SEARCH

SEARCH BY CITATION

Keywords:

  • Stem cells;
  • Rosiglitazone;
  • Adipocyte;
  • Differentiation;
  • Uncoupling protein one;
  • Brown adipose tissue;
  • White adipose tissue

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

In contrast to the earlier contention, adult humans have been shown recently to possess active brown adipose tissue with a potential of being of metabolic significance. Up to now, brown fat precursor cells have not been available for human studies. We have shown previously that human multipotent adipose-derived stem (hMADS) cells exhibit a normal karyotype and high self-renewal ability; they are known to differentiate into cells that exhibit the key properties of human white adipocytes, that is, uncoupling protein two expression, insulin-stimulated glucose uptake, lipolysis in response to β-agonists and atrial natriuretic peptide, and release of adiponectin and leptin. Herein, we show that, upon chronic exposure to a specific PPARγ but not to a PPARβ/δ or a PPARα agonist, hMADS cell-derived white adipocytes are able to switch to a brown phenotype by expressing both uncoupling protein one (UCP1) and CIDEA mRNA. This switch is accompanied by an increase in oxygen consumption and uncoupling. The expression of UCP1 protein is associated to stimulation of respiration by β-AR agonists, including β3-AR agonist. Thus, hMADS cells represent an invaluable cell model to screen for drugs stimulating the formation and/or the uncoupling capacity of human brown adipocytes that could help to dissipate excess caloric intake of individuals. STEM CELLS 2009;27:2753–2760


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

White adipose tissue (WAT) plays a central role in the control of energy homeostasis [1, 2]. Alterations in WAT mass by hyperplasia and/or hypertrophy are often leading to dysregulation of metabolic pathways. In contrast to WAT, brown adipose tissue (BAT) is specialized in adaptive thermogenesis in which the uncoupling protein one (UCP1) plays a key role. Importantly, UCP1 ablation induces obesity in mice living at thermoneutrality [3]. BAT has been long known to be present in small animals and in larger newborn mammals including humans [4, 5]. Of utmost interest, recent data show that active BAT is indeed present in healthy adult individuals [6–10] although, in contrast to WAT, which is easily obtained, the availability of human BAT has remained scarce. The emergence of brown adipocytes within WAT following cold exposure or β-agonist treatment has been well-documented in rodents [11–13]. Treatment with specific PPARγ agonists also promotes the appearance of UCP1-expressing cells within WAT depots in rodents and humans [14–16], but the occurrence within white fat depots of pre-existing BAT precursors cannot be excluded. Reciprocally, the apparent conversion of BAT into WAT observed in human newborns and domestic animals cannot rule out the pre-existence of WAT precursors [17–19]. From a biological viewpoint, the existence of a single precursor cell type giving rise to distinct pools of brown versus white preadipocytes remains unclear [20], as brown and white preadipocytes appear “committed” at that stage and only able to differentiate in vitro into brown and white adipocytes, respectively. On one hand, a myogenic signature of brown adipocytes and the existence of a reservoir of brown adipocyte progenitors in human skeletal muscle favor a distinct origin from that of white adipocytes [11, 21, 22]. On the other hand, brown adipocytes could emerge from different origins and other studies have shown that BMP7 triggers commitment of murine mesenchymal progenitor cells to a brown adipocyte lineage in vitro and in vivo [23], and that white adipocytes can be converted to brown adipocytes through transgenesis [24]. Several transcription factors and cofactors have been shown to play a role in the formation of brown fat cells. Both PGC-1α and PGC1-β play essential and complementary roles in differentiation-induced mitochondrial biogenesis and respiration [25–27]. The zinc-finger protein PRDM16 appears to control brown fat determination through induction of PGC-1α, UCP1, and type 2 iodothyronine deiodinase genes [22, 28]. However, in contrast to primary and clonal preadipocytes of WAT from various species, no primary or clonal precursor cells of human brown adipocytes have been so far obtained that could be used as tools to develop therapeutic drugs and to gain further insights into the molecular mechanisms of brown adipogenesis [29]. Recently, we have isolated mesenchymal stem cell populations from human adipose tissue (termed hMADS cells) that exhibit at a clonal level both normal karyotype and high self-renewal ability and that are not tumorigenic; they are able to differentiate into various lineages including adipocytes and osteoblasts as well as to support in vivo regenerative processes [30–33]. Once differentiated into adipocytes, hMADS cells exhibit the molecular and functional properties of human fat cells that include potent release of adiponectin and leptin as well as responsiveness to insulin, β-AR agonists, and, specific to primates, atrial natriuretic peptide [30]. Thus, hMADS cells represent a unique human cell model to examine whether, in response to appropriate stimuli, conversion to functional brown adipocytes can also take place. Our results show that long-term activation of PPARγ is sufficient to promote this event in vitro and to enhance UCP1-dependent respiratory capacity of hMADS cells. They show also that β-AR agonists, including β3-AR agonists, up-regulate UCP1 expression. In light of the occurrence of functional BAT in adult individuals, hMADS cells represent an invaluable cell model to screen for drugs able to stimulate the formation and/or the energy-dissipating capacity of human brown adipocytes.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Cell Culture

The establishment and characterization of multipotency and self renewal capacity of hMADS cells have been described [30–33]. In the experiments reported herein, hMADS-2 cells, established from the pubic region fat pad of a 5-year-old male donor, were used between passages 16 and 35 corresponding to 35-100 population doublings. Cells were seeded at a density of 4,500 cells/cm2 in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS), 2.5 ng/ml of hFGF2, 60 μg/ml of penicillin, and 50 μg/ml of streptomycin. The medium was changed every other day, and hFGF2 was removed when cells reached confluence and were triggered for differentiation at day 2 post-confluence (designated as day 0). Cells were then maintained in DMEM/Ham's F12 media supplemented with 10 μg/ml of transferrin, 0.85 μM insulin, 0.2 nM triiodothyronine, 1 μM dexamethasone (DEX), and 500 μM isobutyl methylxanthine (IBMX). Three days later, the medium was changed (DEX and isobutyl methylxanthine were omitted) and various amounts of rosiglitazone were added for the indicated periods. Media were then changed every other day, and cells were used at the indicated days. Glycerol-3-phosphate dehydrogenase (GPDH) activity measurements and Oil red O staining were performed as described previously [34, 35].

Isolation of Different Cell Fractions from Stroma Vascular Fraction of Human WAT and Culture of CD34+/CD31 Cells

Human subcutaneous WAT was obtained from healthy women undergoing elective procedures of fat removal for aesthetic purposes. The protocols of fat collection were approved by the Institutional Research Board of Inserm and Toulouse University Hospital. The adipose tissue was digested using collagenase (300 U/ml in phosphate buffered saline (PBS), 2% bovine serum albumin, pH 7.4) for 45 minutes under constant shaking. Following removal of the floating mature adipocytes, the lower layer containing the stroma vascular fraction (SVF) was centrifuged (200 g, 10 minutes) and the pellet was resuspended in erythrocyte lysis buffer (155 mM NH4Cl; 5.7 mM K2HPO4; 0.1 mM EDTA, pH 7.3) for 10 minutes. After successive filtrations through 100, 70, and 40 μm sieves, the cells were resuspended in PBS/2% FCS. The distinct cell fractions of the SVF were isolated using an immunoselection/depletion protocol as previously described [36, 37]. Freshly isolated CD34+/CD31- cells defined as progenitor cells, CD34-/CD31+ cells defined as capillary endothelial cells, CD34-/CD14+ cells defined as macrophages, and CD34-/CD14-/CD3+ cells defined as lymphocytes were either lysed for mRNA extraction or cultured for further analysis.

Isolated progenitor cells (120 000 cells/cm2) were plated on 48-well plastic plates in Endothelial Cell Basal Medium (PromoCell, Heidelberg, Germany, http://www.promocell.com) supplemented with 10% FCS. Adipocyte differentiation was induced under the same conditions used for hMADS cells.

Isolation and Analysis of RNA

Extraction of total RNA, reverse transcriptase reactions, and quantitative reverse transcription-polymerase chain reaction assays were performed as described [30, 32, 33, 35]. The expression of selected genes was normalized to that of TATA-binding protein (TBP) gene and quantified using the comparative-ΔCt method. Oligonucleotide sequences, designed using Primer Express software (PerkinElmer Life Sciences, Boston, http://www.perkinelmer.com), are described in supplemental online Table 1.

Western Blot Analysis

Total cellular lysates were subjected to immunoblotting as previously described [35]. Primary antibodies were rabbit anti-human UCP1 and anti-TBP (Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com). Secondary horseradish peroxidase-conjugated antibodies were purchased from Promega (Charbonnières, France, http://www.promega.com). Enhanced ChemiLuminescence (Millipore, St. Quentin en Yvelines, France, http://www.millipore.com) was used for detection.

High-Resolution O2 Consumption Measurement

Oxygen consumption was measured using a 2-injection-chamber respirometer equipped with a Peltier thermostat, Clark-type electrodes, and integrated electromagnetic stirrers (Oroboros Oxygraph, Innsbruck, Austria, http://www.oroboros.at). Measurements were performed at 37°C with continuous stirring in 2 ml of DMEM containing F12 medium supplemented with 10% newborn calf serum. Before each oxygen-consumption measurement, the medium in the chambers was equilibrated with air for 30 minutes, and freshly trypsinized cells were transferred into the respirometer glass chambers. After observing a steady-state respiratory flux, ATP synthase was inhibited with oligomycin (0.25-0.5 mg/l) and cells were titrated with the uncoupler carbonyl cyanide 3-chlorophenylhydrazone up to optimal concentrations in the 1-2 μM range. The respiratory chain was inhibited by antimycin A (1 μg/ml). Oxygen consumption was calculated using DataGraph software (Oroboros software). Basal respiration was defined as the antimycin A-sensitive oxygen consumption. To test the effect of isoproterenol, the β-agonist was acutely added at 1 μM in respiratory chamber and similar experiments as described above were then carried out.

Statistical Analysis

Data are expressed as mean values ± SEM and are analyzed using the 2-tailed Student's t-test. Differences were considered statistically significant at p ≤ .05.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

UCP1 and Brown Fat Cell Markers Are Expressed During Adipocyte Differentiation of hMADS Cells

Activation of PPARγ by rosiglitazone was required for adipocyte differentiation of the whole population of hMADS-2 cells (Fig. 1A) as described previously [30]. Treatment of post-confluent hMADS-2 cells with increasing concentrations of rosiglitazone for 6 days, that is, between days 3 and 9, induced expression of PPARγ and adipsin, a very late marker of adipogenesis, and that of glycerol-3-phosphate dehydrogenase (GPDH) activity associated with lipid accumulation. There was no significant difference in GPDH activity, PPARγ mRNA and adipsin mRNA levels between cells treated with rosiglitazone chronically for 6 days or a longer time period. At day 16, 20 nM rosiglitazone was sufficient to bring a maximal response, in agreement with the affinity of this ligand for PPARγ (Fig. 1A--1D). Taken together, these results indicate that a 6-day exposure to rosiglitazone was sufficient for hMADS-2 cells to become white fat cells. Next, we investigated whether UCP1 mRNA was expressed during this process. Short-term treatment only led to a very low expression of UCP1 mRNA and UCP1 protein. However, a longer chronic treatment of differentiating hMADS-2 cells with 20 nM rosiglitazone led to a strong expression of UCP1 mRNA (10- to 24-fold increase) and UCP1 protein (Fig. 2A, 2B). In contrast to UCP1, the high expression of UCP2 mRNA, which was already observed between days 3 and 9, became modestly enhanced (∼ 2-fold increase) upon a longer exposure to rosiglitazone (Fig. 2C), and UCP2 protein could then be detected (B. Miroux and D. Ricquier, personal communication). These observations suggest that the duration of rosiglitazone treatment regulates the activation of UCP1 gene. Concurrently, CIDEA gene, the expression of which is tightly associated with that of UCP1 in BAT and may be causally linked to obesity by playing a role in energy expenditure [38–40], was also strongly activated (7- to 11-fold increase) upon long-term treatment (Fig. 2E)}. We measured the levels of mRNA encoding the mitochondrial carnitine palmitoyltransferase (CPT1B) and found that they were markedly increased (4- to 8-fold increase) when differentiating hMADS-2 cells switched from the white to the brown phenotype (Fig. 2E), suggesting enhanced mitochondriogenesis ([14] and vide infra Fig. 5A).

thumbnail image

Figure 1. Effect of rosiglitazone on adipogenesis of human multipotent adipose-derived stem (hMADS) cells. hMADS-2 cells were induced to differentiate into adipocytes as described under the Materials and Methods section. Rosiglitazone was added at the indicated concentrations and periods of treatment. At day 16, (A) cells were fixed and stained by Oil red O, GPDH activity (B) was determined and PPARγ (C) and adipsin (D) mRNA levels were measured by quantitative reverse transcription-polymerase chain reaction. Results are mean ± SEM of three independent experiments performed on different series of cells by taking as 100% in (C and D) the values obtained at day 16 with 500 nM rosiglitazone. ∗, p < .01 versus untreated cells. Abbreviations: GPDH, glycerol-3-phosphate dehydrogenase; Rosi, rosiglitazone.

Download figure to PowerPoint

thumbnail image

Figure 2. Effect of long-term exposure to rosiglitazone on the expression of brown adipocyte markers. hMADS-2 cells were induced to differentiate as described in Figure 1. At day 16, UCP1 (A), UCP2 (C), CIDEA (D), and CPT1B (E) mRNA levels were measured by quantitative reverse transcription-polymerase chain reaction. Results are mean ± SEM of three independent experiments performed on different series of cells and are expressed by taking as 100% the values obtained with 500 nM rosiglitazone. UCP1 protein levels (B) were determined on two different series of cells by immunoblotting with TBP used for equal loading. a, p < .05 and b, p < .01 versus untreated cells; c, p < .01 versus short-term (500 nM rosiglitazone) treated cells (days 3-9). Abbreviations: CPT1B, carnitine palmitoyltransferase; Rosi, rosiglitazone; TBP, TATA-binding protein; UCP1, uncoupling protein one; UCP2, uncoupling protein two.

Download figure to PowerPoint

PRDM16, which controls both in vitro and in vivo the determination of mouse brown fat fate by inducing the expression of UCP1 and PGC-1α genes [22, 28], was also expressed in proliferating and confluent hMADS cells (days −3 and 0, data not shown) as well as in cells untreated or in cells short- or long-term treated with rosiglitazone (supplemental online Fig. 1A). In order to assess this point, adipocyte and stromal vascular fractions of subcutaneous WAT from adult patients were first separated by standard centrifugation. Then adipocyte progenitors, endothelial cells, macrophages, and lymphocytes were obtained from the stromal vascular fraction by an immunoselection procedure [36, 37]. Isolated adipocytes and native CD34+/CD31 adipocyte progenitors expressed significant levels of PRDM16, which was also expressed at a high level in macrophages and to a lower level in native endothelial cells and lymphocytes (supplemental online Fig. 1A). Upon differentiation of adipocyte progenitors, PRDM16 remained expressed at levels similar to those observed in hMADS cells (supplemental online Fig. 1A). Importantly, in contrast to PRDM16, UCP1 was not expressed in the various native cell types, although it became expressed during differentiation of adipocyte progenitors at day 16 to one-third the level observed in hMADS cells (supplemental online Fig. 1B). In both adipocyte progenitors and hMADS cells, UCP1 was not expressed upon short-term rosiglitazone treatment. Whether UCP1-expressing cells from this sub-population are functional remains presently unknown.

White adipocytes from rodents are less prone to undergo apoptosis than brown adipocytes [41, 42], but both express anti-apoptotic Bcl-2 and pro-apoptotic Bax protein [43]. Human white adipocytes appear less resistant to apoptosis than their murine counterparts, possibly due to low expression of anti-apoptotic Bcl-2 and NAIP genes [44]. In hMADS-2 cells, the shift from the white to the brown phenotype was unexpectedly accompanied by an increased expression of the anti-apoptotic Bcl-2 gene and a decreased expression of the pro-apoptotic Bax gene, increasing the Bcl-2/Bax ratio from 1-3.7 (supplemental online Fig. 2). This result emphasizes striking differences in adipocytes of different species in the expression pattern of apoptosis-related genes.

As both UCP1 (Fig. 2), β3-AR (Fig. 3A), and β2-AR (not shown) were expressed in hMADS-2 cells exposed between days 3 and 16 to rosiglitazone, responsiveness of brown adipose hMADS cells to β-agonists was assessed. As shown in Figure 3B and 3C, expression of UCP1 mRNA and UCP1 protein was significantly enhanced upon a 6-hour exposure in the 10-100 nM range to isoproterenol, a pan-agonist of β-ARs, as well as to 10-1000 nM CL316243, a selective β3-AR agonist.

thumbnail image

Figure 3. Expression of β3-adrenergic receptor and up-regulation of UCP1 in response to β-AR agonists. (A) hMADS-2 cells were induced to differentiate upon exposure to 100 nM rosiglitazone at the indicated periods of treatment and β3-adrenergic receptor mRNA levels were determined at day 17. (B, C) hMADS-2 cells were induced to differentiate upon exposure to 100 nM rosiglitazone from day 3 to day 16, in the absence or the presence of increasing concentrations of β-AR agonists for the last 6 hours. Results are mean ± SEM of three (A) and two (B) independent experiments performed on different series of cells; they are expressed by taking as 100% the values obtained for the 3-9 days of treatment (A) or in the absence of β-AR agonists (B, C). ∗, p < .05 versus untreated cells. Abbreviations: UCP1, uncoupling protein one.

Download figure to PowerPoint

The Regulation of UCP1 Expression Occurs in hMADS-2 Cells First Differentiated into White Fat Cells

From the above experiments, it remained unclear whether long-term continuous exposure of confluent hMADS-2 cells to rosiglitazone was required to acquire the brown fat cell phenotype or whether a short-term rosiglitazone treatment of white adipocytes was sufficient to trigger the brown fat lineage. Consequently, hMADS-2 cells were first differentiated in the presence of rosiglitazone between days 3 and 9, then rosiglitazone was removed and re-added between days 14 and 16. The results presented in Figure 4A show that a 2-day treatment was sufficient to stimulate the expression of UCP1, CIDEA, and CPT1B genes. This effect was confined to PPARγ activation since the PPARα and PPARδ agonists, Wy14643 and L165041, respectively, were unable to induce the expression of UCP1 protein (Fig. 4B). Furthermore, treatment with known activators/ligands of PPARs in lieu of rosiglitazone (arachidonic acid, docosahexaenoic acid, and eicosapentaenoic acid at 10 μM) had no effect on UCP1 mRNA expression (data not shown). Altogether, these observations demonstrate that a short-term activation of PPARγ was sufficient to trigger a switch of hMADS-2 cells from a white to a brown adipocyte phenotype. Noteworthy, the levels of PPARα, PGC-1α, and PGC-1β, though well expressed during differentiation, were not further raised when hMADS cells acquired the brown phenotype (supplemental online Fig. 3).

thumbnail image

Figure 4. Acquisition of white fat cells of a brown phenotype depends upon PPARγ activation. hMADS-2 cells were first induced to differentiate into white adipocytes in the presence of 100 nM rosiglitazone between days 3 and 9. After removal, rosiglitazone was re-added or not at day 14 at the indicated concentrations for 2 days: (A) UCP1, CIDEA, and CPT1B mRNA levels were determined at day 16 by quantitative reverse transcription-polymerase chain reaction. (B) UCP1 protein expression was determined at day 16 by immunoblotting after treatment by the different PPAR agonists between days 14 and 16 at the indicated concentrations. Results are expressed as fold increase by taking as 1 the values obtained at day 16 after treatment between day 3 and 9 with 100 nM rosiglitazone; they are mean ± SEM of two independent experiments performed on different series of cells. ∗, p < .01 versus short-term (100 nM rosiglitazone) treated cells (days 3-9). Abbreviations: CPT1B, carnitine palmitoyltransferase; Rosi, rosiglitazone; TBP, TATA-binding protein; UCP1, uncoupling protein one.

Download figure to PowerPoint

The effects of rosiglitazone on UCP1 expression are not unique to hMADS-2 cells as similar results were also obtained with hMADS-1 and hMADS-3 cells [31], which had been isolated from the umbilical fat pad of a 31-month-old female donor and the prepubic fat pad of a 4-month-old male donor, respectively (supplemental online Fig. 4 and data not shown).

Oxygen Consumption and Uncoupled Respiration in hMADS-2 Cells

A major feature of brown fat cells versus white fat cells is their higher abundance of mitochondria associated with high rates of cellular and uncoupled respiration [5]. Oxygen consumption was measured in differentiated hMADS-2 cells using an oxygen-sensitive electrode to calculate relative rates of respiration. These experiments revealed a striking effect of long-term rosiglitazone treatment on the levels of both total and oligomycin-sensitive respiration. hMADS-2 cells differentiated in the presence of rosiglitazone for 20 days and expressing brown adipocyte markers displayed a threefold increase in total respiration and a 2.5-fold increase in oligomycin-sensitive respiration relative to cells exposed to rosiglitazone only between days 3 and 9 and expressing white adipocyte markers (Fig. 5A and 5B). Similarly, hMADS cells, first differentiated into white fat cells by treatment with the PPARγ ligand between days 3 and 9 and then exposed to rosiglitazone for an additional 4-day period from day 16, also exhibited a lower but still significant increase in total oxygen consumption and oligomycin-sensitive respiration. In order to definitively demonstrate the UCP1-mediated respiration, we acutely stimulated hMADS cells with isoproterenol in respiratory chamber. As shown in Figure 5C, a potent stimulation occurred in long-term rosiglitazone-treated cells and in short-term rosiglitazone-treated cells, which had been first differentiated into white fat cells. Furthermore, oligomycin-sensitive respiration was significantly higher after isoproterenol stimulation in long-term treated compared to short-term treated cells (ratio of 0.75 ± 0.09 vs 0.35 ± 0.07, p < .05, respectively).

thumbnail image

Figure 5. Effect of short-term and long-term exposure to rosiglitazone on oxygen consumption and uncoupling respiration. Stimulation of basal respiration by isoproterenol. hMADS-2 cells were induced to differentiate in the presence of 100 nM rosiglitazone between days 3 and 9, between days 3 and 20, or between days 3 and 9 and then between days 16 and 20. At day 20, oxygen consumption (A), uncoupled activity (B), and isoproterenol-stimulated respiration (C, i.e., the ratio of rate of oxygen consumption before and after addition of 1 μM isoproterenol) were determined. Results are mean ± SEM of four independent experiments performed on different series of cells. ∗, p < .05 versus short-term (100 nM rosiglitazone) treated cells (days 3-9).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Recently, using fluorodeoxyglucose positron emission associated with computerized tomography to trace tumor metastasis, the presence in healthy adult humans of brown adipose tissue at various sites distinct from white adipose depots has been reported [6]. Human BAT activity is acutely cold-induced and stimulated by the sympathetic nervous system [7–10]. The amount of BAT correlates inversely with body-mass index [8, 45]. These important observations suggest the possibility of increasing BAT metabolic activity to modulate energy expenditure and to control white fat mass, in light also of the fact that UCP1 invalidation triggers obesity at thermoneutrality, a condition that prevails in people of industrialized countries [3]. Therefore, the development of a suitable model for studying formation and energy-dissipating functions of human brown adipocytes appears valuable. hMADS cells exhibit several significant advantages as they display a normal karyotype and a self-renewal ability and differentiation occurs in serum-free chemically-defined medium. Moreover, recent studies have shown striking differences between rodent and human mesenchymal stem cells with respect to Hedgehog signaling in adipocyte differentiation, underlying the importance of using human in lieu of murine cell models [46]. From a biological perspective, the properties of hMADS cells, which undergo differentiation to osteocytes, chondroblasts, myoblast-like cells, white adipocytes [30, 31, 33], and herein brown adipocytes, are raising new issues. Elegant studies have shown that brown adipocytes share a common signature with myoblasts [11, 22] and that a reservoir of brown adipocyte progenitors exists in human skeletal muscle [21]. However, during proliferation and differentiation, hMADS cells did not express Myf5, and their myogenic potential in vitro and in vivo was low unless forced expression of MyoD occurred [47]. Therefore, a common signature with myoblasts can be ruled out. We postulate that the striking plasticity of hMADS cells that allows formation of brown fat cells is related to the fact that they originate from young donors [31] as compared to CD34+/CD31- cells, which originate from adult patients and in which the number of lineages has become more restricted [37]. Once differentiated, hMADS cells fulfill key criteria of brown adipocytes by expressing genes encoding UCP1, CIDEA, PGC-1α, PGC-1β, and PRDM16 as well as key members of the PPAR family. The upregulation of UCP1 expression both by isoproterenol and CL316 243 indicate that the β-AR signaling pathway, in particular through β3-AR, is functional in these cells. So far, the presence and the role of β3-AR in human adipocytes have remained controversial [48]. Although brown adipocytes from young baboons express low levels of β3-AR, no lipolysis was observed in response to four β3-agonists [49]. Human immortalized brown adipocytes expressing β3-AR exhibited a weak lipolytic response when exposed to CGP12177A, a partial β3-agonist, and the receptor appeared poorly coupled to adenylate cyclase [50, 51]. In both cases, no stimulation of UCP1 expression occurred in response to a specific β3-agonist, as we observed herein.

Rosiglitazone belongs to the thiazolidinedione family, a class of insulino-sensitizing compounds, which are effective in the treatment of non-insulino-dependent diabetes [52]. In vitro, UCP1 expression is detected in cultured human preadipocytes upon rosiglitazone treatment [53]. Ligand-activated PPARγ acts on both white and brown precursor cells and promotes their differentiation into white and brown adipocytes, respectively [30, 54-56]. Recently, rosiglitazone has been shown to induce a non-adrenergic recruitment of functional brown adipocytes from mouse BAT preadipocytes [57], and we assume that a similar phenomenon occurs in both hMADS and CD34+/CD31- cells. Regarding signaling, early stimulation of the PKA pathway activation is insufficient to induce fat cell formation. After DEX/IBMX removal, the acquisition of a white and subsequently a brown fat cell phenotype depends only on the duration of PPARγ activation, demonstrating that the same synthetic ligand promotes the formation of both types of fat cells. Importantly, PGC-1α, PGC-1β, and PRDM16 were already expressed at maximal levels in cells exhibiting the white phenotype. Recent studies have shown that PRDM16 controls the determination of brown fat fate in vitro and in vivo, that is, inducing the expression of UCP1 and PGC-1α genes, whereas ligand-activated PPARγ appears necessary for the expression of CIDEA and mitochondrial components [28]. Our results are consistent with these observations and in agreement with the occurrence of functional Peroxisome Proliferator Response Element (PPRE) in the promoter of CIDEA gene [40]. In contrast, expression of PRDM16 was not specific to the brown fat phenotype as it was also expressed in white adipocyte progenitors and macrophages isolated from WAT of adult humans, which might be due to differences between rodent and human cell models. Therefore, besides the expression of PRDM16, PGC-1α, and PGC-1β and the ubiquitous expression of CtBP-1 gene ([58] and data not shown), we conclude from our data that additional molecular event(s), which may occur upon a longer-term exposure to rosiglitazone, are required to acquire a complete brown fat cell phenotype. Studies on the transcriptomic signature of short-term versus long-term rosiglitazone-treated hMADS cells should shed some light on that issue, that is, in deciphering whether a rapid biological switch or a smooth continuum is required for white adipocytes to become brown adipocytes. Further experiments should be helpful in determining whether differentiation into brown adipocytes involves a selected sub-population or the whole cell population.

Rosiglitazone, despite normalizing glycemia and insulinemia, promotes increase of body weight in rodents and most patients [59–62]. In vivo treatment of rodents with this drug also increases BAT mass, which is accompanied by an increase of UCP1 expression [54, 63-65]. In humans, it cannot be ruled out that this drug may also, but in an insufficient manner, increase the amount of BAT now recognized to be present in a fairly large percentage of healthy adult individuals [6]. The contribution of BAT to energy expenditure, that is, nonshivering and diet-induced thermogenesis, is well-established in rodents. In humans, body mass index is inversely correlated with BAT mass. This observation could be due to differences between subjects in their ability to maintain and/or to increase BAT mass via the formation of brown adipocytes. Hopefully, our human cell model should be useful to bring some answers to these important health-related issues.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

In conclusion, we show that human multipotent adipose-derived stem cells of young donors are able to differentiate into functional brown adipocytes. In light of the occurrence of functional BAT in adult individuals, hMADS cells represent an invaluable cell model to screen for drugs able to stimulate the formation and/or the energy-dissipating capacity of human brown adipocytes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

This work was supported by the Centre National de la Recherche Scientifique and by a grant from: “Equipe FRM, soutenue par la Fondation pour la Recherche Médicale”. Work in the K.K. laboratory was supported by the Danish Natural Science Research Council, the NOVO Foundation. Part of the work was carried out as a part of the research program of the Danish Obesity Research Centre (DanORC) supported by The Danish Council for Strategic Research (Grant No. 2101-06-0005). The authors are very grateful to B. Miroux and D. Ricquier for performing Western blot analysis of UCP2. We are grateful for the skilled technical assistance of Nathalie Techer and Mansour Djedaini (CNRS UMR 6543), Mireille André (CNRS UMR 5241), and Pauline Decaunes (INSERM U858), as well as the skillful secretarial assistance of Geneviève Oillaux. C.E. and C.C. were recipients of fellowships from Fondation pour la Recherche Médicale.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional supporting information available online.

FilenameFormatSizeDescription
STEM_200_sm_SuppFig1.tif1946KSupplemental Figure 1. Expression of PRDM16 and UCP-1 in hMADS cells and in native human adipose tissue cells before and after differentiation Mature adipocytes (MA) and stromal-vascular fractions of subcutaneous WAT from adult patients were first separated by standard centrifugation. Then adipocyte progenitors (PROG), endothelial cells (E), macrophages (M) and lymphocytes (L) were obtained from the stromal-vascular fraction by an immunoselection/depletion procedure. PRDM16 (A) and UCP1 (B) mRNA levels were measured by quantitative RT-PCR in native cells and in PROG and hMADS cells induced to differentiate under identical culture conditions. mRNA levels were measured for PROG cells at day 0 and 16 for cells treated with 100 nM rosiglitazone, and at day 16 for hMADS cells treated or not (-) for the indicated time period. Results are mean ± SEM of three independent experiments.
STEM_200_sm_SuppFig2.tif778KSupplemental Figure 2. Expression of apoptosis-related genes upon acquisition of a white fat versus a brown fat phenotype. hMADS-2 cells were induced to differentiate as described in Figure 1 in the presence of 100 nM rosiglitazone. At day 16, Bcl2 and Bax mRNA levels were measured by quantitative RT-PCR. Results are mean ± SEM of three independent experiments performed on different series of cells and are expressed by taking as 100% the values obtained at day 9 after rosiglitazone treatment between days 3 and 9. * p ≤ 0.05.
STEM_200_sm_SuppFig3.tif922KSupplemental Figure 3. Expression of transcription-related factors and co-factors upon acquisition of a white fat versus a brown fat phenotype. hMADS-2 cells were induced to differentiate for different periods of time in the absence or the presence of the indicated concentrations of rosiglitazone. PGC-1α, PGC-1β and PPARα mRNA levels were measured by quantitative RT-PCR at day 16. Results are mean ± SEM of three independent experiments performed on different series of cells by taking as 100% the values obtained with 500 nM of rosiglitazone.
STEM_200_sm_SuppFig4.tif9832KSupplemental Figure 4. Effect of rosiglitazone on adipogenesis of hMADS-1 cells. hMADS-1 cells were induced to differentiate as described in Figure 1. Rosiglitazone was added at the indicated concentrations for different periods of time. At day 16, (A) cells were fixed and stained by Oil red O, (B) GPDH activity was determined and (C) UCP1 and CIDEA mRNA levels were measured by quantitative RT-PCR. Results are mean ± SEM of two independent experiments performed on different series of cells and are expressed in (C) by taking as 100% the values obtained at 100 nM rosiglitazone at day 16. * p < 0.05 vs. untreated cells
STEM_200_sm_SuppTab1.doc36KSupplemental Table 1. Sequence of primers used for gene expression analysis.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.