Introduction

Angiogenesis corresponds to the formation of new vessels from the pre-existing vascular network. This phenomenon, critical for embryonic development, still occurs in the adult and plays a key role in physiological (reproductive cycle in women) and pathophysiological remodeling processes (wound healing, atherosclerosis, tumor growth or metastasis formation).¹ Angiogenesis is also strongly suspected of promoting non-neglasic tissue growth and particularly excessive adipose tissue development leading to obesity. It is thus tempting to speculate that inhibition of the angiogenic process might be an interesting approach to reduce fat mass development. A recent work supports this hypothesis since Rupnick et al² have demonstrated that antiangiogenic treatments performed on several murine models of obesity resulted in significant body weight and adipose tissue loss. However, the regulation of angiogenesis in adipose tissue is still not well characterized. In vitro studies have identified various proangiogenic factors produced by the adipocytes and involved in different steps of the angiogenic process. We and others have demonstrated the inductive effects of leptin on endothelial cell proliferation and migration.^3,4 Recently, we have shown that adipocytes produce gelatinase matrix metalloproteinases (MMPs), MMP-2 and -9,⁵ both enzymes known to enhance the degradation of the basement membrane and extracellular matrix allowing endothelial cell migration as well as proangiogenic factor release.⁶ Besides leptin and MMPs, adipocytes also secrete vascular endothelial growth factor (VEGF).⁷ It is thus suggested that, through secretion of proangiogenic factors, adipocytes may directly affect the growth and organization of the surrounding endothelial cells. However, there are no data available concerning mechanisms regulating adipocyte proangiogenic factor secretion during adipose mass growth.

Hypoxia is known to be a potent angiogenic inducer particularly in tumor progression.⁸ The best described hypoxia-dependent signaling pathway involves the activation of hypoxia inducible factors (HIFs), heterodimeric transcription factors consisting of a hypoxia-sensitive subunit (HIF-1, -2) and a constitutive subunit (ARNT/HIF-1, ARNT-2).⁹ Excessive adipose mass development involves adipocyte hypertrophia, due to increased lipogenesis associated with decreased lipolysis, as well as enhanced proliferation and differentiation of preadipocytes into adipocytes.¹⁰ Adipocyte hypertrophy, that leads to the formation of cells with a diameter up to 150 m, might favour the presence of local hypoxic areas within the adipose tissue since the diffusion limit of oxygen is considered at 100 m. Although the occurrence of hypoxic areas in growing fat pads remains to be demonstrated, hypoxia might be a promotor of angiogenesis in adipocytes.

We performed the present study to determine whether hypoxia regulates the expression of various adipocyte-secreted proangiogenic factors in the established murine preadipocyte model, the 3T3-F442A cell line. Differentiated cells were submitted to three different hypoxic stimuli: low oxygen pressure (5% O₂), cobalt chloride (CoCl₂) and desferrioxamine (DFO), both hypoxia mimics.^11,12 We demonstrated that hypoxia upregulated VEGF and leptin expression and was associated with a higher accumulation of HIF-1 in hypoxic preadipocyte nuclei. MMP-2 and -9 activities were also induced.

Materials and methods

Materials

Chemicals were obtained from Sigma (Saint Quentin Fallavier, France) and cell culture reagents from either Life Technologies (Cergy Pontoise, France) or Roche Diagnostics (Meylan, France). Primary antibodies were purchased from Santa Cruz Biotechnology (Le Perray-en-Yvelines, France) for the polyclonal rabbit antibody directed against VEGF and from Abcam (Novus Biologicals, Littleton, USA) for the polyclonal mouse antibody directed against the subunit of HIF-1. Peroxidase-conjugated antibodies were from Chemicon (Euromedex, Souffelweyersheim, France) and the enhanced chemiluminescence (ECL) kit from Pierce (Bezons, France). The prestained protein marker and the protein determination kit were obtained from Bio-Rad (Ivry/Seine, France). Protease inhibitor tablets (Complete mini) were from Roche (Meylan, France). Lactate and glycerol concentration, lactate dehydrogenase (LDH) activity and glucose release were determined using kits from Sigma (Saint Quentin Fallavier, France) and from Biotrol Diagnostics (Chennevières, France), respectively. Determination of leptin protein level was performed using an ELISA kit specific for murine leptin (R&D Systems, Wiesbaden-Nordenstadt, Germany). RNA concentration was determined using Ribogreen (Molecular Probes, Leiden, Netherlands).

Cell culture and treatments

Cells from the 3T3-F442A murine preadipocyte cell line were seeded at a density of 1500 cells/cm². Cells were cultured until confluence in medium composed of DMEM supplemented by 10% of donor calf serum and an antibiotic mixture (50 U/ml penicillin and 50 l/ml streptomycin) in an atmosphere of 95% air–5% CO₂ at 37°C. When confluent, the cells were differentiated in DMEM containing 10% fetal calf serum supplemented with 50 nmol/l insulin. The differentiating medium was changed every 2 days.

After 6 days in the differentiating medium, the cells were rinsed with PBS. At this state of differentiation, 3T3-F442A murine adipocytes were shown to express the proangiogenic factors studied here.⁵ Adipocytes were placed in insulin- and serum-free DMEM containing 0.1% BSA and submitted, or not, to chemical or ambient hypoxia during the indicated times. Chemical hypoxia was induced by adding 1 mol/l of CoCl₂ or 100 mol/l of DFO to the cells maintained in normoxic conditions. Both compounds were diluted in insulin-free DMEM supplemented with 0.1% BSA. A preliminary study was performed to determine the optimal dose of each hypoxia mimic required to induce a maximal cell response (personal data). Untreated cells maintained in normoxia for the same time- period of treatment were used as controls. For ambient hypoxia, cells were incubated in a temperature- and humidity-controlled environmental chamber with an atmosphere containing 5% CO₂, 5% O₂, balance N₂. For each time-period incubation in low oxygen pressure, appropriate controls corresponding to cells maintained in normoxic atmosphere (95% air–5% CO₂) were performed in parallel. Conditioned media were then collected and stored at -20°C until analysis. Cells were rinsed twice with PBS and frozen at -20°C. To assess a putative toxicity of treatments, we determined the LDH activity in the conditioned media.

Determination of MMP activity by zymography

MMP-2 and -9 gelatinase activities were identified by electrophoresis in the presence of SDS in 10% polyacrylamide gels containing 1 mg/ml gelatin. Conditioned media (20 l) were loaded directly on gels and, after electrophoresis, proteins were renaturated in 2.5% Triton X-100 (15 min incubation repeated twice). Gels were incubated at 37°C for 16 h in 50 mmol/l Tris-HCl (pH 8.5), 0.02% NaN₃, and 5 mmol/l CaCl₂ to allow the enzymes to digest the gelatin. After incubation, the gels were stained with Coomassie Blue and clear areas indicated the presence of gelatinolytic activity. After scanning densitometry, gels were quantified using the NIH Image program (developed at the US National Institute of Health).

Protein extraction and Western blot analysis

Total cell proteins were extracted in 100 l of lysis buffer containing 20 mmol/l Tris-HCl (pH 7.5), 1% Nonidet P-40, 10% glycerol, 150 mmol/l NaCl, 1 mmol/l CaCl₂, 1 mmol/l MgCl₂, 1 mmol/l Na₃VO₄ and an antiprotease cocktail. For nuclear protein extraction, 3T3-F442A cells were harvested in PBS. After centrifugation (1000 g, 5 min, 4°C), the nuclei were isolated by resuspension of the cell pellets in 100 l of buffer A (plasmic membrane lysis buffer) (10 mmol/l HEPES, 10 mmol/l KCl, 0.1 mmol/l EGTA, 0.1 mmol/l EDTA, 0.1 mol/l dithiotreitol (DTT), 10 mg/ml phenylmethylsulfonyl fluoride (PMSF) and a mixture of protease inhibitors). After 15 min in ice, 8 l of 10% Nonidet P-40 was added. After centrifugation (17 000 g, 30 s, 4°C), the pellet-containing nuclei was disrupted by adding 30 l of a buffer B (nuclear envelope lysis buffer) containing 20 mmol/l HEPES (pH 7.9), 0.4 mol/l NaCl, 0.1 mmol/l EGTA, 0.1 mmol/l EDTA, 0.1 mol/l DTT, 10 mg/ml PMSF and a cocktail of protease inhibitors. The lysates, after shaking for 15 min at 4°C, were centrifuged at 17 000 g for 5 min at 4°C. The supernatants containing the protein nuclear extracts were then collected. Protein concentrations were determined using a protein determination kit.

30 to 40 g of proteins were separated by SDS-PAGE under denaturing conditions. After transfer to nitrocellulose membranes, Ponceau staining was performed to verify the equal loading of the lanes. Membranes were then blocked overnight at 4°C with a blocking buffer (50 mmol/l Tris-HCl (pH 7.5), 200 mmol/l NaCl, 0.05% polyethylene-sorbitan monolaurate (Tween), 3% BSA and 10% horse serum). They were incubated with a primary antibody for 90 min, rinsed, blocked for 40 min and incubated with the secondary antibody for 60 min. The immunocomplexes were visualized using the chemiluminescence reagent kit. The autoradiographs were scanned by an imaging densitometer and quantified using the NIH Image program.

RNA extraction and RT-PCR analysis

Total RNAs were extracted from 3T3-F442A cells according to the method of Chomczynski and Sacchi.¹³ RNA concentrations were determined using a fluorimetric assay (Ribogreen). For reverse transcription (RT), 2 g total RNA was incubated with 200 U reverse transcriptase (SuperScript II, Gibco), dNTP (0.5 mmol/l), oligo(dT) (25 ng/l), DTT (0.01 mol/l) and reaction buffer in a final volume of 20 l at 42°C for 50 min and then at 70°C for 15 min. In some reaction mixtures, enzyme was omitted to ensure the absence of amplification of contaminating genomic DNA, which was always the case in the present study. After a final denaturation at 94°C for 4 min, 10 l cDNA was subjected to PCR consisting of a denaturation at 94°C for 1 min followed by 1 min of annealing at T°C (depending on the target gene) and 90 s of elongation at 72°C for n cycles. The last cycle ended with 7 min of elongation at 72°C. The annealing temperature (T) and number (n) of PCR cycles was optimized for each pair of primers: MMP-2 and -9, leptin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH): 56°C, 35 cycles; VEGF: 54°C, 35 cycles. Expected size of PCR products: MMP-2: 407 pb, MMP-9: 374 pb, leptin: 353 pb, GAPDH: 319 pb. The specific primers used for VEGF were shown to amplify all VEGF splice variants.Table 1¹⁴

Table 1.

Full table

The PCR contained 0.4 mol/l of each primer, dNTP (0.2 mmol/l), MgCl₂ (1.5 mmol/l) reaction buffer and 5 U Taq polymerase (Promega) in a final volume of 50 l. The amplified cDNAs were size-fractioned by 1.5% agarose gel electrophoresis, visualized under UV with a 1% ethidium bromide staining. The gels were scanned and analyzed using the NIH Image program. Individual data were calculated as means of independent experiments and expressed as the ratio of MMP-2, MMP-9, leptin or VEGF over GAPDH expression.

Statistics

Results are expressed as control percentages.e.m. from (n) independent experiments. Statistical analysis was performed using Student's t-test for unpaired data or one-way ANOVA followed by the Bonferroni t-test when appropriate. Values of P<0.05 were considered statistically significant.

Results

Metabolic adaptative response of 3T3-F442A adipocytes to low oxygen pressure and chemical hypoxia

At 6 days after confluence, differentiated 3T3-F442A adipocytes were cultured in 5% O₂ atmosphere or maintained in normoxia for 24 h. In parallel, cells were or not treated with hypoxia mimics (1 mol/l CoCl₂ or 100 mol/l DFO) for 24 h. To assess the potential toxicity of the applied treatments, the activity of the LDH was determined in the conditioned media. No significant LDH activity was detected whatever the treatment (data not shown), which implies the absence of cell suffering in response to ambient or chemical hypoxia. As one of the hallmarks of hypoxic cells is to enhance anaerobic metabolism,¹⁵ we determined extracellular lactate release and glucose uptake from the conditioned media. As shown in Figure 1, hypoxic treatments significantly increased the secretion of lactate (Figure 1a): two-fold increase in response to ambient hypoxia and 20% increase in response to both hypoxia mimics, CoCl₂ and DFO, compared to untreated cells maintained in normoxia for the same time period. The glucose uptake of 3T3-F442A cells (Figure 1b) was also markedly enhanced: 30% increase in 5% O₂ atmosphere or after treatment with CoCl₂ and two-fold enhancement in response to DFO. In the same conditions, the secretion of glycerol in the extracellular media, assessed as an index of the basal lipolytic activity, was not significantly modified (Figure 1c). Moreover, no alteration of the expression of the hormone-sensitive lipase (HSL), the rate limiting enzyme of the lipolytic process, was evidenced whatever the hypoxic treatment (Figure 2a). Thus, in hypoxic conditions (real or mimicked with CoCl₂ or DFO), the metabolism of the differentiated 3T3-F442A adipocytes switched to anaerobic metabolism, but without alteration of the basal lipolytic activity.

Figure 1.

Ambient and chemical hypoxia induced a switch to anaerobic metabolism, but did not modify the basal lipolytic activity. Differentiated 3T3-F442A adipocytes were maintained in normoxia (ct) or placed in 5% O₂ atmosphere (Hyp) for 24 h (n=8). In parallel, murine cells were treated or not for 24 h with 1 mol/l CoCl₂ or 100 mol/l DFO (n=3). Extracellular lactate secretion (a), glucose concentration (b) as well as glycerol release (c) were determined in the collected media. Values are meanss.e.m. expressed as percentage of the values obtained for the untreated cells maintained in normoxia for 24 h (ct). ^*P<0.05 and ^**P<0.01 vs normoxic untreated cells.

Full figure and legend (55K)

Figure 2.

Ambient and chemical hypoxia increased VEGF protein level without alteration in the amount of HSL protein. Differentiated murine 3T3-F442A adipocytes were placed in normoxia (Norm) or in 5% O₂ atmosphere (Hyp) for 24 h. In parallel, 3T3-F442A cells maintained in normoxia were treated or not (Ct) for 24 h with 1 mol/l CoCl₂ or 100 mol/l DFO. Total proteins were extracted and the expression of the hormone-sensitive lipase, HSL, (a) and of the VEGF, (b) were assessed by Western blot analyses. Representative autoradiographs from three independent experiments are shown.

Full figure and legend (52K)

Effects of ambient and chemical hypoxia on the protein levels of VEGF and leptin and on the activity of MMPs in differentiated adipocytes

To determine whether hypoxia regulated in differentiated adipocytes the expression of VEGF and of leptin, 3T3-F442A cells were incubated under 5% O₂ atmosphere or maintained in normoxia for 24 h. In parallel, cells were or not submitted to 1 mol/l CoCl₂ or 100 mol/l DFO for 24 h. Western blot analyses were performed on total cellular protein extracts using an antibody directed against the VEGF. As shown in Figure 2b, the amounts of the VEGF protein were significantly increased in response to both low oxygen pressure or hypoxia mimics (from 1.5- to two-fold increase in cells cultured in hypoxic conditions compared to untreated cells maintained in normoxia for the same time period, n=3, P<0.05). As the level of leptin secretion in differentiated 3T3-F442A cells is very low and could not be detected by ELISA analyses performed on conditioned media (data not shown), the influence of hypoxia on leptin secretion could not be studied.

To assess the effect of hypoxia on the activity of MMP-2 and -9, differentiated adipocytes were placed in 5% O₂ atmosphere for 2, 4, 8 or 24 h or maintained for the same time period in normoxia. In parallel, cells were or not submitted to 1 mol/l CoCl₂ or 100 mol/l DFO for 24 h. The gelatin zymography performed on conditioned media clearly showed that in cells cultured in ambient hypoxia, the gelatinase activity of MMP-2 and -9 was increased compared to cells maintained in normoxia for the same time-period (Figure 3). Interestingly, the extent of the enhancement in MMP activity was maximal after 2 h-period in low oxygen pressure. The stimulatory effect of hypoxia was less pronounced with increasing time period of incubation in hypoxic atmosphere but still remained present. In parallel, gelatin zymography performed on the media of cells treated with CoCl₂ or DFO for 24 h, also evidenced that both hypoxia mimic treatments led to an increase in MMP-2 and -9 activity compared to untreated cells (Figure 3). It can be noticed that similarly to ambient hypoxia, chemical hypoxia had a more pronounced regulatory effect on MMP-9 than MMP-2 activity.

Figure 3.

Ambient and chemical hypoxia enhanced the gelatinase activity of MMP-2 and MMP-9. Differentiated 3T3-F442A adipocytes were placed in normoxia (Norm) or in 5% O₂ atmosphere (Hyp) for the indicated times (2, 4, 8 or 24 h) (n=8). In parallel, 3T3-F442A cells maintained in normoxia were treated or not (Ct) for 24 h with 1 mol/l CoCl₂ or 100 mol/l DFO (n=4). The media were collected and analyzed by gelatin zymography. Representative Coomassie Blue staining of electrophoresis gels are shown. Densitometric analyses of the lytic area corresponding to the activity of MMP-9 and -2 were performed and results are expressed as percentage of the values obtained in normoxic untreated cells at the corresponding times. Values are meanss.e.m. of (n) independent experiments. ^*P<0.05 and ^**P <0.01 vs normoxic untreated cells at the corresponding times.

Full figure and legend (121K)

Effects of ambient and chemical hypoxia on the expression of VEGF, leptin and MMPs in differentiated adipocytes

To determine the effect of hypoxic conditions on the mRNA levels of the proangiogenic factors, differentiated adipocytes were placed in 5% O₂ atmosphere for 2, 4, 8 or 24 h or maintained for the same time period in normoxia. In parallel, cells were or not submitted to 1 mol/l CoCl₂ or 100 mol/l DFO for 24 h. RT-PCR analyses were performed on total RNAs extracted using specific primers for leptin, VEGF, MMPs and GAPDH cDNAs. GAPDH was used as internal control and the amounts of the amplified leptin, VEGF and MMPs cDNAs were normalized to the amount of the corresponding amplified GAPDH cDNAs. As shown Figure 4a, the levels of the amplified leptin cDNA were found to be increased when cells were cultured under low oxygen pressure from 2 to 8 h, with a maximal effect observed after 2h incubation period (1.5-fold increase compared to cells maintained in normoxia for the same time period). No regulatory effect was observed when cells were cultured under low oxygen pressure for up to 24 h whereas in normoxic cells treated with the hypoxia mimics for 24 h, 1.8- to 2.5-fold (for DFO and CoCl₂, respectively) increase in leptin expression was evidenced compared to untreated cells.

Figure 4.

Ambient and chemical hypoxia increased the expression of leptin and VEGF. Differentiated 3T3-F442A adipocytes were placed in normoxia (Norm) or in 5% O₂ atmosphere (Hyp) for the indicated times (2, 4, 8 or 24 h) (n=8). In parallel, differentiated murine cells maintained in normoxia were treated or not (Ct) for 24 h with 1 mol/l CoCl₂ or 100 mol/l DFO (n=3). Total RNAs were extracted from the cells and RT-PCR analyses were performed using specific primers for leptin (a), VEGF (b) and GAPDH cDNAs. Representative ethidium bromide staining of electrophoresis gels are shown. Densitometric analyses of the amplified cDNAs normalized to GAPDH were performed and results are expressed as percentage of the values obtained in normoxic untreated cells at the corresponding times. Values are meanss.e.m. of (n) independent experiments. ^*P<0.05 and ^** P <0.01 vs normoxic untreated cells at the corresponding times.

Full figure and legend (182K)

Concerning the expression of VEGF, the levels of the three isoforms of VEGF (120, 164 and 188) were found to be upregulated when differentiated adipocytes were placed in 5% O₂ atmosphere for 4–8 h compared to cells maintained in normoxia for the same time period (Figure 4b). Similarly to leptin, no alteration of the VEGF mRNA amounts was observed after 24h incubation period in ambient hypoxia whereas for the same time period, both hypoxia mimics led to an increase in VEGF expression (Figure 4b).

Finally, concerning the expression of MMPs, no effect of low oxygen pressure was observed whatever the time period of incubation on the level of the amplified cDNAs for MMP-2 and -9 (Figure 5). However, both DFO and CoCl₂ treatments of normoxic cells for 24 h were associated with a significant increase in MMP-2 and -9 mRNA amounts (1.5- to 6-fold increase for MMP-2 and -9 expression, respectively (n=3)).

Figure 5.

Ambient hypoxia did not regulate the expression of MMP-2 and -9 at the transcript level whereas chemical hypoxia did. Differentiated 3T3-F442A adipocytes were placed in normoxia (Norm) or in 5% O₂ atmosphere (Hyp) for the indicated times (2, 4, 8 or 24 h) (n=8). In parallel, 3T3-F442A cells maintained in normoxia were treated or not (Ct) for 24 h with 1 mol/l CoCl₂ or 100 mol/l DFO (n=3). Total RNAs were extracted from the cells and RT-PCR analyses were performed using specific primers for MMP-2 and -9 cDNAs. Representative ethidium bromide staining of electrophoresis gels are shown.

Full figure and legend (65K)

Influence of low oxygen pressure and hypoxia mimics on HIF-1 protein accumulation in 3T3-F442A adipocyte nuclei

To define whether hypoxic proangiogenic factor regulation was associated with HIF-1 stabilization into the adipocyte nuclei, differentiated 3T3-F442A cells were cultured in 5% O₂ atmosphere or maintained in normoxia for 2, 4, 8 or 24 h in the presence or not of 100 mol/l DFO. Nuclear proteins were extracted, and Western blot analyses were performed using a specific antibody against HIF-1. Both ambient and chemical hypoxia were shown to induce a time-dependent increase in the HIF-1 protein amount detected in the nuclei (Figure 6). The maximal increase in the accumulation of HIF-1 protein in nuclei was observed after 4 h-DFO treatment and 8 h-ambient hypoxia (3.5-fold increase as compared with untreated normoxic cell, P<0.01, n=4). It recovered basal levels after 24 h.

Figure 6.

Induction of HIF-1 expression in 3T3-F442A adipocyte nuclei in response to ambient and chemical hypoxia. Murine 3T3-F442A adipocytes were submitted or not (Norm or CH) either to 5% O₂ atmosphere, Hyp, (a) or 100 mol/l DFO (b) for the indicated times (2, 4, 8 and 24 h). Nuclear proteins were extracted and the expression of HIF-1 was assessed by Western blot analyses. Representative autoradiographs are shown. Densitometric analyses were performed and results are expressed as percentage of the values obtained for normoxic untreated cells. Values are meanss.e.m. of four independent experiments. ^*P<0.05 and ^**P <0.01 vs normoxic untreated cells at the corresponding times.

Full figure and legend (152K)

Discussion

During obesity settlement, both increases in adipocyte size and number are supposed to be supported by the extension of the local microcirculation. Although angiogenesis occurrence within growing adipose tissue still remains to be demonstrated, angiogenic properties of adipose grafts¹⁶ as well as adipocyte-derived proangiogenic factor secretion are largely described. However, mechanisms inducing such angiogenic process are completely unknown. Inflammation¹⁷ and hypoxia,¹⁵ the main angiogenesis regulators described in non-neoplasic tissues, may represent two possible candidates for angiogenesis regulation in growing adipose tissue.

To study the effect of hypoxia on the secretory activity of adipocytes, experiments were performed in vitro using the well-defined cellular model of murine 3T3-F442A adipocytes whose features are closely related to those of white adipocytes when differentiated in a medium supplemented with serum and insulin.¹⁸ 3T3-F442A adipocytes were submitted to either low oxygen pressure (5% O₂) or to hypoxia mimics, CoCl₂ or DFO. To first determine adipocyte sensitivity to hypoxia, the glycolytic activity of 3T3-F442A adipocytes was studied. Indeed, hypoxia is known to switch the cellular metabolism to limit oxygen consumption and compensate for the ATP synthesis decrease,¹⁵ resulting in enhanced lactate production¹⁹ and glucose uptake.²⁰ Both glycolytic indexes were markedly increased in adipocytes in response to a 24 h period in low oxygen pressure or in the presence of hypoxia mimics (Figure 1a and b). Thus, 3T3-F442A adipocytes are sensitive to both hypoxic conditions used in the present study. It has to be noticed that adipose tissue has already been described to be an in vivo lactate producer.²¹ Our results suggest that the release of lactate in vivo by adipocytes might represent a relevant index of oxygen level within adipose mass. In the same conditions, no alteration of the basal lipolytic activity, assessed by the extracellular glycerol production (Figure 1c) as well as by the basal expression of HSL (Figure 2a) was observed. This result implies that the basal lipolytic activity is independent of oxygen pressure and confirms that the hypoxic treatments do not lead to cellular toxicity as demonstrated by the lack of release of LDH activity. Moreover, it also agrees with an in vivo study reporting that hypobaric hypoxia did not alter basal lipolysis but decreased its hormonal control.²²

In parallel, under the same conditions a marked increased in VEGF protein as well as its mRNA levels was observed, whatever the kind of hypoxia applied to the adipocytes (Figures 2b and 4b). This is in agreement with a study performed on rat omental adipose tissue showing that hypoxia is an inducer of VEGF expression in omental adipocytes.²³ Thus, hypoxic adipocytes are able to adapt not only their metabolism but also their secretory activity to oxygen level. The adipocyte secretory activity is an emerging observation. Indeed, increasing number of studies have described the ability of adipocytes to produce and secrete numerous factors such as cytokines and growth factors.²⁴ Leptin is one of the most studied adipocyte-derived factor. Our group has also recently demonstrated that adipocytes produce two gelatinase matrix metalloproteinases involved in the degradation of the basement membrane and extracellular matrix, MMP-2 and -9.⁵ The present results clearly demonstrate that, in addition to VEGF expression, hypoxia also enhanced the expression of leptin (Figure 4a) and MMPs (Figures 3 and 5). The extent of the increase in leptin expression under hypoxic conditions is similar to that described in a recent study performed in human PAZ6 cells.²⁵ Concerning the effect of hypoxia on MMPs, studies performed on other cell types showed that the effects of hypoxia on MMP expression and activity are highly dependent on the duration of the treatment and on cell type.^26,27 Under our conditions, although both chemically mimicked and real hypoxia markedly upregulated MMP-2 and -9 activity secreted in conditioned media, only hypoxia mimic treatments were associated with an increase in both gelatinase mRNAs. Different intracellular pathways specifically induced by each kind of hypoxia generating system might explain the difference between ambient and chemical hypoxia. Indeed, both hypoxia mimics are thought to act mainly by inhibiting the newly discovered oxygen- and Fe²⁺-dependent enzyme, HIF-1 prolyl-4-hydroxylase (HIF-PH),^28,29 leading to HIF-1 stabilization and the activation of cellular hypoxia-dependent pathways under normal oxygen level.⁸ However, we observed the upregulation of MMP activity under a short time period of low oxygen pressure without a concomitant increase in mRNA transcripts. This suggests that post-transcriptional mechanisms could be quickly involved, for instance an enhanced maturation involving other MMP partners such as the membrane-type (MT) 1-MMP, MMP-3³⁰ and/or the endogenous tissue inhibitor (TIMP-2),³¹ known to be hypoxia-sensitive. TIMP-2 has already been described in adipocytes,⁵ but other investigations are required to clarify the presence of MT-MMP and MMP-3 as well as their role in the regulation of MMP-2 and -9 activities by hypoxia.

Finally, most hypoxia-sensitive genes are regulated through the activity of the transcription factor HIF³² which binds to hypoxia responsive elements (HREs) contained in their promoter region.³³ Therefore, we analyzed the pattern of HIF-1 protein expression in hypoxic 3T3-F442A adipocyte nuclei. Both chemical and ambient hypoxia were found to be associated with a rapid increase of HIF-1 protein amounts in nuclei (Figure 6). Moreover, the time course of HIF-1 accumulation was consistent with the hypoxia-induced stabilization of HIF-1 protein.⁹ Since the HRE sequence has been recently evidenced in the leptin promoter region,^34,35 it is suggested that an HIF-1-dependent pathway may be involved in the hypoxia-mediated upregulation of adipocyte-derived leptin expression as is well known for VEGF in other cell types.³⁶ However, HIF-1 involvement in the increase of leptin mRNA amount in response to 2 h-ambient hypoxia seems to be compromised as the accumulation of HIF protein into hypoxic adipocyte nuclei is significant only after 4 h in low 5% O₂ atmosphere. It can thus be supposed that leptin expression is enhanced by another hypoxia-sensitive transcription factor than HIF at least for short time period of ambient hypoxia.³⁷

In conclusion, 3T3-F442A adipocytes are sensitive to hypoxic conditions and adapt their secretory activity to the oxygen levels. Since VEGF, leptin and MMP-2 and -9 are factors involved in the process of angiogenesis, it is suggested that hypoxic adipocytes may increase their secretion of angiogenic factors to stimulate the formation of new blood vessels. Obesity is associated, in humans, with hyperleptinemia³⁸ and is due to adipocyte hypertrophy and hyperplasia.¹⁰ Although physiological occurrence and effects of hypoxia remain to be demonstrated in growing adipose tissue, it is tempting to speculate that reduction of oxygen supply to adipocytes could initiate an increased expression of adipocyte-derived angiogenic factors and the stimulation of the neovascularization. Physiological relevance merits further investigation.

References

1. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000; 407: 249–257. | Article | PubMed | ISI | ChemPort |
2. Rupnick MA, Panigrahy D, Zhang CY, Dallabrida SM, Lowell BB, Langer R, Folkman MJ. Adipose tissue mass can be regulated through the vasculature. Proc Natl Acad Sci USA 2002; 29: 29.
3. Bouloumie A, Drexler HC, Lafontan M, Busse R. Leptin, the product of Ob gene, promotes angiogenesis. Circ Res 1998; 83: 1059–1066. | PubMed | ISI | ChemPort |
4. Sierra-Honigmann MR, Nath AK, Murakami C, Garcia-Cardena G, Papapetropoulos A, Sessa WC, Madge LA, Schechner JS, Schwabb MB, Polverini PJ, Flores-Riveros JR. Biological action of leptin as an angiogenic factor. Science 1998; 281: 1683–1686. | Article | PubMed | ISI | ChemPort |
5. Bouloumie A, Sengenes C, Portolan G, Galitzky J, Lafontan M. Adipocyte produces matrix metalloproteinases 2 and 9: involvement in adipose differentiation. Diabetes 2001; 50: 2080–2086. | Article | PubMed | ISI | ChemPort |
6. Chang C, Werb Z. The many faces of metalloproteases: cell growth, invasion, angiogenesis and metastasis. Trends Cell Biol 2001; 11: S37–S43. | Article | PubMed | ISI | ChemPort |
7. Claffey KP, Wilkison WO, Spiegelman BM. Vascular endothelial growth factor. Regulation by cell differentiation and activated second messenger pathways. J Biol Chem 1992; 267: 16317–16322. | PubMed | ISI | ChemPort |
8. Brahimi-Horn C, Berra E, Pouyssegur J. Hypoxia: the tumor's gateway to progression along the angiogenic pathway. Trends Cell Biol 2001; 11: S32–S36. | Article | PubMed | ISI | ChemPort |
9. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helix–loop–helix-PAS heterodimer regulated by cellular O₂ tension. Proc Natl Acad Sci USA 1995; 92: 5510–5514. | Article | PubMed | ChemPort |
10. Hirsch J, Batchelor B. Adipose tissue cellularity in human obesity. Clin Endocrinol Metab 1976; 5: 299–311. | PubMed |
11. Wang GL, Semenza GL. General involvement of hypoxia-inducible factor 1 in transcriptional response to hypoxia. Proc Natl Acad Sci USA 1993; 90: 4304–4308. | Article | PubMed | ChemPort |
12. Wang GL, Semenza GL. Desferrioxamine induces erythropoietin gene expression and hypoxia-inducible factor 1 DNA-binding activity: implications for models of hypoxia signal transduction. Blood 1993; 82: 3610–3615. | PubMed | ISI | ChemPort |
13. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal Biochem 1987; 162: 156–159. | Article | PubMed | ISI | ChemPort |
14. Poltorak Z, Cohen T, Sivan R, Kandelis Y, Spira G, Vlodavsky I, Keshet E, Neufeld G. VEGF145, a secreted vascular endothelial growth factor isoform that binds to extracellular matrix. J Biol Chem 1997; 272: 7151–7158. | Article | PubMed | ISI | ChemPort |
15. Wenger RH. Cellular adaptation to hypoxia: O₂-sensing protein hydroxylases, hypoxia-inducible transcription factors, and O₂-regulated gene expression. FASEB J 2002; 16: 1151–1162. | Article | PubMed | ISI | ChemPort |
16. Cooper JD, Pearson FG, Patterson GA, Todd TR, Ginsberg RJ, Goldberg M, DeMajo WA. Technique of successful lung transplantation in humans. J Thorac Cardiovasc Surg 1987; 93: 173–181. | PubMed |
17. Murdoch C, Finn A. Chemokine receptors and their role in vascular biology. J Vasc Res 2000; 37: 1–7. | Article | PubMed | ISI | ChemPort |
18. Green H, Kehinde O. Spontaneous heritable changes leading to increased adipose conversion in 3T3 cells. Cell 1976; 7: 105–113. | Article | PubMed | ISI | ChemPort |
19. Firth JD, Ebert BL, Pugh CW, Ratcliffe PJ. Oxygen-regulated control elements in the phosphoglycerate kinase 1 and lactate dehydrogenase A genes: similarities with the erythropoietin 3' enhancer. Proc Natl Acad Sci USA 1994; 91: 6496–6500. | Article | PubMed | ChemPort |
20. Loike JD, Cao L, Brett J, Ogawa S, Silverstein SC, Stern D. Hypoxia induces glucose transporter expression in endothelial cells. Am J Physiol 1992; 263: C326–C333. | PubMed | ChemPort |
21. Jansson PA, Smith U, Lonnroth P. Evidence for lactate production by human adipose tissue in vivo. Diabetologia 1990; 33: 253–256. | PubMed |
22. de Glisezinski I, Crampes F, Harant I, Havlik P, Gardette B, Jammes Y, Souberbielle JC, Richalet JP, Riviere D. Decrease of subcutaneous adipose tissue lipolysis after exposure to hypoxia during a simulated ascent of Mt Everest. Pflugers Arch 1999; 439: 134–140. | PubMed |
23. Zhang QX, Magovern CJ, Mack CA, Budenbender KT, Ko W, Rosengart TK. Vascular endothelial growth factor is the major angiogenic factor in omentum: mechanism of the omentum-mediated angiogenesis. J Surg Res 1997; 67: 147–154. | Article | PubMed | ISI | ChemPort |
24. Ailhaud G, Grimaldi P, Negrel R. Cellular and molecular aspects of adipose tissue development. Annu Rev Nutr 1992; 12: 207–233. | Article | PubMed | ISI | ChemPort |
25. Grosfeld A, Zilberfarb V, Turban S, Andre J, Guerre-Millo M, Issad T. Effects of hypoxia on human PAZ6 adipocytes. Diabetologia 2002; 45: 527–530. | PubMed |
26. Ben-Yosef Y, Lahat N, Shapiro S, Bitterman H, Miller A. Regulation of endothelial matrix metalloproteinase-2 by hypoxia/reoxygenation. Circ Res 2002; 90: 784–791. | Article | PubMed | ISI | ChemPort |
27. Canning MT, Postovit LM, Clarke SH, Graham CH. Oxygen-mediated regulation of gelatinase and tissue inhibitor of metalloproteinases-1 expression by invasive cells. Exp Cell Res 2001; 267: 88–94. | Article | PubMed | ISI | ChemPort |
28. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, Salic A, Asara JM, Lane WS, Kaelin Jr WG. HIF alpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O₂ sensing. Science 2001; 292: 464–468. | PubMed | ISI | ChemPort |
29. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell SJ, Kriegsheim A, Hebestreit HF, Mukherji M, Schofield CJ, Maxwell PH, Pugh CW, Ratcliffe PJ. Targeting of HIF-alpha to the von Hippel–Lindau ubiquitylation complex by O₂-regulated prolyl hydroxylation. Science 2001; 292: 468–472. | PubMed | ISI | ChemPort |
30. Ogata Y, Enghild JJ, Nagase H. Matrix metalloproteinase 3 (stromelysin) activates the precursor for the human matrix metalloproteinase 9. J Biol Chem 1992; 267: 3581–3584. | PubMed | ISI | ChemPort |
31. Stetler-Stevenson WG. Matrix metalloproteinases in angiogenesis: a moving target for therapeutic intervention. J Clin Invest 1999; 103: 1237–1241. | PubMed | ISI | ChemPort |
32. Semenza GL. Oxygen-regulated transcription factors and their role in pulmonary disease. Respir Res 2000; 1: 159–162. | PubMed |
33. Semenza GL, Jiang BH, Leung SW, Passantino R, Concordet JP, Maire P, Giallongo A. Hypoxia response elements in the aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. J Biol Chem 1996; 271: 32529–32537. | Article | PubMed | ISI | ChemPort |
34. Ambrosini G, Nath AK, Sierra-Honigmann MR, Flores-Riveros J. Transcriptional activation of the human leptin gene in response to hypoxia: involvement of hypoxia-inducible factor 1. J Biol Chem 2002; 25: 25.
35. Grosfeld A, Turban S, André J, Cauzac M, Challier J-C, Hauguel-de Mouzon S, Guerre-Millo M. Transcriptional effect of hypoxia on placental leptin. FEBS 2001; 502: 122–126.
36. Minchenko A, Salceda S, Bauer T, Caro J. Hypoxia regulatory elements of the human vascular endothelial growth factor gene. Cell Mol Biol Res 1994; 40: 35–39. | PubMed | ISI | ChemPort |
37. Dachs GU, Tozer GM. Hypoxia modulated gene expression: angiogenesis, metastasis and therapeutic exploitation. Eur J Cancer 2000; 36: 1649–1660. | Article | PubMed | ISI | ChemPort |
38. Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens TW, Nyce MR, Ohannesian JP, Marco CC, McKee LJ, Bauer TL. Serum immunoreactive-leptin concentrations in normal-weight and obese humans. N Engl J Med 1996; 334: 292–295. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Christine Arragon for excellent technical assistance and Dr Cecilia Holm (Department of Cell and Molecular Biology, Lund University, Sweden) for providing us polyclonal chicken antibody against the hormone-sensitive lipase (HSL).