosatrienoic acids.

Discussion

This study shows that MSC-derived adipocytes exhibited decreased activity of the arachidonic acid metabolic pathway that yields EETs, that is, epoxygenase levels. The inability of MSC-derived adipocytes to sustain normal levels of EETs may be the result of the increased levels of sEH coupled with decreased expression of P450 epoxygenases. We further report that expression of CYP2J2 was significantly decreased in adipocytes. Further, we demonstrate that MSCs stay in a pluripotent condition. We have previously shown that MSCs are pleiotrophic cells that can differentiate to other lineage such as osteoblasts as a result of crosstalk by specific signaling pathways, including HO-1/-2 expression [7,34].

In this report, we show that epoxygenase product activities, EETs, are effective in suppression of adipogenesis. 8,9-EET is more effective in suppression of adipogenesis compared to 5,6- and 11,12-EET but equally effective as 14,15-EET. Further, inhibition of sEH potentiated the EET-mediated decrease of adipogenesis. Adipocyte stem cells in culture treated with AUDA, an inhibitor of sEH, caused an increase in effectiveness of both 11,12- and 14,15-EET in suppression of adipocyte stem cell differentiation. The antiadipogenic effect of an EET agonist, when taken together with the inhibition of sEH, highlights the therapeutic potential of EET in the management of cardiovascular disease [35,36]. An association of sEH gene polymorphism with insulin resistance has been reported, implying that sEH plays an important role in the pathogenesis of insulin resistance [37]. EET agonist decreased O₂ ⁻ production and prevented the rapid degradation of EET and subsequent activation of pAKT [32,38]. Thus, EETs appear capable of reprogramming adipocyte stem cells, resulting in expression of a new phenotype that contains adipocytes of reduced cell size, that are associated with an increase in adiponectin and a decrease in inflammatory cytokines. In this report, we also demonstrate in vitro that the EET agonist 12-(3-hexylureido)dodec-8(Z)-enoic acid is very effective in suppression of adipogenesis and that suppression occurs in a dose-dependent manner.

Several reports show that oxidative stress and O₂ ⁻ reduce EET levels [33,39,40]. EETs are rapidly degraded by O₂ ⁻ to DHET [40], and EETs are also inactivated by sEH to DHETre [33]. Since EET agonists induce HO-1 and vice versa, HO-1 induction increases EET levels [22]. HO-1-mediated induction of EETs may be related to the ability of HO-1 to decrease oxidative stress and O₂ ⁻. Previously, it was shown that overexpression of HO-1 attenuated and AngII mediated oxidative stress [41] and vascular injury and dysfunction in hypertensive rats [42]. The mechanism by which HO-1 attenuated ROS involves an increase in extracellular superoxide dismutase (EC-SOD) and the restoration of mitochondrial function [43]. In addition, a lack of HO-2 creates a setting that promotes oxidative-stress-related disturbances, including increases in O₂ ⁻ and decreases in EC-SOD [44]. The fact that HO-1 and −2 serve an antioxidative function and preserve EET levels suggests that the activation of this system in adipose tissue in obesity, a condition of high oxidative stress, represents an adaptive mechanism that confers MSCs resistance against oxidative stress and inhibits adipogenesis.

The role of EETs in decreasing oxidative stress via an increase in HO-1-mediated adiponectin and pAMPK is in agreement with the report that EETs increase ERK 1/2 MAP kinase.

EETs have been shown to mediate MAP kinase activation and ERK 1/2 MAP kinase phosphorylation [32]. To date, the effect of EET on adipocyte MAP kinase has not been investigated. In addition, the crosstalk between AMPK–AKT and activation of AMPK is essential for the cellular processes that are controlled by the energy state of the cell. AMPK is activated by a decrease in ATP and rise in cellular AMP [45,46], which leads to the phosphorylation of eNOS and key enzymes that subsequently inhibit the synthesis of cholesterol and increase glucose uptake. Activation of AMPK in skeletal muscle leads to an enhancement of glucose transport mediated by the translocation of GLUT4 to the membrane and this appears to be additive to the stimulation in response to insulin [45]. In addition, pharmacological activation of AMPK by AICAR in obese Zucker rats improves glucose tolerance and reduces systolic blood pressure [47].

This study provides direct evidence that the EET-agonist-mediated inhibition of adipogenesis is accompanied by the decrease of FAS, PPARγ, and C/EBPα in MSC-derived adipocyte stem cells. Additionally, these perturbations occur in sequence, commencing with increased levels of HO-1 expression and decreased lipid accumulation. The lipid-lowering effect of the EET agonist was completely blocked by pharmacological suppression of HO activity. In addition, PPARγ and C/EBPα are known to increase adipogenesis [48]. The ability of the EET agonist to stimulate pAKT and decrease FAS, PPARγ, and C/EBPα supports this hypothesis and that EETs have a negative effect on adipogenesis. These effects can be duplicated by inducers of HO-1 such as CoPP and/or L-4F [7,18], suggesting that HO-1 plays a key role in lipid metabolism. These novel observations underscore the importance of EETs in regulation of MSCs to adipocyte lineages.

Since pharmacological activation of AKT and AMPK in obese Zucker rats improves obesity and glucose tolerance and reduces systolic blood pressure [49], our finding of the effect of the EET-HO-1 module on AKT activation in adipocytes may be crucial to increase glucose uptake, lipid homeostasis, and inhibition of PPARγ and C/EBPα.

FAS mRNA levels were shown to be increased dramatically during 3T3-L1 adipocyte differentiation [50]. In our experiments, expression of FAS, PPARγ, and C/EBPα increased during adipogenesis; however, FAS, PPARγ, and C/EBPα expression decreased after EET agonist treatment. The action of EET agonist treatment as manifest by increased levels of HO-1 and pAKT is associated in an improvement in glucose uptake. Further, EET agonist effectively restored expression of adiponectin, which was accompanied with a significant increase in cellular glucose uptake. In agreement with our results, adiponectin-deficient cells showed marked downregulation of GLUT4, and adipose triglyceride lipase [51]. As seen in Fig. 7, inhibition of pAKT by LY294002 increased adipogenesis. In agreement with this, LY294002 was shown to inhibit GLUT4 translocation [52]. This suggests that EET agonist treatment may increases translocation of GLUT4.

Conclusions

We have presented novel results that indicate the existence of epoxygenase-mediated generation of EETs in MSCs and a molecular crosstalk between EETs and HO-1 that regulates MSC–adipocyte stem cell differentiation and development to mature adipocytes. This novel action of EETs provides a mechanistic basis for the EET-mediated control of adipogenesis via HO-1 and adiponectin (Fig. 8). In support of this conclusion, EET agonist administration has been shown to inhibit adiposity, increase insulin sensitivity, and improve vascular function in obese animal model [20]. Thus, targeting MSCs to increase EET levels could be employed therapeutically to address the metabolic impairment in MSC-derived adipocyte function associated with vascular diseases, including obesity, diabetes, and hypertension at levels of MSCs.

FIG. 8.

Proposed mechanism for the EET agonist-mediated suppression of MSCs-derived adipocyte differentiation and lipid accumulation. EET agonist-activating HO-1 expression increase phosphorylation of AMPK and AKT which in turn decrease FAS, thereby leading to decrease in lipid droplets.

Footnotes

Acknowledgments

This work was supported by NIH grants DK068134, HL55601 (N.G.A.), and HL34300 (M.L.S.), and The Robert A. Welch Foundation and GM31278 (J.R.F.). This research was also supported, in part, by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (Z01 ES025034)(DZ). The authors are indebted to Dr. Attallah Kappas and The Beatrice Renfield Foundation for their support.

Author Disclosure Statement

No competing financial interests exist.

References

Salem

, Thiemermann

. 2010. Stem Cells, 28:585–596Review.

Song

, Cha

, Song

, Kim

, Chang

, Lim

, Choi

, Ham

, Lee

, Chung

, Jang

, Hwang

. 2010. Reactive oxygen species inhibit adhesion of mesenchymal stem cells implanted into ischemic myocardium via interference of focal adhesion complex. Stem Cells, 28:555–563.

Bai

, Lu

, Bai

, Zheng

, Ke

, Li

, Luo

. 2004. Oxidative stress inhibits osteoblastic differentiation of bone cells by ERK and NF-kappaB. Biochem Biophys Res Commun, 314:197–207.

Sambuceti

, Morbelli

, Vanella

, Kusmic

, Marini

, Massollo

, Augeri

, Corselli

, Ghersi

, Chiavarina

, Rodella

, L’ Abbate

, Drummond

, Abraham

, Frassoni

. 2009. Diabetes impairs the vascular recruitment of normal stem cells by oxidant damage; reversed by increases in pAMPK, heme oxygenase-1 and adiponectin. Stem Cells, 27:399–407.

Wellen

, Hotamisligil

. 2003. Obesity-induced inflammatory changes in adipose tissue. J Clin Invest, 112:1785–1788.

Houstis

, Rosen

, Lander

. 2006. Reactive oxygen species have a causal role in multiple forms of insulin resistance. Nature, 440:944–948.

Vanella

, Kim

, Asprinio

, Peterson

, Barbagallo

, Vanella

, Goldstein

, Ikehara

, Abraham

. 2009. HO-1 expression increases mesenchymal stem cell-derived osteoblast and decrease adipocyte lineages. Bone, 46:236–243.

Verda

, Kim

, Ikehara

, Statkute

, Bronesky

, Petrenko

, Oyama

, He

, Link

, Vahanian

, Burt

. 2008. Hematopoietic mixed chimerism derived from allogeneic embryonic stem cells prevents autoimmune diabetes mellitus in NOD mice. Stem Cells, 26:381–386.

Abraham

, Li

, Vanella

, Peterson

, Ikehara

, Asprinio

. 2008. Bone marrow stem cell transplant into intra-bone cavity prevents type 2 diabetes: role of heme oxygenase-adiponectin. J Autoimmun, 30:128–135.

10.

Abraham

. 1991. Molecular regulation—biological role of heme in hematopoiesis. Blood Rev, 5:19–28.

11.

Iwaki

, Matsuda

, Maeda

, Funahashi

, Matsuzawa

, Makishima

, Shimomura

. 2003. Induction of adiponectin, a fat-derived antidiabetic and antiatherogenic factor, by nuclear receptors. Diabetes, 52:1655–1663.

12.

Kusmic

, L'Abbate

, Sambuceti

, Drummond

, Barsanti

, Matteucci

, Cao

, Piccolomini

, Cheng

, Abraham

. 2010. Improved myocardial perfusion in chronic diabetic mice by the up-regulation of pLKB1 and AMPK signaling. J Cell Biochem, 109:1033–1044.

13.

, Keaney

Jr.

2010. AMP-activated protein kinase: a stress-responsive kinase with implications for cardiovascular disease. Curr Opin Pharmacol, 10:111–115Review.

14.

Walsh

. 2009. Adipokines, myokines and cardiovascular disease. Circ J, 73:13–18.

15.

Kim

, Burgess

, Li

, Tsenovoy

, Addabbo

, McClung

, Puri

, Abraham

. 2008. Heme oxygenase-mediated increases in adiponectin decrease fat content and inflammatory cytokines, tumor necrosis factor-alpha and interleukin-6 in Zucker rats and reduce adipogenesis in human mesenchymal stem cells. J Pharmacol Exp Ther, 325:833–840.

16.

, Kim

, Tsenovoy

, Peterson

, Rezzani

, Rodella

, Aronow

, Ikehara

, Abraham

. 2008. Treatment of obese diabetic mice with a heme oxygenase inducer reduces visceral and subcutaneous adiposity, increases adiponectin levels, and improves insulin sensitivity and glucose tolerance. Diabetes, 57:1526–1535.

17.

Nicolai

, Li

, Kim

, Peterson

, Vanella

, Positano

, Gastaldelli

, Rezzani

, Rodella

, Drummond

, Kusmic

, L'Abbate

, Kappas

, Abraham

. 2009. Heme oxygenase-1 induction remodels adipose tissue and improves insulin sensitivity in obesity-induced diabetic rats. Hypertension, 53:508–515.

18.

Peterson

, Drummond

, Hyun

, Li

, Kruger

, Ikehara

, Abraham

. 2008. L-4F treatment reduces adiposity, increases adiponectin levels and improves insulin sensitivity in obese mice. J Lipid Res, 49:1658–1669.

19.

Peterson

, Kim

, Li

, Positano

, Vanella

, Rodella

, Piccolomini

, Puri

, Gastaldelli

, Kusmic

, L'Abbate

, Abraham

. 2009. The L-4F mimetic peptide prevents insulin resistance through increased levels of HO-1, pAMPK, and pAKT in obese mice. J Lipid Res, 50:1293–1304.

20.

Sodhi

, Inoue

, Gotlinger

, Canestraro

, Vanella

, Kim

, Manthati

, Koduru

, Falck

, Schwartzman

, Abraham

. 2009. Epoxyeicosatrienoic acid agonist rescues the metabolic syndrome phenotype of HO-2-null mice. J Pharmacol Exp Ther, 331:906–916.

21.

Sacerdoti

, Bolognesi

, Di

, Gatta

, McGiff

, Schwartzman

, Abraham

. 2006. Rat mesenteric arterial dilator response to 11,12-epoxyeicosatrienoic acid is mediated by activating heme oxygenase. Am J Physiol Heart Circ Physiol, 291:H1999–H2002.

22.

Sacerdoti

, Colombrita

, Di

, Schwartzman

, Bolognesi

, Falck

, Gatta

, Abraham

. 2007. 11,12-Epoxyeicosatrienoic acid stimulates heme-oxygenase-1 in endothelial cells. Prostaglandins Other Lipid Mediat, 82:155–161.

23.

Zeldin

. 2001. Epoxygenase pathways of arachidonic acid metabolism. J Biol Chem, 276:36059–36062.

24.

Campbell

, Falck

. 2007. Arachidonic acid metabolites as endothelium-derived hyperpolarizing factors. Hypertension, 49:590–596.

25.

Abraham

, Feldman

, Falck

, Lutton

, Schwartzman

. 1991. Modulation of erythropoiesis by novel human bone marrow cytochrome P450-dependent metabolites of arachidonic acid. Blood, 78:1461–1466.

26.

Grimm

, Baeuerle

. 1993. The inducible transcription factor NF-kappa B: structure-function relationship of its protein subunits. Biochem J, 290,Pt 2:297–308.

27.

Burns

, Capdevila

, Wei

, Breyer

, Homma

, Harris

. 1995. Role of cytochrome P-450 epoxygenase metabolites in EGF signaling in renal proximal tubule. Am J Physiol, 269:C831–C840.

28.

Tavian

, Colombo

. 2007. Improved cytochemical method for detecting Jordans' bodies in neutral lipid storage diseases. J Clin Pathol, 60:956–958.

29.

Seubert

, Yang

, Bradbury

, Graves

, Degraff

, Gabel

, Gooch

, Foley

, Newman

, Mao

, Rockman

, Hammock

, Murphy

, Zeldin

. 2004. Enhanced postischemic functional recovery in CYP2J2 transgenic hearts involves mitochondrial ATP-sensitive K+ channels and p42/p44 MAPK pathway. Circ Res, 95:506–514.

30.

Falck

, Kodela

, Manne

, Atcha

, Puli

, Dubasi

, Manthati

, Capdevila

, Yi

, Goldman

, Morisseau

, Hammock

, Campbell

. 2009. 14,15-Epoxyeicosa-5,8,11-trienoic acid (14,15-EET) surrogates containing epoxide bioisosteres: influence upon vascular relaxation and soluble epoxide hydrolase inhibition. J Med Chem, 52:5069–5075.

31.

Sacerdoti

, Abraham

, Oyekan

, Yang

, Gatta

, McGiff

. 2004. Role of the heme oxygenases in abnormalities of the mesenteric circulation in cirrhotic rats. J Pharmacol Exp Ther, 308:636–643.

32.

Gutterman

. 2002. Vascular dysfunction in hyperglycemia: is protein kinase C the culprit? Circ Res, 90:5–7.

33.

Imig

. 2005. Epoxide hydrolase and epoxygenase metabolites as therapeutic targets for renal diseases. Am J Physiol Renal Physiol, 289:F496–F503.

34.

Barbagallo

, Vanella

, Peterson

, Kim

, Tibullo

, Giallongo

, Vanella

, Parrinello

, Palumbo

, Di Raimond

, Abraham

, Asprinio

. 2010. Over expression of heme oxygenase-1 increases human osteoblast stem cell differentiation. J Bone Miner Metab, 28:276–288.

35.

Spector

, Norris

. 2007. Action of epoxyeicosatrienoic acids on cellular function. Am J Physiol Cell Physiol, 292:C996–C1012.

36.

Imig

. 2006. Cardiovascular therapeutic aspects of soluble epoxide hydrolase inhibitors. Cardiovasc Drug Rev, 24:169–188.

37.

Ohtoshi

, Kaneto

, Node

, Nakamura

, Shiraiwa

, Matsuhisa

, Yamasaki

. 2005. Association of soluble epoxide hydrolase gene polymorphism with insulin resistance in type 2 diabetic patients. Biochem Biophys Res Commun, 331:347–350.

38.

Bodiga

, Zhang

, Jacobs

, Larsen

, Tampo

, Manthati

, Kwok

, Zeldin

, Falck

, Gutterman

, Jacobs

, Medhora

. 2009. Protective actions of epoxyeicosatrienoic acid: dual targeting of cardiovascular PI3K and KATP channels. J Mol Cell Cardiol, 46:978–988.

39.

Larsen

, Gutterman

, Sato

, Toyama

, Campbell

, Zeldin

, Manthati

, Falck

, Miura

. 2008. Hydrogen peroxide inhibits cytochrome p450 epoxygenases: interaction between two endothelium-derived hyperpolarizing factors. Circ Res, 102:59–67.

40.

Dhanasekaran

, Al-Saghir

, Lopez

, Zhu

, Gutterman

, Jacobs

, Medhora

. 2006. Protective effects of epoxyeicosatrienoic acids on human endothelial cells from the pulmonary and coronary vasculature. Am J Physiol Heart Circ Physiol, 291:H517–H531.

41.

Asija

, Peterson

, Stec

, Abraham

. 2007. Targeting endothelial cells with heme oxygenase-1 gene using VE-cadherin promoter attenuates hyperglycemia-mediated cell injury and apoptosis. Antioxid Redox Signal, 12:2065–2074.

42.

Olszanecki

, Rezzani

, Omura

, Stec

, Rodella

, Botros

, Goodman

, Drummond

, Abraham

. 2007. Genetic suppression of HO-1 exacerbates renal damage: reversed by an increase in the antiapoptotic signaling pathway. Am J Physiol Renal Physiol, 292:F148–F157.

43.

Di Noia

, Van

, Palmieri

, Yang

, Quan

, Goodman

, Abraham

. 2006. Heme oxygenase-1 enhances renal mitochondrial transport carriers and cytochrome C oxidase activity in experimental diabetes. J Biol Chem, 281:15687–15693.

44.

Turkseven

, Drummond

, Rezzani

, Rodella

, Quan

, Ikehara

, Abraham

. 2007. Impact of silencing HO-2 on EC-SOD and the mitochondrial signaling pathway. J Cell Biochem, 100:815–823.

45.

Murgia

, Elbenhardt

, Cusinato

, Garcia

, Richter

, Schiaffino

. 2009. Multiple signalling pathways redundantly control GLUT4 gene transcription in skeletal muscle. J Physiol, 587:4319–4327.

46.

Habets

, Coumans

, El

, Zarrinpashneh

, Bertrand

, Viollet

, Kiens

, Jensen

, Richter

, Bonen

, Glatz

, Luiken

. 2009. Crucial role for LKB1 to AMPKalpha2 axis in the regulation of CD36-mediated long-chain fatty acid uptake into cardiomyocytes. Biochim Biophys Acta, 1791:212–219.

47.

Buhl

, Jessen

, Pold

, Ledet

, Flyvbjerg

, Pedersen

, Schmitz

, Lund

. 2002. Long-term AICAR administration reduces metabolic disturbances and lowers blood pressure in rats displaying features of the insulin resistance syndrome. Diabetes, 51:2199–2206.

48.

Ross

, Hemati

, Longo

, Bennett

, Lucas

, Erickson

, MacDougald

. 2000. Inhibition of adipogenesis by Wnt signaling. Science, 289:950–953.

49.

Elmarakby

, Imig

. 2010. Obesity is the major contributor to vascular dysfunction and inflammation in high fat diet hypertensive rats. Clin Sci (Lond), 118:291–301.

50.

Moustaid

, Sul

. 1991. Regulation of expression of the fatty acid synthase gene in 3T3-L1 cells by differentiation and triiodothyronine. J Biol Chem, 266:18550–18554.

51.

, Li

, Yang

, Liu

, Li

, Boden

. 2009. Effect of shRNA-mediated adiponectin/Acrp30 down-regulation on insulin signaling and glucose uptake in the 3T3-L1 adipocytes. J Endocrinol Invest, 33:96–102.

52.

Funaki

, Randhawa

, Janmey

. 2004. Separation of insulin signaling into distinct GLUT4 translocation and activation steps. Mol Cell Biol, 24:7567–7577.