Mitochondrial Reactive Oxygen Species Control the Transcription Factor CHOP-10/GADD153 and Adipocyte Differentiation

Recent reports emphasize the importance of mitochondria in white adipose tissue biology. In addition to their crucial role in energy homeostasis, mitochondria are the main site of reactive oxygen species generation. When moderately produced, they function as physiological signaling molecules. Thus, mitochondrial reactive oxygen species trigger hypoxia-dependent gene expression. Therefore the present study tested the implication of mitochondrial reactive oxygen species in adipocyte differentiation and their putative role in the hypoxia-dependent effect on this differentiation. Pharmacological manipulations of mitochondrial reactive oxygen species generation demonstrate a very strong and negative correlation between changes in mitochondrial reactive oxygen species and adipocyte differentiation of 3T3-F442A preadipocytes. Moreover, mitochondrial reactive oxygen species positively and specifically control expression of the adipogenic repressor CHOP-10/GADD153. Hypoxia (1% O2) strongly increased reactive oxygen species generation, hypoxia-inducible factor-1 and CHOP-10/GADD153 expression, and inhibited adipocyte differentiation. All of these hypoxia-dependent effects were partly prevented by antioxidants. By using hypoxia-inducible factor-1α (HIF-1α)-deficient mouse embryonic fibroblasts, HIF-1α was shown not to be required for hypoxia-mediated CHOP-10/GADD153 induction. Moreover, the comparison of hypoxia and CoCl2 effects on adipocyte differentiation of wild type or HIF-1α deficient mouse embryonic fibroblasts suggests the existence of at least two pathways dependent or not on the presence of HIF-1α. Together, these data demonstrate that mitochondrial reactive oxygen species control CHOP-10/GADD153 expression, are antiadipogenic signaling molecules, and trigger hypoxia-dependent inhibition of adipocyte differentiation. Recent reports emphasize the importance of mitochondria in white adipose tissue biology. In addition to their crucial role in energy homeostasis, mitochondria are the main site of reactive oxygen species generation. When moderately produced, they function as physiological signaling molecules. Thus, mitochondrial reactive oxygen species trigger hypoxia-dependent gene expression. Therefore the present study tested the implication of mitochondrial reactive oxygen species in adipocyte differentiation and their putative role in the hypoxia-dependent effect on this differentiation. Pharmacological manipulations of mitochondrial reactive oxygen species generation demonstrate a very strong and negative correlation between changes in mitochondrial reactive oxygen species and adipocyte differentiation of 3T3-F442A preadipocytes. Moreover, mitochondrial reactive oxygen species positively and specifically control expression of the adipogenic repressor CHOP-10/GADD153. Hypoxia (1% O2) strongly increased reactive oxygen species generation, hypoxia-inducible factor-1 and CHOP-10/GADD153 expression, and inhibited adipocyte differentiation. All of these hypoxia-dependent effects were partly prevented by antioxidants. By using hypoxia-inducible factor-1α (HIF-1α)-deficient mouse embryonic fibroblasts, HIF-1α was shown not to be required for hypoxia-mediated CHOP-10/GADD153 induction. Moreover, the comparison of hypoxia and CoCl2 effects on adipocyte differentiation of wild type or HIF-1α deficient mouse embryonic fibroblasts suggests the existence of at least two pathways dependent or not on the presence of HIF-1α. Together, these data demonstrate that mitochondrial reactive oxygen species control CHOP-10/GADD153 expression, are antiadipogenic signaling molecules, and trigger hypoxia-dependent inhibition of adipocyte differentiation. White adipose tissue is the main energy store in adult mammals and displays great plasticity according to the energy needs of the organism. Adipocyte differentiation results from a subtle balance of sequential and interdependent transcription factors expression that activate or inhibit promoters of adipogenic genes. Three members of the CAAT/enhancer binding protein (C/EBP) 1The abbreviations used are: C/EBP, CAAT/enhancer binding protein; CHOP-10/GADD153, C/EBP homologous protein/growth arrest and DNA damage 153; ROS, reactive oxygen species; HIF-1, hypoxia-inducible factor-1; PDTC, pyrrolidine dithiocarbamate; CCCP, carbonyl cyanide m-chlorophenylhydrazone; MEF, mouse embryonic fibroblast(s); GPDH, glycerol-3-phosphate-dehydrogenase; MnTBAP, manganese (III) tetrakis (4-benzoic acid) porphyrin; H2-DCF, 6-carboxy-2′, 7′-dichlorodihydrofluorescein; H2-DCFDA, 6-carboxy-2 ′, 7′-dichlorodihydrofluorescein diacetate, diacetoxymethylester. family, C/EBPβ, C/EBPδ, and CHOP-10/GADD153 (C/EBP homologous protein also identified as growth arrest and DNA damage 153), are expressed early in the adipocyte differentiation process. By forming heterodimers with the other C/EBPs, CHOP-10/GADD153 acts as a dominant negative regulator of C/EBP (1Ron D. Habener J.F. Genes Dev. 1992; 6: 439-453Crossref PubMed Scopus (986) Google Scholar). During the clonal expansion phase, CHOP-10/GADD153 expression falls and, simultaneously, C/EBPβ and C/EBPδ are transiently induced. This mediates the later expression of C/EBPα and peroxisome proliferation-activated receptor-γ, subsequently triggering full-blown adipocyte differentiation (2Fajas L. Fruchart J.C. Auwerx J. Curr. Opin. Cell Biol. 1998; 10: 165-173Crossref PubMed Scopus (253) Google Scholar). Adipogenic hormones and nutrients are well known to stimulate adipocyte differentiation (3Ailhaud G. Ann. Endocrinol. 1995; 562: 93-95Google Scholar). However, hypoxia strongly inhibits this process (4Yun Z. Maecker H.L. Johnson R.S. Giaccia A.J. Dev. Cell. 2002; 2: 331-341Abstract Full Text Full Text PDF PubMed Scopus (386) Google Scholar). The involvement of mitochondria in adipose tissue plasticity has been neglected until now. Indeed, it was only recently demonstrated that mitochondrial apparatus was a fundamental aspect of white fat and adipocyte differentiation (5Prunet-Marcassus B. Moulin K. Carmona M.C. Villarroya F. Pénicaud L. Casteilla L. FEBS Lett. 1999; 464: 184-188Crossref PubMed Scopus (17) Google Scholar, 6Wilson-Fritch L. Burkart A. Bell G. Mendelson K. Leszyk J. Nicoloro S. Czech M. Corvera S. Mol. Cell. Biol. 2003; 23: 1085-1094Crossref PubMed Scopus (381) Google Scholar). Thus, the concept that mitochondria could play an essential role in white adipose tissue development is now emerging. This is consistent with another study, which reports that genes encoding components of the mitochondrial respiratory chain play an essential role in lipid storage (7McKay R.M. McKay J.P. Avery L. Graff J.M. Dev. Cell. 2003; 4: 131-142Abstract Full Text Full Text PDF PubMed Scopus (249) Google Scholar). Beside their key role in ATP production, mitochondria constitute the primary source of reactive oxygen species (ROS) generation in many cells. Indeed, during respiration, a small but significant proportion of O2 molecules are converted to superoxide anion radical ( O2·¯) by complex I and complex III according to their steady state reduction (8Turrens J.F. Biosci. Rep. 1997; 17: 3-8Crossref PubMed Scopus (755) Google Scholar) (see Fig. 1A). This radical is subsequently diverted into secondary products such as hydrogen peroxide (H2O2) and hydroxyl radical (·OH) via O2·¯ dismutation and Fenton reaction, respectively. The continuous and high generation of mitochondrial ROS has generally been considered in the context of degenerative diseases and aging (9Cadenas E. Davies K.J. Free Radic. Biol. Med. 2000; 29: 222-230Crossref PubMed Scopus (2409) Google Scholar). However, the concept that according to the yield and the time course of mitochondrial ROS generation they also function as physiological signaling molecules is now accepted (10Casteilla L. Rigoulet M. Pénicaud L. IUBMB Life. 2001; 52: 181-188Crossref PubMed Scopus (146) Google Scholar, 11Finkel T. Curr. Opin. Cell Biol. 1998; 10: 248-253Crossref PubMed Scopus (1014) Google Scholar, 12Nemoto S. Takeda K. Yu Z.X. Ferrans V.J. Finkel T. Mol. Cell. Biol. 2000; 20: 7311-7318Crossref PubMed Scopus (336) Google Scholar). For instance, various reports indicate that mitochondrial ROS act as a second messenger in the O2 sensing mechanism to trigger erythropoietin and vascular endothelium growth factor gene transcription via hypoxia-inducible factor-1 (HIF-1) (13Chandel N.S. McClintock D.S. Feliciano C.E. Wood T.M. Melendez J.A. Rodriguez A.M. Schumacker P.T. J. Biol. Chem. 2000; 275: 25130-25138Abstract Full Text Full Text PDF PubMed Scopus (1579) Google Scholar). Previously, we demonstrated that mitochondrial ROS influence the size of the white preadipocyte pool (14Carrière A. Fernandez Y. Rigoulet M. Pénicaud L. Casteilla L. FEBS Lett. 2003; 550: 163-167Crossref PubMed Scopus (96) Google Scholar). This led us to postulate that mitochondrial ROS could also participate in adipocyte differentiation. Using pharmacological strategies, first we confirmed this hypothesis. Second, we focused our investigation on CHOP-10/GADD153 transcription factor as a putative target of mitochondrial ROS. Indeed among the early expressed transcription factors, it is well known to be sensitive to ROS. Finally, we further tested whether hypoxia-dependent inhibition of adipocyte differentiation could be mediated by mitochondrial ROS generation. Chemicals—Mitochondrial inhibitors (antimycin, rotenone, and propofol), pyrrolidine dithiocarbamate (PDTC), carbonyl cyanide m-chlorophenylhydrazone (CCCP), CoCl2, N-acetylcysteine, and actinomycin D were purchased from Sigma. Manganese (III) tetrakis (4-benzoic acid) porphyrin (MnTBAP) was obtained from Calbiochem. Cell Culture Conditions—3T3-F442A preadipocytes were routinely cultivated (14Carrière A. Fernandez Y. Rigoulet M. Pénicaud L. Casteilla L. FEBS Lett. 2003; 550: 163-167Crossref PubMed Scopus (96) Google Scholar) in a 21% O2, 74% N2, and 5% CO2 humidified atmosphere. For the different experiments, cells were seeded at 5500 cells/cm2 and were grown to confluence (day 0). At day 0, medium was changed, mitochondrial inhibitors were added as indicated, and cell cultures were stopped at different times. When cell cultures were maintained more than 2 days after day 0, medium without inhibitors (except when indicated) was replaced. In hypoxia experiments, confluent cells were maintained in hypoxic conditions (1% O2, 94% N2, 5% CO2; work station BugBox, Jouan S.A.) and cultures were stopped at different times. Cells were preincubated with antioxidants at day –1. In some experiments, an uncoupler (CCCP) was added 5 min before the addition of antimycin. Antioxidants and CCCP also were present during the incubation period. Mouse embryonic fibroblast (MEF) cells were induced to differentiate 2 days after confluence with the same medium as for 3T3-F442A (14Carrière A. Fernandez Y. Rigoulet M. Pénicaud L. Casteilla L. FEBS Lett. 2003; 550: 163-167Crossref PubMed Scopus (96) Google Scholar) supplemented with 10 μg/ml insulin, 1 μm dexamethasone, 0.5 mm isobutylmethylxanthine, and 5 μm rosiglitazone. Three days later, medium containing 1 μg/ml insulin and 5 μm rosiglitazone was replaced and then changed every day (4Yun Z. Maecker H.L. Johnson R.S. Giaccia A.J. Dev. Cell. 2002; 2: 331-341Abstract Full Text Full Text PDF PubMed Scopus (386) Google Scholar). To evaluate the hypoxia or CoCl2 effects on adipogenesis, cells were treated 2 days after confluence. Determination of Intracellular ROS Generation—Intracellular ROS generation was assessed using 6-carboxy-2′, 7′-dichlorodihydrofluorescein diacetate, diacetoxymethylester (H2-DCFDA) (Molecular Probes) (15Nègre-Salvayre A. Augé N. Duval C. Robbesyn F. Thiers J.C. Nazzal D. Benoist H. Salvayre R. Methods Enzymol. 2002; 352: 62-71Crossref PubMed Scopus (80) Google Scholar). This probe enters the cells and is hydrolyzed to 6-carboxy-2′, 7′-dichlorodihydrofluorescein (H2-DCF). H2O2 and other peroxides cause oxidation of H2-DCF, yielding the fluorescent product DCF. Confluent cells were loaded with H2-DCFDA (2 μm) 30 min before the end of the incubation period (1 h in presence of the mitochondrial inhibitor). After washing twice with phosphate-buffered saline, cells were collected in water to disrupt cell membranes. Fluorescence (excitation 493 nm, emission 527 nm) and protein content (16Lowry O.H. Rosebrough N.J. Farr A.L. Randall R.J. J. Biol. Chem. 1951; 193: 265-275Abstract Full Text PDF PubMed Google Scholar) then were measured. The intensity of fluorescence was expressed as arbitrary units per milligram of protein. For ROS measurement in hypoxic conditions, cells were incubated with H2-DCFDA for 2 h while the head space was gassed with 1% O2, 94% N2, and 5% CO2 (13Chandel N.S. McClintock D.S. Feliciano C.E. Wood T.M. Melendez J.A. Rodriguez A.M. Schumacker P.T. J. Biol. Chem. 2000; 275: 25130-25138Abstract Full Text Full Text PDF PubMed Scopus (1579) Google Scholar). Cells were immediately treated as described above. Data are normalized to values obtained from normoxic controls. Measurement of Glycerol-3-phosphate-dehydrogenase (GPDH) Activity and Triglyceride Content—GPDH activity was measured as described previously (17Pairault J. Green H. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 5138-5142Crossref PubMed Scopus (223) Google Scholar). Enzyme activity is reflected by the disappearance of NADH, and results were expressed as nmol/min/mg of protein. Cellular triglyceride content was determined using a commercially available test combination (Triglycerides enzymatiques PAP 150, Biomerieux). Western Blot Analysis—Cells were collected on ice in 2 mm EDTA-phosphate-buffered saline. After centrifugation at 600 × g (4 °C, 10 min) pellets were resuspended in lysis buffer (50 mm Hepes, 250 mm NaCl, 0.5% Triton X-100, pH 7), containing 1 mm Na3VO4 and protease inhibitors (10 μg/ml leupeptin, 10 μg/ml aprotinin, and 1 mm phenylmethanesulfonyl fluoride). 40 μg of proteins were separated by 12.5% SDS-PAGE and subjected to Western blot. Antibodies used were: anti-CHOP-10/GADD153, anti-C/EBPβ (both diluted 1:300, Santa Cruz or proteins were using the and Blot from cells was using μg of were on and subjected to product of CHOP-10/GADD153 into were and from to of the phase, as of was by with probe A.M. M. S. G. 17: PubMed Scopus Google Scholar). The were with using the were using were and significant between two were determined by test at Adipocyte a well known of complex III was used at a of nm, which only inhibition data not of cell At this and as induced an in ROS generation with for days inhibited activity and triglyceride content control data not two of adipocyte differentiation. To a putative confluent cells were transiently treated with and activity was later assessed on day Fig. that or h of from confluence was to inhibit h of a effect as the data demonstrate that adipocyte differentiation is inhibited by and that to this during of preadipocytes to is for of h was for further of Pharmacological of ROS on Adipocyte mitochondrial inhibitors in not only ROS generation but also were to demonstrate the effects by ROS. the of effects by the addition of antioxidants was MnTBAP, a superoxide with and PDTC, a and were used to ROS generation of (50 μm) and which effect per on ROS generation and activity not an of fluorescence induced by was prevented when or were added to the medium In in and cells were with with Triglyceride content in and cells was also with with and and data not Second, it has been demonstrated E. A. J. PubMed Scopus Google Scholar, White J. Biol. Chem. 1995; Full Text Full Text PDF PubMed Scopus Google Scholar) that an CCCP, is to stimulate O2·¯ generation induced by we tested the addition of CCCP on effects on ROS generation and adipocyte differentiation. ROS generation in 1 μm cells was strongly and increased with with At this CCCP effect not This was with inhibition of activity when CCCP was added to the medium with with Moreover, triglyceride content was in cells with with not to demonstrate the involvement of mitochondrial ROS in adipocyte we two inhibitors of complex and also has the to PubMed Scopus Google Scholar). for we used of nm) and which only a in cell not Fig. and the effects of and on fluorescence and In these conditions, increased ROS generation and activity and triglyceride content data not the fluorescence and increased activity and triglyceride content data not these demonstrate a very strong and negative correlation between changes in mitochondrial ROS and adipocyte differentiation. of Pharmacological of ROS on inhibition of adipocyte differentiation by mitochondrial ROS early in the differentiation process led us to on CHOP-10/GADD153, a transcription factor that is expressed early and is known to be A.J. H.L. 2001; PubMed Scopus Google Scholar, N.J. J. PubMed Scopus Google Scholar). of confluent cells with for h strongly increased CHOP-10/GADD153 protein The addition of or PDTC, prevented the of CHOP-10/GADD153 Moreover, CHOP-10/GADD153 protein content was when cells were treated with and CCCP with with Finally, CHOP-10/GADD153 was increased after was with Thus, in tested conditions, changes in CHOP-10/GADD153 protein content were positively with mitochondrial ROS generation and with adipocyte differentiation. To test the of these we the expression of C/EBPβ in the same Three of C/EBPβ are present in and proteins known as protein and a protein known as protein. The mitochondrial inhibitors (antimycin, rotenone, and induced a of the protein but these were not with of ROS generation with inhibition of adipocyte differentiation. these results demonstrate that mitochondrial ROS positively and specifically control CHOP-10/GADD153. on Adipocyte and CHOP-10/GADD153 by cells to hypoxia demonstrated a significant of ROS generation which is not activity was inhibited by hypoxia with This was with a strong of CHOP-10/GADD153 protein content addition partly but prevented a hypoxia-dependent of ROS generation as well as inhibition of activity and an in CHOP-10/GADD153 results were obtained with a not Hypoxia induced an in protein and a in protein which was not prevented by addition not these data that mitochondrial ROS generation and CHOP-10/GADD153 could specifically hypoxia-dependent effects on adipocyte differentiation. of CHOP-10/GADD153 a the mechanism of hypoxia-dependent CHOP-10/GADD153 were After hypoxia a very of CHOP-10/GADD153 was of actinomycin D (10 a transcription prevented CHOP-10/GADD153 In these conditions, of was not results demonstrate that hypoxia-dependent of CHOP-10/GADD153 a of CHOP-10/GADD153 HIF-1α was described as hypoxia effects J. Biol. Chem. 2003; Full Text Full Text PDF PubMed Scopus Google Scholar). activity on the between expressed and hypoxic conditions J. Biol. Chem. 2003; Full Text Full Text PDF PubMed Scopus Google Scholar). This led us to putative involvement in hypoxia-dependent inhibition of adipocyte differentiation. In conditions, HIF-1α was not and hypoxia induced the of HIF-1α protein content in preadipocytes The addition of and partly prevented this and data not as it for CHOP-10/GADD153 HIF-1α and CHOP-10/GADD153 our conditions a putative between hypoxia effects on wild type cells and cells deficient for the HIF-1α gene HIF-1α by Western the of the not of CHOP-10/GADD153 was in cell that HIF-1α is not required for hypoxia-mediated CHOP-10/GADD153 induction. CoCl2 is used as a hypoxia for to HIF-1α we the effect of CoCl2 on CHOP-10/GADD153. CoCl2 μm) strongly induced HIF-1α not small in CHOP-10/GADD153 protein content with that obtained after hypoxia was in 3T3-F442A cells with data not Moreover, CoCl2 effect on CHOP-10/GADD153 protein content in cell not results that CoCl2 not the hypoxic conditions for CHOP-10/GADD153 induction. of Hypoxia and CoCl2 on in and CoCl2 μm) inhibited activity of cells the were not by CoCl2 that inhibition of adipocyte differentiation is dependent on HIF-1α expression After of and were inhibited and Fig. with than the results that hypoxia effects on adipocyte differentiation are not by CoCl2 and could be by at least two signaling and the other In this study, we strong that mitochondrial ROS control CHOP-10/GADD153 expression and adipocyte differentiation. Moreover, we that this mechanism could be in hypoxia-dependent inhibition of adipocyte differentiation. mitochondrial inhibitors be used to the of the respiratory chain and ROS generation. their has effects on This led us to different to demonstrate the effects by mitochondrial the putative of of complex effects by the putative of effects by the addition of an CCCP, as previously demonstrated in mitochondria E. A. J. PubMed Scopus Google Scholar) and in cells White J. Biol. Chem. 1995; Full Text Full Text PDF PubMed Scopus Google and comparison between and effects at on This is very ROS by inhibition of complex I which also inhibits complex has the to at this site PubMed Scopus Google Scholar). the changes of mitochondrial ROS generation are and with the of adipocyte differentiation. This a great of adipocyte differentiation to mitochondrial ROS generation, which strongly in the control of this process. to the strong generation of ROS generally we induced only changes in mitochondrial ROS generation by very in mitochondria This is consistent with a role for mitochondrial ROS as physiological signaling molecules that control adipocyte differentiation. Moreover, we demonstrated that mitochondrial ROS at the time of is to later inhibit adipocyte differentiation. This led us to postulate that transcription factors, which play a key role early in adipocyte C/EBP transcription factors be for mitochondrial ROS. C/EBP transcription factor CHOP-10/GADD153 expression has been shown to be by various H2O2 A.J. H.L. 2001; PubMed Scopus Google Scholar, N.J. J. PubMed Scopus Google mitochondrial J. S. N.J. J. 2002; PubMed Scopus Google as well as hypoxia 1992; 52: Google Scholar). results are consistent with these data and demonstrate that in preadipocytes CHOP-10/GADD153 but not other C/EBP expression was strongly to the yield of mitochondrial ROS generation. we also for a between CHOP-10/GADD153 and inhibition of adipocyte differentiation. in such cells their differentiation the adipocyte N. D. J. 1995; PubMed Scopus Google Scholar, D. Habener J.F. Genes Dev. 1992; 6: 439-453Crossref PubMed Scopus (986) Google this strongly suggests that CHOP-10/GADD153 could the mitochondrial inhibition of adipocyte differentiation. ROS are according to the on respiratory chain are as a to the cell to the S. Takeda K. Yu Z.X. Ferrans V.J. Finkel T. Mol. Cell. Biol. 2000; 20: 7311-7318Crossref PubMed Scopus (336) Google Scholar) but also as a of the O2 sensing The between hypoxia and ROS is In a first ROS are by a in proportion to the O2 and a in ROS generation hypoxia-dependent effects J. S. J. PubMed Scopus Google Scholar). In a second hypoxia inhibits mitochondrial changes in the that ROS generation (13Chandel N.S. McClintock D.S. Feliciano C.E. Wood T.M. Melendez J.A. Rodriguez A.M. Schumacker P.T. J. Biol. Chem. 2000; 275: 25130-25138Abstract Full Text Full Text PDF PubMed Scopus (1579) Google Scholar). that measurement of ROS could be to between the two Cell. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar). In our conditions, hypoxia induced an in ROS generation in confluent the second inhibition of adipocyte differentiation was with an of HIF-1α and CHOP-10/GADD153 protein All these hypoxia effects were prevented by the addition of antioxidants that mitochondrial ROS be at the of hypoxia signaling pathways in adipose cells. HIF-1α and CHOP-10/GADD153 and the involvement of a mechanism on CHOP-10/GADD153 hypoxia a putative between However, the of mouse embryonic fibroblasts demonstrated that HIF-1α is not required for hypoxia-mediated CHOP-10/GADD153 induction. This with a study that identified a of genes by hypoxia in an Y. J.M. D. K. M. M. F. R. L. D. E. E. 1998; PubMed Scopus Google Scholar). However, the involvement of HIF-1α in hypoxia-dependent inhibition of adipocyte differentiation was (4Yun Z. Maecker H.L. Johnson R.S. Giaccia A.J. Dev. Cell. 2002; 2: 331-341Abstract Full Text Full Text PDF PubMed Scopus (386) Google Scholar). The of the was using a hypoxia with their Indeed, inhibition of adipocyte differentiation is dependent on the presence of HIF-1α. In such conditions, it is that CoCl2 not strongly CHOP-10/GADD153 expression as hypoxia as described on cells K. C.E. M. H. T. A.L. PubMed Scopus Google Scholar). hypoxia-mediated inhibition of adipocyte differentiation is only partly dependent on the presence of HIF-1α activity in is inhibited by hypoxia in a than in the this suggests the involvement of at least two pathways in the hypoxia-mediated inhibition of adipocyte dependent or of HIF-1α. This signaling be by CHOP-10/GADD153. This study that mitochondrial ROS be considered as antiadipogenic signaling molecules. also that mitochondrial ROS and CHOP-10/GADD153 to the O2 sensing mechanism of adipose cells. This signaling could be in growth of adipose tissue as described in other Y. 1997; PubMed Scopus Google Scholar, B. 2002; PubMed Scopus Google Scholar). Moreover, ROS generation to mitochondrial could be in adipose with PubMed Scopus Google Scholar). these data emphasize the importance of mitochondria and ROS in white adipose tissue development and C. and C. for work station and and G. and A. for and are also to N. for in the of the