HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 14, 10853-10860, April 6, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/14/10853    most recent M008010200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chevillotte, E. Articles by Vidal, H. Search for Related Content PubMed PubMed Citation Articles by Chevillotte, E. Articles by Vidal, H. Social Bookmarking                      What's this?

The Regulation of Uncoupling Protein-2 Gene Expression by -6 Polyunsaturated Fatty Acids in Human Skeletal Muscle Cells Involves Multiple Pathways, Including the Nuclear Receptor Peroxisome Proliferator-activated Receptor ^*

Emmanuel Chevillotte, Jennifer Rieusset, Marina Roques, Michel Desage^§, and Hubert Vidal

From the INSERM U449, Faculté de Médecine René Laennec, Université Claude Bernard Lyon-1, and ^§ CRNHL Faculté de Médecine René Laennec, Université Claude Bernard Lyon-1, 69372 Lyon, France

Received for publication, September 1, 2000, and in revised form, November 29, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Fatty acids have been postulated to regulate uncoupling protein (UCP) gene expression in skeletal muscle in vivo. We have identified, at least in part, the mechanism by which polyunsaturated fatty acids increase UCP-2 expression in primary culture of human muscle cells. -6 fatty acids and arachidonic acid induced a 3-fold rise in UCP-2 mRNA levels possibly through transcriptional activation. This effect was prevented by indomethacin and mimicked by prostaglandin (PG) E₂ and carbaprostacyclin PGI₂, consistent with a cyclooxygenase-mediated process. Incubation of myotubes for 6 h with 100 µM arachidonic acid resulted in a 150-fold increase in PGE₂ and a 15-fold increase in PGI₂ in the culture medium. Consistent with a role of cAMP and protein kinase A, both prostaglandins induced a marked accumulation of cAMP in human myotubes, and forskolin reproduced the effect of arachidonic acid on UCP-2 mRNA expression. Inhibition of protein kinase A with H-89 suppressed the effect of PGE₂, whereas cPGI₂ and arachidonic acid were still able to increase ucp-2 gene expression, suggesting additional mechanisms. We found, however, that the MAP kinase pathway was not involved. Prostaglandins, particularly PGI₂, are potent activators of the peroxisome proliferator-activated receptors. A specific agonist of peroxisome proliferator-activated receptor (PPAR)  (L165041) increased UCP-2 mRNA levels in myotubes, whereas activation of PPAR or PPAR was ineffective. These results suggest thus that ucp-2 gene expression is regulated by -6 fatty acids in human muscle cells through mechanisms involving at least protein kinase A and the nuclear receptor PPAR.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Uncoupling protein-2 (UCP-2)¹ and uncoupling protein-3 (UCP-3) are members of the mitochondrial carrier family that show sequence identity with UCP-1 (1-6), the brown adipose tissue-specific uncoupling protein involved in the control of cold-induced nonshivering thermogenesis (7, 8). UCP-1 mediates the dissipation of the mitochondrial proton gradient generated by the respiratory chain resulting in the production of heat instead of ATP (7, 8). Because there is little brown fat in adult, UCP-1 activity may not contribute to a large extent to energy expenditure in humans. UCP-2 mRNA is present in many tissues including adipose tissue and skeletal muscle (1, 2), whereas UCP-3 mRNA is preferentially expressed in skeletal muscle (3, 4). Biochemical studies have demonstrated that UCP-2 and UCP-3, like UCP-1, are located in the inner mitochondrial membrane and have uncoupling activity (5, 6). Therefore, they are candidates to explain the mitochondrial proton leak in tissues devoid of UCP-1. To date, the proposed functions for UCP-2 and UCP-3 include the control of different components of energy expenditure, the control of reactive oxygen species (ROS) production generated by the mitochondrial electron transport chain, the regulation of ATP synthesis, and the regulation of fatty acid oxidation (5, 6).

Several lines of evidence implicate fatty acids in the regulation of ucp-2 and ucp-3 gene expression. For example, high fat diet for 2 weeks produces a marked increase in UCP-2 mRNA expression in white adipose tissue in mice (9). The induction of ucp-3 gene expression in mouse muscle just after birth correlates with the increase in plasma levels of NEFAs due to the initiation of suckling (10). In rats, an increase in plasma NEFA level induced by intralipid plus heparin infusion causes a rise in skeletal muscle UCP-3 mRNA level (11). In humans also, data support a role for fatty acids in the regulation of ucp-2 and ucp-3 gene expression in vivo. A positive correlation was reported between plasma NEFA levels and total UCP-3 mRNA levels in skeletal muscle of obese subjects (12). We have previously shown that UCP-2 and UCP-3 mRNA expression is up-regulated during severe calorie restriction (13, 14), a condition associated with increased adipose tissue lipolysis and plasma nonesterified fatty acid (NEFA) concentrations. In addition, we have observed positive relationships between variations in plasma NEFA levels during calorie restriction and changes in UCP-3 mRNA abundance in the muscle of obese nondiabetic and type 2 diabetic patients (15). Finally, we have demonstrated that the rise in NEFA plasma levels and in lipid oxidation during triglyceride infusion is associated with increased expression of UCP-3 in skeletal muscle of healthy lean subjects (16).

Little data are available regarding the nature of the fatty acids involved in the regulation of UCP genes expression in muscle and in adipose tissue, and thus their mechanisms of action are not yet clearly defined. So far, in vitro studies have mainly tested the role of the peroxisome proliferator-activated receptors (PPARs) because these nuclear receptors have been proposed to mediate the transcriptional effect of polyunsaturated fatty acids (PUFAs) or their derivatives on a number of lipid-related genes (17, 18). In white adipocytes, PPAR seems to be involved in the regulation of ucp-2 gene expression in undifferentiated cells (19), whereas PPAR is likely to play a major role in differentiated adipocytes (19 -21). In muscle, depending upon the cell line used, the data are less clear. Oleic acid was found to induce ucp-3 gene expression in several models (22, 23). This effect can be reproduced by a specific agonist of PPAR in mouse C2C12 cells (22). In rat L6 myotubes, however, PPAR activation does not modify ucp-3 gene expression, and PPAR was postulated to play a role in this process (23). Nevertheless, in the L6 cell model, UCP-2 expression was reported to be induced by thiazolidine diones and PPAR activation (21). Finally, in cardiomyocytes, activation of PPAR, and not of PPAR, promoted a strong induction of UCP-2 expression (24).

Additional pathways, distinct from PPAR activation, participate in the regulation of gene expression by fatty acids (25). Fatty acids or their derivatives (acyl-coenzyme A, oxidized fatty acids) may directly bind to and modify the transcriptional activity of other transcription factors (25). In addition, eicosanoids can bind cell surface receptors that are linked to G proteins and then activate several signaling cascades (MAP kinase, protein kinase A, protein kinase C) (25). The concomitant participation of multiple pathways in the transcriptional action of fatty acids has been recently illustrated in a study of the regulation of Glut 4 gene expression by arachidonic acid (26).

Therefore, in the present work we attempted to identify the nature and the mechanism of action of the fatty acids that may be involved in the previously observed induction of UCP expression in human skeletal muscle in vivo (14, 15, 16). To this end, we used a model of human differentiated myotubes in primary culture (27). We found that -6 PUFAs, but not the -3, up-regulate ucp-2 gene expression in human muscle cells through the production of prostaglandins and multiple intracellular pathways including protein kinase A-dependent activation and involvement of the nuclear receptor PPAR.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Materials-- Culture media were from Life Technologies, Inc., or from Biomedia (Boussens, France). Fetal calf serum was purchased from Biomedia. All fatty acids and prostaglandins were obtained from Sigma and Biomol Research Laboratories (Plymouth Meeting, PA). Indomethacin, nordihydroguaiaretic acid (NDGA), and forskolin were purchased from Sigma. H-89 (protein kinase A inhibitor) and PD-98059 (MAP kinase inhibitor) were from Calbiochem. Wy-14643 was obtained from Biomol Research Laboratories. Rosiglitazone was kindly provided by SmithKline Beecham (Harlow, UK). L-165041 was a gift from Drs. J. Berger and D. Moller from Merck.

Primary Culture of Human Skeletal Muscle Cells-- Biopsies of the lumbar mass (erector spinae) muscle (about 1 g) were taken, with the consent of the patient, during surgical procedure. The Ethics Committee of Lyon Hospitals approved the experimental protocol. In the present study, biopsies were taken from healthy lean subjects (49 ± 5 years) with no familial or personal history of diabetes, dyslipidemia, or hypertension. The satellite cells were isolated from the muscle biopsy by trypsin digestion and were grown in Ham's F-10 medium supplemented with 20% fetal calf serum, 1% chicken embryo extract, and 1% antibiotics (100 units/ml penicillin and 100 µg/ml streptomycin) as described previously in detail (27). Confluent myoblast were allowed to differentiate into myotubes in -minimal essential medium containing 2% fetal calf serum and 1% antibiotics. Twelve to 16 days after induction of the differentiation process, most of the cells showed a multinuclear status that characterized mature myotubes (27). In addition, the differentiated cells expressed muscle-specific markers (myosin, sarcomeric -actin, and creatine kinase), and they were characterized by an insulin-sensitive glucose transport (27). The human myotubes were serum-starved overnight before incubations with fatty acids or other products.

Cell Treatments-- Stock solutions (50 mM) of fatty acid were prepared in absolute ethanol (EtOH) and stored at 20 °C in glass tubes, away from light. To obtain the desired concentrations, small amounts of stock solution were added, while stirring gently, to pre-warmed (37 °C) -minimal essential medium containing 1% antibiotics and 1% free fatty acid bovine serum albumin (Roche Molecular Biochemicals). Dilutions were held at 37 °C for at least 1 h before addition to the cells. The molar ratio of fatty acid to albumin was 0.8, and the final concentration of EtOH was always kept below 0.2%. Solutions of prostaglandins, indomethacin, and NDGA were prepared in EtOH and kept at 20 °C under the same conditions as for the fatty acids. Proper amounts were directly added to the cell medium. Concentrated (10² M) stock solutions of the different PPAR agonists, PD-98059 and forskolin, were done in dimethyl sulfoxide (Me₂SO) and stored at 20 °C away from the light. The final concentration was made by dilution with culture media. The final concentration of Me₂SO was always kept less than 0.1%. All experiments with differentiated myotubes were done in 6-well culture plates, and incubations with vehicle (EtOH or Me₂SO) alone were systematically performed as controls.

Quantification of Target mRNAs-- At the end of the incubation periods, cells were scraped in the presence of 350 µl of the lysis buffer from the RNeasy kit for total RNA preparation (Qiagen, Courtaboeuf, France). Total RNA was purified following the instructions of the manufacturer, resuspended in 40 µl of RNase-free water, and stored at 80 °C until quantification of mRNAs. Concentration and purity of each sample were assessed by absorbance measurement at 260 nm and by the 260:280 nm ratio, respectively. The yield of total RNA averaged 1.3 µg/well from a 6-well culture plate, corresponding to ~250,000 myoblasts per well at confluence, before induction of the differentiation process.

The mRNA levels UCP-2, UCP-3, and the different isoforms of PPARs were determined using the RT-competitive PCR (RT-cPCR) assays that were described previously and validated (14, 28). The RT reactions were performed with 0.2 µg of total RNA in the presence of a specific antisense primer and a thermostable reverse transcriptase enzyme to warrant optimal synthesis of first strand cDNA (29). Cy-5 5'-end-labeled sense primers were used during the cPCR to generate fluorescent PCR products that were analyzed using an automated laser fluorescence DNA sequencer (ALFexpress, Amersham Pharmacia Biotech) in 4% denaturating polyacrylamide gels. The sequences of the primers were identical to those reported previously (14, 28). The initial concentration of the target mRNA was determined at the competition equivalence point, as described in detail (29).

Determination of Prostaglandin Production-- Myotubes were incubated with 100 µM arachidonic acid under the conditions described above. Incubation medium (3 ml) was collected from the well and filtrated to remove cells and cellular debris using 0.45-µm Minisart filter units (Sartorius AG, Goettingen, Germany). The clarified medium was kept at 4 °C in the presence of 40 µl of butylated hydroxytoluene as antioxidant. Extraction of the lipid fraction was performed with chloroform/methanol (2:1 v/v) (4.5 ml at pH 3). After centrifugation, the organic phase was removed and dried under nitrogen. Prostaglandins E₂, F₂, and PGI₂ (measured as its stable metabolite 6-oxo-PGF₁) concentrations were determined by gas chromatography-mass spectrometry using the protocol described by Tsikas (30). Briefly, lipids in the dried organic phase were first derivatized in 100 µl of 10% pentafluorobenzyl bromide and 30 µl of 10% ethyldiisopropylamine (both in acetonitrile) for 30 min at 40 °C. The medium was then dried under nitrogen. Next, the oxo groups of the prostaglandins were methoximated with methoxamine hydrochloride (30 µl) in pyridine (30 µl) for 60 min at 60 °C. After drying under nitrogen, the hydroxy groups were esterified in 100 µl of pure N,O-bis(trimethylsilyl)-trifluoroacetamide for 30 min at 60 °C. Analysis was performed on a Hewlett-Packard 5973 gas chromatography-mass spectrometer (Les Ulis, France) in negative ions chemical ionization using CH₄ as reagent gas.

Determination of Cyclic AMP Production-- The production of cAMP was estimated according to the procedure described by Kasagi et al. (31), by measuring the amount of cAMP released from the cultured myotubes into 1.5 ml of medium (Hanks' solution without NaCl supplemented with 220 mM sucrose and 1 mM of 3-isobutyl-1-methylxanthine). Cyclic AMP was directly quantified in 5 µl of medium using the Biotrack cAMP ¹²⁵I assay system from Amersham Pharmacia Biotech.

Determination of Phosphorylated ERK-1 and ERK-2 by Western Blotting-- After incubation, differentiated myotubes were washed twice with ice-cold phosphate-buffered saline buffer and then lysed in 350 µl of ice-cold lysis buffer (150 mM NaCl, 20mM NaH₂PO₄, 1% Triton X-100, 10 mM EDTA, 20 mM 1% glycerol, 50 mM HEPES, pH 7.4) containing 200 mM NaF and 4 mM NaVO₄. After 1 h on ice, cells were scraped and the lysates were cleared of nuclei and detergent-insoluble materials by centrifugation (14,000 rpm) for 10 min at 4 °C. Proteins (35 µg) were resolved by electrophoresis through 9% SDS-polyacrylamide gel and electrophoretically transferred to polyvinylidine difluoride membrane. Phosphorylated MAP kinase (ERK-1 and ERK-2) were detected using a human anti-phospho-MAP kinase antibody (Upstate Biotechnology Inc.) as described by Sarbassov et al. (32). The phosphorylated MAP kinases were visualized using the Enzyme-catalyzed Fluorescence Substrate for Western blotting from Amersham Pharmacia Biotech and quantified using a FluorImager SI and ImageQuant software (Molecular Dynamics, Sunnyvale, CA).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Expression of UCP-2 and UCP-3 in Human Myotubes and Effects of Fatty Acids-- UCP-2 and UCP-3 mRNA levels was determined by RT-competitive PCR in human skeletal muscle cells after an overnight incubation in the absence of serum in the culture medium. Fig. 1 shows the comparison of the results obtained in myotubes with UCP-2 and UCP-3 mRNA levels measured in human skeletal muscle biopsies. UCP-2 mRNA concentration was approximately two times lower in cultured cells than in muscle biopsies. However, the expression level was readily measurable, and the values were within the lower range of the individual data in muscle biopsies (ranging from 1.6 to 7 amol/µg total RNA) (14). In contrast, UCP-3 mRNA levels were dramatically reduced in cultured myotubes when compared with muscle biopsies (0.05 ± 0.08 versus 12 ± 2 amol/µg total RNA). The measured values were close to the detection limit of the RT-cPCR methodology when using 0.2 µg of total RNA in the reverse transcription reaction (29). Extremely low level of expression of UCP-3 is a classical feature of skeletal muscle cells in culture. Hwang and Lane (22) have indicated that detection of UCP-3 mRNA in mice C2C12 cells required the use of 5 µg of poly(A)⁺ mRNA. Under such conditions, they have obtained a signal, using Northern blot, which was about 10⁴ less important than the glyceraldehyde-3-phosphate dehydrogenase signal (22). In the same study, in contrast, UCP-3 mRNA was readily measurable using total RNA preparations from mice skeletal muscle (22). The reasons for the down-regulation of ucp-3 gene expression in cultured muscle cells are unclear but may be related to the lack of signals arising from the innervation. A recent report demonstrates indeed that muscle denervation in mice led to a marked decrease in UCP-3 mRNA levels in gastrocnemius muscle (33).

View larger version (9K): [in this window] [in a new window]   Fig. 1.   Comparison of UCP-2 and UCP-3 mRNA levels in cultured human myotubes and in human skeletal muscle biopsies. UCP mRNA levels were determined by RT-competitive PCR in primary culture of myotubes from five subjects (closed bars) and in vastus lateralis muscle biopsies from eight subjects (open bars). Data are means ± S.E.

To study the effects of fatty acids on UCP expression, 100 µM of C-18 and C-20 fatty acids that were representative of the different families (saturated, -3, -6 and -9), were added for 24 h to human myotubes, and the changes in UCP-2 and UCP-3 mRNA levels were monitored by RT-cPCR. Regarding UCP-3, oleic acid (C18:1, -9) produced an ~3-fold increase in UCP-3 mRNA levels (data not shown). This result is in agreement with preceding reports (22, 23). However, the amplitude of the effect showed large variations (0.2-18-fold increase, n = 6). This was probably due to the difficulty to measure accurately the low levels of UCP-3 mRNA by RT-cPCR. Therefore, we decided not to investigate the regulation of UCP-3 in this cell model in more detail, and we focused our attention on the effects of fatty acids on ucp-2 gene expression. Because a number of data show coordinate variations in UCP-2 and UCP-3 mRNA levels during metabolic changes in rodents and in humans (6, 14), it could be hypothesized that similar mechanisms are involved in the regulation of both genes by fatty acids.

Fig. 2 shows the effects of various fatty acids on UCP-2 mRNA expression. Stearic acid (C18:0) reduced UCP-2 mRNA levels. In contrast, UCP-2 mRNA levels were significantly increased by -linolenic acid (C18:3, -6), di-homo--linolenic acid (C20:3, -6), and arachidonic acid (C20:4, -6). In rat L6 muscle cells, an induction of UCP-2 mRNA expression by linoleic acid (C18:2, -6) has been observed previously (21). Under our experimental conditions, incubation of human primary muscle cells with linoleic acid did not affect UCP-2 expression. However, all the -6 PUFAs that are directly derived from linoleic acid induced a robust induction of UCP-2 mRNA (Fig. 2). This suggests that either longer incubations with linoleic acid are required to allow its metabolism or that enzymes metabolizing linoleic acid are lacking in the human myotube model. In contrast to the -6 PUFAs, fatty acids of the -3 series, as well as oleic acid (C18:1, -9), did not modify the expression of UCP-2 in human muscle cells in culture. Therefore, these results suggest that the up-regulation of UCP-2 mRNA expression by fatty acids is mainly dependent on the PUFAs from the -6 series. This effect was dependent on the concentration of fatty acid added into the culture medium. The half-maximal effect of -linolenic acid or arachidonic acid was observed with concentrations between 1 and 5 µM (data not shown). To verify whether the induction of UCP-2 mRNA reflected a transcriptional activation of the ucp-2 gene or a post-transcriptional stabilization of the mRNA, myotubes were preincubated for 24 h with or without 100 µM arachidonic acid before addition of 10⁶ M actinomycin D, a potent inhibitor of RNA polymerase II. The kinetics of the decrease in UCP-2 mRNA concentration were similar under the two conditions (data not shown). The results from three independent preparations of human myotubes demonstrated that arachidonic acid does not significantly affect the half-life of UCP-2 mRNA (14 ± 2 versus 11 ± 1 h, without versus with 100 µM arachidonic acid, respectively). This result suggests that -6 PUFAs and arachidonic acid affect the transcriptional regulation of the ucp-2 gene.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effect of fatty acids on UCP-2 mRNA levels. Myotubes were incubated for 24 h with 100 µM of the indicated fatty acids, as described under "Experimental Procedures." UCP-2 mRNA levels were determined by RT-cPCR. Results are the means ± S.E. with cell preparations from at least four different subjects. Effect of -6 PUFAs are indicated in closed bars. *, p  0.01 versus vehicle (ethanol) with six different preparations of myotubes (Student's t test for paired data).

Mechanism of Action of -6 PUFAs on ucp-2 Gene Expression-- When human myotubes were incubated for 6 h instead of 24 h, -linolenic acid (100 µM) had no effect, whereas arachidonic acid significantly increased UCP-2 mRNA levels (Fig. 3, upper panel). Because it is well known that C-18 and C-20 -6-PUFAs can be rapidly transformed into arachidonic acid which is the major metabolically active fatty acid of the -6 series, these results suggested that the regulation of ucp-2 gene expression by -6 PUFAs could be mediated by arachidonic acid. To address the mechanism by which arachidonic acid induces UCP-2 expression, we have investigated its metabolism in the human myotubes. Arachidonic acid is rapidly converted to a number of eicosanoids that have potent physiological properties (34). To identify the involvement of either the lipo- or cyclooxygenase pathway in the generation of active metabolites that can induce ucp-2 gene expression, inhibitors of the two enzyme systems were used (Fig. 3, lower panel). Myotubes were pretreated with nordihydroguaiaretic acid (NDGA), an inhibitor of lipoxygenases or with indomethacin, an inhibitor of cyclooxygenases, before addition of arachidonic acid. The effect of arachidonic acid was completely prevented by indomethacin, whereas treatment with NDGA did not affect the induction of UCP-2 mRNA (Fig. 3). In addition to lipo- and cyclooxygenases, the cytochrome P-450 epoxygenase pathway may eventually be involved in the metabolism of arachidonic acid. It has been demonstrated that NDGA is also an inhibitor of arachidonic acid epoxygenase activity at concentrations higher than 50 µM (35). Since NDGA did not prevent the effect of arachidonic acid on UCP-2 expression under our experimental conditions (Fig. 3), the contribution of the epoxygenase pathway was not likely to be involved. Therefore, these results are consistent with a major role of the cyclooxygenases to convert arachidonic acid to prostanoids which in turn mediate the induction of ucp-2 gene expression. Of note, indomethacin also prevented the effect of -linolenic acid in cells incubated for 24 h (data not shown), indicating that the -6 PUFAs do not act directly on ucp-2 gene expression but rather that their effect is mediated by products of the cyclooxygenase pathway.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Induction of UCP-2 expression is dependent on arachidonic acid metabolism. Upper panel, kinetics of -6 action on PUFAs. Cells were incubated with 100 µM of either -linolenic acid (hatched bars) or arachidonic acid (closed bars) for the indicated times. Data are means ± S.E. for 5 different experiments. *, p < 0.01 versus vehicle (ethanol) (open bars). Lower panel, effect of inhibitors of arachidonic acid metabolism. Myotubes were preincubated for 30 min with 50 µM indomethacin or 200 µM NDGA before addition of 100 µM arachidonic acid (A.A.) for 6 h (closed bars). Results are means ± S.E. with five different cells preparations. *, p < 0.02 in the presence versus in the absence of arachidonic acid (open bars).

To identify which prostanoids were involved in the induction of ucp-2 gene expression, we verified whether the main prostaglandins derived from arachidonic acid elicited the same effect when added directly into the culture medium. Because PGI₂ (prostacyclin) is known to be highly labile, we used its nonmetabolizable analog cPGI₂ (carbaprostacyclin). Fig. 4 (upper panel) shows that both PGE₂ and cPGI₂ induced a robust increase in UCP-2 mRNA levels after 24 h of incubation. In contrast, neither PGD₂ nor PGF₂ affected the mRNA levels of UCP-2 in human myotubes in culture. The magnitude of the effect of PGE₂ was slightly lower, albeit not significantly different, than the effect of arachidonic acid. On the other hand, cPGI₂ produced a significantly more important increase in UCP-2 mRNA levels than arachidonic acid or PGE₂ did (Fig. 4). Cyclooxygenase can also directly convert dihomo--linolenic acid into prostaglandins (34) that might be active on UCP2 expression. In agreement, PGE₁ increased UCP2 mRNA levels to a similar extend as PGE₂ did (2.6 ± 0.9 versus 1.1 ± 0.3 amol/µg total RNA in the presence of 1 µM PGE₁ versus vehicle, n = 3).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Involvement of prostaglandins in the regulation of ucp-2 gene expression. Upper panel, effects of prostaglandins on UCP-2 mRNA levels. Myotubes were incubated for 24 h with either 100 µM arachidonic acid (A.A.) or 1 µM of the indicated prostaglandin. Data are means ± S.E. for six different experiments. *, p < 0.01 versus vehicle (ethanol), and $, p < 0.05 versus incubation with arachidonic acid and with PGE2. Lower panel, production of prostaglandins. Myotubes were incubated with 100 µM arachidonic acid for 6 h, and the amounts of PGE2 and PGI2 were determined in the incubation medium by gas chromatography-mass spectrometry according to the method described under "Experimental Procedures." PGI2 was measured in the form of its stable metabolite 6-oxo-PGF1. The data are presented as arbitrary units corresponding to the area under the curves of the mass spectrometry analysis. Triangles show the data obtained in the presence of 50 µM indomethacin. There was no change in the level of PGE2 or PGI2 when cells were incubated in the absence of arachidonic acid.

To our knowledge, the nature of the major prostaglandins synthesized from arachidonic acid in human skeletal muscle cells is not known. In chicken and rat muscle cells in culture, it has been reported that PGE₂ and PGF₂ are produced in noticeable amounts (36). In addition, several studies have demonstrated that PGI₂ is generally one of the major prostaglandins derived from arachidonic acid in various cell systems (34, 37). In human adipocytes, for example, the production of PGI₂ was found to be about 5-fold more important than the production of PGE₂ (38). We determined, in the culture medium of human myotubes, the productions of the prostaglandins that have been found to induce ucp-2 gene expression. In the absence of arachidonic acid, using a C-19 saturated fatty acid as internal standard to estimate the basal concentrations, we found that the amounts of PGE₂, PGF₂, and PGI₂ (measured as its stable metabolite 6-oxo-PGF₁) were very low (5-10 fmol/µg of protein) and similar for the three prostaglandins. These levels did not change during 6 h of incubation without arachidonic acid (data not shown). In contrast, addition of 100 µM arachidonic acid produced an ~150-fold increase in PGE₂ levels after 6 h of incubation (Fig. 4, lower panel). Under the same conditions, the levels of PGI₂ increased about 15-fold (Fig. 4) although there was no change in the amounts of PGF₂ (data not shown). Moreover, the production of both PGE₂ and PGI₂ was strongly reduced in the presence of 50 µM indomethacin (Fig. 4) which prevented the induction of UCP-2 mRNA expression by arachidonic acid (Fig. 3). It should be also indicated that there was no significant production of PGE₁ in myotubes incubated with 100 µM of -linolenic acid for 6 h (data not shown). Taken together these results strongly suggest that the effect of -6 PUFAs on UCP-2 mRNA expression is probably mediated by arachidonic acid-derived PGE₂ and PGI₂ that are both produced rapidly and in large amounts in human myotubes.

Signal Transduction Pathways Involved in the Effect of Prostaglandins on ucp-2 Gene Expression-- Several high affinity specific receptors for prostaglandins have been described in various cell models (39). In human myotubes, the induction of UCP-2 mRNA expression by PGE₂ and cPGI₂ was dependent upon the concentrations of prostaglandins, with half-maximal effect occurring with concentrations of about 5 nM for PGE₂ and about 10 nM for cPGI₂ (data not shown, n = 3). PGE₂ and PGI₂ are known to stimulate adenylate cyclase through activation of G protein-coupled membrane receptors (39). Fig. 5 (upper panel) shows that both PGE₂ and cPGI₂ (1 µM) produced a time-dependent increase in cAMP levels in human myotubes. In addition, incubation of human myotubes with forskolin, a potent inducer of adenylate cyclase activity, produced a significant increase in UCP-2 mRNA (Fig. 5, lower panel). Carbaprostacyclin (cPGI₂) induced a more important increase in UCP-2 mRNA concentration than PGE₂ did (p < 0.05). This could be related to the more important production of cAMP in the presence of cPGI₂ (Fig. 5, upper panel). All these results suggest thus that the production of cAMP and the activation of the protein kinase A-dependent pathway are likely to be involved in the effect of prostaglandins and arachidonic acid on ucp-2 gene expression. To confirm this hypothesis further, we tested the effect of H-89, a specific inhibitor of protein kinase A (40). Fig. 6 clearly shows that addition of H-89 (10 µM) completely suppressed the effect of PGE₂. In contrast, both cPGI₂ and arachidonic acid were still able to increase significantly UCP-2 mRNA expression in the presence of the inhibitor. It should be noted, however, that the magnitude of the effect of cPGI₂ on UCP-2 mRNA level was reduced in the presence of H-89 (+132 ± 42% versus +241 ± 23% with versus without H-89, n = 4). The effect of arachidonic acid appeared to be less affected by H-89 (+154 ± 20% versus +199 ± 64% with versus without H-89, n = 4). Taken together, these results suggest that elevation of cAMP level and activation of the protein kinase A pathway are involved in the regulation of ucp-2 gene expression by -6 PUFAs and prostaglandins. However, it seemed that additional mechanisms could also play a role.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Effect of forskolin on UCP-2 mRNA expression. Upper panel, production of cAMP in human myotubes. Cells were incubated for the indicated times with 1 µM of either PGE2 (closed symbols) or PGI2 (open symbols) under the conditions described under "Experimental Procedures." Data are means ± S.E. for three different experiments. Cyclic AMP concentration remained at basal level in the absence of prostaglandins (data not shown). Lower panel, effect of forskolin on UCP-2 mRNA expression. Myotubes were incubated for 24 h with 1 µM of forskolin or of the indicated prostaglandin. Data are means ± S.E. for five different experiments. *, p < 0.01 versus vehicle (ethanol), and $, p < 0.05 versus incubation with forskolin and with PGE2.

View larger version (57K): [in this window] [in a new window]   Fig. 6.   Effect of H-89, an inhibitor of protein kinase A, on UCP-2 mRNA expression. Cells were incubated for 24 h in the presence H-89 (10 µM) and the indicated inducers of UCP-2 expression. PGE2 and PGI2 were used at a concentration of 1 µM and arachidonic acid (A.A.) at a concentration of 100 µM. Data are means ± S.E. with four different experiments. *, p < 0.01 versus control condition (ethanol + H-89).

Prostaglandins activate the MAP kinase pathways in various cell models (39). To verify whether a stimulation of the MAP kinases is implicated in the regulation of ucp-2 gene expression, we studied the effect of PD-98059, a classically used MAP kinase kinase inhibitor. We found that addition of PD-98059 (50 µM) to human myotubes did not prevent the induction of UCP-2 mRNA by arachidonic acid, PGE₂, or cPGI₂ (data not shown). In addition, whereas PGF₂, which is known to stimulate the MAP kinase pathway in other cell models (39), increased the phosphorylation of ERK-1 and ERK-2 (p44 and p42 MAP kinases), we found that these kinases were not modified by either PGE₂ or cPGI₂ (data not shown), indicating that these prostaglandins did not activate the MAP kinase pathway in human myotubes. Because PGF₂ did not affect UCP-2 mRNA levels (Fig. 3), these results clearly demonstrate that activation of the MAP kinase pathway is not involved in the regulation of ucp-2 gene expression by -6 PUFAs in human myotubes.

Involvement of Peroxisome Proliferator-activated Receptors-- Prostanoids are also potent activators of the nuclear receptors of the PPAR family (41, 42). cPGI₂ has been demonstrated to bind and activate the isoforms  and  of the PPARs, whereas PGE₂ is a less potent agonist (41). We have previously reported the expression profile of the three PPAR mRNAs in human skeletal muscle (28). Muscles are characterized by extremely low mRNA concentrations of PPAR, whereas PPAR and PPAR mRNAs are expressed at a higher level. In cultured human myotubes, we found that the average mRNA levels of the three PPARs were grossly similar to what was previously determined in muscle biopsies (28). We found low levels of PPAR mRNA (0.2 ± 0.1 amol/µg total RNA, n = 5). PPAR mRNA (2.1 ± 0.3 amol/µg total RNA, n = 5, p < 0.001) was about 10-fold more abundant than PPAR. There were intermediate levels of PPAR mRNA (0.7 ± 0.1 amol/µg total RNA, n = 5). By using selective agonists of the different PPAR isoforms, we then tested their possible involvement in the effect of arachidonic acid on ucp-2 gene expression. Fig. 7 shows that WY-14643 (PPAR agonist) and Rosiglitazone (PPAR agonist) did not alter UCP-2 mRNA levels in human myotubes. 15-Deoxy-^12,14-PGJ₂, the proposed natural ligand of PPAR (43), also did not affect UCP-2 mRNA expression (data not shown), further confirming that PPAR is not involved in the regulation of ucp-2 gene in human myotubes. In contrast, incubation with L-165041, a recently developed agonist of PPAR (44), produced a robust induction of UCP-2 mRNA levels (Fig. 10). Because PGI₂ is a potent activator of PPAR (41, 42) and because we found that only the activation of this nuclear receptor resulted in an activation of ucp-2 gene expression, it is thus likely that part of the action of PGI₂ can be mediated by a direct activation of PPAR, in addition to the stimulation of the protein kinase A pathway.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   Effect of PPAR agonists on UCP-2 expression. Myotubes were incubated for 24 h with 1 µM of the indicated PPAR agonists. Wy-14643 is a PPAR agonist, Rosiglitazone a PPAR agonist, and L-165041 a PPAR agonist. Results are means ± S.E. with six different preparations of myotubes. *, p < 0.01 versus vehicle (DMSO).

A role of PPAR in the regulation of ucp-2 gene expression has already been postulated in preadipocyte cells (19). In contrast, in fully differentiated adipocytes it has been clearly demonstrated that activation of PPAR can induce ucp-2 gene expression (19, 20, 45). PPAR is expressed at very high levels in mature adipocytes but is virtually absent in preadipocytes, whereas PPAR is constitutively expressed (46). This suggests that the regulation of ucp-2 gene may implicate either PPAR or PPAR, depending on the relative abundance of the two nuclear receptors. In human skeletal muscle (28) and in cultured muscle cells (Ref. 23 and this work), the expression pattern of PPAR and PPAR is rather similar to what has been reported in preadipocytes. PPAR mRNA levels are extremely low, and PPAR is the major isoform of the PPARs. Under such conditions, the regulation of ucp-2 gene expression by PPAR agonists, and probably also by fatty acid derivatives, implicates PPAR. Recently, a similar conclusion has been proposed, but not tested directly, for the induction of UCP-3 expression by fatty acids in rat L6 myotubes (23).

UCP-2 and PPAR are expressed ubiquitously (6, 47). The highest levels of UCP-2 mRNA have been found in intestine, lung, spleen, and monocytes/macrophages (6, 5), tissues that are also characterized by high expression levels of PPAR (47, 48) and by an important metabolism of prostanoids. Therefore, the observed up-regulation of ucp-2 gene expression by -6 PUFAs and PGI₂ may occur also in other cell types than in myotubes and in preadipocytes. To date, the exact role of UCP-2 is not clearly defined (5, 6), but increasing body of evidence suggests a participation of UCP-2 in the regulation of the production of reactive oxygen species (ROS) (6, 49). An excess of ROS can produce cell damage, and it has been demonstrated that respiration uncoupling limits ROS synthesis (6). It might thus be of interest to verify whether the regulatory mechanism of ucp-2 gene expression described in the present work participates in an antioxidant defense process of the cells (6, 49).

As for UCP-2, the functions of PPAR are poorly known, in contrast to the well established functions of PPAR in adipocyte differentiation and of PPAR in the -oxidation of fatty acids and other aspects of lipid metabolism in the liver (17, 18). It has been proposed that PPAR plays a role in the initiation of the adipocyte differentiation process and in the effect of long chain fatty acids on post-confluent preadipocyte proliferation (46, 50, 51). However, PPAR is expressed in almost all tissues or cells both in rodents (47) and in humans (28, 48), suggesting additional functions. The availability of a specific agonist of PPAR (44) will probably help to rapidly elucidate the role of this nuclear receptor. To date, only few studies have investigated the consequences of an activation PPAR in vivo in rodents. Treatment for 2 weeks with L-165041 did not affect plasma glucose and triglyceride concentrations in the insulin resistant db/db mice (44, 52), in contrast to what has been observed during treatments with a PPAR agonist (44). On the other hand, mice treated with L-165041 showed increased plasma cholesterol concentrations and elevation of circulating high density lipoprotein levels, suggesting that PPAR may be involved in the regulation of cholesterol metabolism in vivo (52). The target genes of PPAR involved in this regulation have yet to be identified. At present, only a limited number of target genes of the nuclear receptor PPAR have been described. Activation of PPAR by fatty acids has been shown to induce the transcription of the genes encoding fatty acid transporter and adipocyte lipid-binding protein both in preadipocytes and in fibroblasts overexpressing PPAR (46, 50). Our work suggests that UCP-2 may be an additional target gene of PPAR. Consensus sequences for peroxisome proliferator-responsive elements have been identified in the 3.3-kilobase pair sequence of the promoter region of the human UCP-2 gene recently published by Tu et al. (53). However, the confirmation that the UCP-2 gene is a genuine target gene of PPAR will now require the direct study of the regulation of the promoter.

In summary, our study demonstrates that -6 PUFAs up-regulate ucp-2 gene expression in human skeletal muscle cells in primary culture. We identified two potential mechanisms, both requiring the oxidative metabolism of arachidonic acid by cyclooxygenases, as follows: 1) the increase of intracellular cAMP levels and the activation of the protein kinase A pathway in response to the production of PGE₂ and PGI₂, and 2) the direct activation of the nuclear receptor PPAR, probably by PGI₂. In addition, we demonstrate that the MAP kinase pathway is not involved in the regulation of ucp-2 gene expression by -6 PUFAs. This work thus provides potential mechanisms for the regulation of ucp-2 gene expression by fatty acids, a process initially proposed on the basis of observations made in human skeletal muscle in vivo (13). In addition, the involvement of PPAR in the regulation of ucp-2 gene expression suggests a possible role of this nuclear receptor in the control of mitochondrial uncoupling activity and its related consequences.

	ACKNOWLEDGEMENTS

We are grateful to M. Odeon and A. Stefanutti for technical assistance with the assay of prostaglandins and cAMP and to K. Bouzakri for assistance in setting up the Western blots of the phosphorylated MAP kinases. We thank Prof. M. Laville, Dr. D. Langin, and Dr. P.A. Grimaldi for discussions.

	FOOTNOTES

^* This work was supported in part by grants from ALFEDIAM-Novo Nordisc, from Institut de Recherche Servier, and from INSERM Progres 4P020D.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a Ph.D. grant from the French Ministère de la Recherche et de l'Enseignement Supérieur. To whom correspondence should be addressed: INSERM U449, Faculté de Médecine René Laennec, Rue G. Paradin, F-69372 Lyon Cedex 08, France. Tel.: 33 478 77 86 29; Fax: 33 478 77 87 62; E-mail: vidal@laennec.univ-lyon1.fr.

Published, JBC Papers in Press, January 12, 2001, DOI 10.1074/jbc.M008010200

	ABBREVIATIONS

The abbreviations used are: UCP, uncoupling protein; PUFAs, polyunsaturated fatty acids; NEFAs, nonesterified fatty acids; PPAR, peroxisome proliferator-activated receptor; NDGA, nordihydroguaiaretic acid; EtOH, ethanol; PG, prostaglandin; ROS, reactive oxygen species; MAP, mitogen-activated protein; PCR, polymerase chain reaction; cPCR, competitive PCR; RT, reverse transcriptase; ERK, extracellular signal-regulated kinase; cPGI₂, carbaprostacyclin PGI2.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. Bugge, L. Grontved, M. M. Aagaard, R. Borup, and S. Mandrup The PPAR{gamma}2 A/B-Domain Plays a Gene-Specific Role in Transactivation and Cofactor Recruitment Mol. Endocrinol., June 1, 2009; 23(6): 794 - 808. [Abstract] [Full Text] [PDF]

				E. Fosslien Cancer Morphogenesis: Role of Mitochondrial Failure Ann. Clin. Lab. Sci., January 1, 2008; 38(4): 307 - 330. [Abstract] [Full Text] [PDF]

				P. Garcia-Roves, J. M. Huss, D.-H. Han, C. R. Hancock, E. Iglesias-Gutierrez, M. Chen, and J. O. Holloszy Raising plasma fatty acid concentration induces increased biogenesis of mitochondria in skeletal muscle PNAS, June 19, 2007; 104(25): 10709 - 10713. [Abstract] [Full Text] [PDF]

				R. Nielsen, L. Grontved, H. G. Stunnenberg, and S. Mandrup Peroxisome Proliferator-Activated Receptor Subtype- and Cell-Type-Specific Activation of Genomic Target Genes upon Adenoviral Transgene Delivery Mol. Cell. Biol., August 1, 2006; 26(15): 5698 - 5714. [Abstract] [Full Text] [PDF]

				B. Desvergne, L. Michalik, and W. Wahli Transcriptional Regulation of Metabolism Physiol Rev, April 1, 2006; 86(2): 465 - 514. [Abstract] [Full Text] [PDF]

				F. Djouadi, F. Aubey, D. Schlemmer, and J. Bastin Peroxisome Proliferator Activated Receptor {delta} (PPAR{delta}) Agonist But Not PPAR{alpha} Corrects Carnitine Palmitoyl Transferase 2 Deficiency in Human Muscle Cells J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1791 - 1797. [Abstract] [Full Text] [PDF]

				H. M. Milne, C. J. Burns, P. E. Squires, N. D. Evans, J. Pickup, P. M. Jones, and S. J. Persaud Uncoupling of Nutrient Metabolism From Insulin Secretion by Overexpression of Cytosolic Phospholipase A2 Diabetes, January 1, 2005; 54(1): 116 - 124. [Abstract] [Full Text] [PDF]

				Y. Gosmain, E. Lefai, S. Ryser, M. Roques, and H. Vidal Sterol Regulatory Element-Binding Protein-1 Mediates the Effect of Insulin on Hexokinase II Gene Expression in Human Muscle Cells Diabetes, February 1, 2004; 53(2): 321 - 329. [Abstract] [Full Text]

				P. M. Jones, C. J. Burns, V. D. Belin, H. M. Roderigo-Milne, and S. J. Persaud The Role of Cytosolic Phospholipase A2 in Insulin Secretion Diabetes, February 1, 2004; 53(90001): S172 - 178. [Abstract] [Full Text]

				K. Bouzakri, M. Roques, P. Gual, S. Espinosa, F. Guebre-Egziabher, J.-P. Riou, M. Laville, Y. Le Marchand-Brustel, J.-F. Tanti, and H. Vidal Reduced Activation of Phosphatidylinositol-3 Kinase and Increased Serine 636 Phosphorylation of Insulin Receptor Substrate-1 in Primary Culture of Skeletal Muscle Cells From Patients With Type 2 Diabetes Diabetes, June 1, 2003; 52(6): 1319 - 1325. [Abstract] [Full Text] [PDF]

				M. Zackova, E. Skobisova, E. Urbankova, and P. Jezek Activating {omega}-6 Polyunsaturated Fatty Acids and Inhibitory Purine Nucleotides Are High Affinity Ligands for Novel Mitochondrial Uncoupling Proteins UCP2 and UCP3 J. Biol. Chem., May 30, 2003; 278(23): 20761 - 20769. [Abstract] [Full Text] [PDF]

				A. J. Gilde, K. A.J.M. van der Lee, P. H.M. Willemsen, G. Chinetti, F. R. van der Leij, G. J. van der Vusse, B. Staels, and M. van Bilsen Peroxisome Proliferator-Activated Receptor (PPAR) {alpha} and PPAR{beta}/{delta}, but not PPAR{gamma}, Modulate the Expression of Genes Involved in Cardiac Lipid Metabolism Circ. Res., March 21, 2003; 92(5): 518 - 524. [Abstract] [Full Text] [PDF]

				D. Cameron-Smith, L. M Burke, D. J Angus, R. J Tunstall, G. R Cox, A. Bonen, J. A Hawley, and M. Hargreaves A short-term, high-fat diet up-regulates lipid metabolism and gene expression in human skeletal muscle Am. J. Clinical Nutrition, February 1, 2003; 77(2): 313 - 318. [Abstract] [Full Text] [PDF]

				D. M. Muoio, P. S. MacLean, D. B. Lang, S. Li, J. A. Houmard, J. M. Way, D. A. Winegar, J. C. Corton, G. L. Dohm, and W. E. Kraus Fatty Acid Homeostasis and Induction of Lipid Regulatory Genes in Skeletal Muscles of Peroxisome Proliferator-activated Receptor (PPAR) alpha Knock-out Mice. EVIDENCE FOR COMPENSATORY REGULATION BY PPARdelta J. Biol. Chem., July 12, 2002; 277(29): 26089 - 26097. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/14/10853    most recent M008010200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chevillotte, E. Articles by Vidal, H. Search for Related Content PubMed PubMed Citation Articles by Chevillotte, E. Articles by Vidal, H. Social Bookmarking                      What's this?