Phosphatidylinositol 3-Kinase Is a Requirement for Insulin-Like Growth Factor I-Induced Differentiation, but not for Mitogenesis, in Fetal Brown Adipocytes

Abstract

In the present study we have examined the role of phosphatidylinositol 3-kinase (PI 3-kinase) in the insulin-like growth factor I (IGF-I)-signaling pathways involved in differentiation and in mitogenesis in fetal rat brown adipocytes. Activation of PI 3-kinase in response to IGF-I was markedly inhibited by two PI 3-kinase inhibitors (wortmannin and LY294002) in a dose-dependent manner. IGF-I-stimulated glucose uptake was also inhibited by both compounds. The expression of adipogenic-related genes such as fatty acid synthase, malic enzyme, glycerol 3-phosphate dehydrogenase, and acetylcoenzyme A carboxylase induced by IGF-I was totally prevented in the presence of IGF-I and any of those inhibitors, resulting in a marked decrease of the cytoplasmic lipid content. Moreover, the expression of the thermogenic marker uncoupling protein induced by IGF-I was also down-regulated in the presence of wortmannin/LY294002. IGF-I-induced adipogenic- and thermogenic-related gene expression was only partly inhibited by the p70^S6k inhibitor rapamycin. In addition, pretreatment of brown adipocytes with either wortmannin or LY294002, but not with rapamycin, blocked protein kinase C ζ activation by IGF-I. In contrast, IGF-I-induced fetal brown adipocyte proliferation was PI 3-kinase-independent. Our results show for the first time an essential requirement of PI 3-kinase in the IGF-I-signaling pathways leading to fetal brown adipocyte differentiation, but not leading to mitogenesis. In addition, protein kinase C ζ seems to be a signaling molecule also involved in the IGF-I differentiation pathways downstream from PI 3-kinase.

INTRODUCTION

Insulin-like growth factor I (IGF-I) has proven to be a mitogenic peptide with important functions in the regulation of growth, development, and differentiation in eukaryotic cells (reviewed in Refs. 1 and 2). Although the intracellular events that mediate IGF-I action are not fully understood, the regulation of both protein and lipid phosphorylation are thought to play a prominent role. It is assumed, however, that the signals may merge and diverge at the levels of several key intermediates that become activated upon binding to its receptors on the cell surface. The activated IGF-I receptor phosphorylates a variety of cellular proteins on tyrosine residues. One of the best studied substrates of the IGF-I/insulin receptor is the insulin receptor substrate-1 (IRS-1) (3–5). Tyrosine phosphorylated IRS-1 binds and activates several signaling molecules, including the 85-kDa subunit of phosphatidylinositol 3-kinase (PI 3-kinase) (6–8).

PI 3-kinase is a family of heterodimeric enzymes composed of a p85- regulatory subunit and a p110 catalytic subunit (9, 10). PI 3-kinase is a lipid kinase capable of phosphorylating phosphoinositides at the 3′-position of the inositol ring (11), and these lipids have been postulated as second messengers (11–13). Although during recent years much information has beed accumulated on the role of PI 3-kinase activities in tyrosine kinase receptor signal transduction or vesicle trafficking, little is known about the possible role of PI 3-kinase in cell differentiation. In this regard, Kaliman et al. (14) have recently implicated PI 3-kinase as an essential positive regulator of terminal differentiation of skeletal muscle cells.

Fetal brown adipocyte primary cultures offer a nonfibroblastic mesenchymal cell model that has proven to be an excellent system in which to study both proliferation and differentiation processes (15–19) and signal transduction (20). These cells show a high level of IGF-I receptor mRNA expression and bear a high number of high-affinity IGF-I binding sites per cell. In fetal brown adipocyte primary cultures, IGF-I behaved as a mitogen per se in a p21 ras protein content-dependent manner (15, 17, 21). With regard to differentiation, rat brown adipocytes differentiate at the end of fetal life on the basis of two programs: the adipogenic program related to lipid synthesis and the thermogenic program related to heat production associated with the expression of the uncoupling protein (UP) to yield an identifiable and functional tissue at birth (19, 22). The UP expression constitutes a unique molecular marker that distinguishes this cell type from any other mammalian adipose cell. Our previous work has shown that IGF-I, which stimulates PI 3-kinase activity in brown adipocytes (20), is also capable of inducing the expression of both adipogenic (23) and thermogenic (17) genes. However, the signal transduction pathways by which IGF-I is involved in the adipogenic and thermogenic differentiation of fetal brown adipocytes have not yet been clarified.

Accordingly, in the present study we demonstrate that treatment of brown adipocytes with PI 3-kinase inhibitors (wortmannin/LY294002), but not with p70^S6k inhibitor (rapamycin), impaired the IGF-I-induced effect on the expression of adipogenic- and thermogenic-related genes, while brown adipocyte proliferation remained unaltered. Our results show for the first time that other signaling molecules such as protein kinase C ζ (PKCζ) may be involved in IGF-I-induced brown adipocyte differentiation process.

RESULTS

The Effect of PI 3-Kinase Inhibitors on IGF-I-Induced IRS-1-Associated PI 3-Kinase Activity in Fetal Brown Adipocytes

We have recently demonstrated the stimulation of PI 3-kinase enzymatic activity in fetal brown adipocyte primary cultures treated with IGF-I (20). Wortmannin has proven to be a specific PI 3-kinase inhibitor that directly binds to, and inhibits, the catalytic (110 kDa) subunit of PI 3-kinase (24, 25). To establish concentrations of wortmannin that effectively inhibited PI 3-kinase activity in fetal brown adipocytes stimulated with IGF-I, quiescent cells were either stimulated for 5 min with 10 nm IGF-I or preincubated for 15 min with various doses of wortmannin and subsequently stimulated with 10 nm IGF-I for a further 5 min. Then, whole cell lysates (600 μg of protein) were subjected to immunoprecipitation with the anti-IRS-1 antibody, and the resulting immune complexes were assayed for PI 3-kinase activity as described in Materials and Methods. As shown in Fig. 1A (upper panel), there was a 90% inhibition of IGF-I-stimulated PI 3-kinase activity when cells were pretreated for 15 min with 10 nm wortmannin, and total inhibition was observed with the dose of 20 nm.

Figure 1.

Effect of Wortmannin and LY294002 on IGF-I-Stimulated PI 3-Kinase Activity and the Tyrosine Phosphorylation of the IGF-I Receptor and IRS in Fetal Brown Adipocytes A, 20-h serum-starved fetal brown adipocytes were incubated for 5 min at 37 C with 10 nm IGF-I or preincubated for 15 min with various doses of wortmannin and LY294002 followed by treatment for 5 min with 10 nm IGF-I. Control cells were cultured in the presence of wortmannin (20 nm) (C_w) or LY294002 (10 μm) (C_LY) or received an equivalent volume of dimethylsulfoxide (DMSO) (C). Cells were lysed and immunoprecipitated with an anti-IRS-1 antibody. The immune complexes were washed and immediately used for an in vitro phosphatidylinositol kinase assay as described in Materials and Methods. The conversion of phosphatidylinositol to phosphatidylinositol phosphate in the presence of[γ -³²P]ATP was analyzed by TLC. Results are representative of three independent experiments. B, Cells were treated for 5 min with 10 nm IGF-I or preincubated for 15 min with 20 nm wortmannin or 10 μm LY204002 and treated with 10 nm IGF-I for a further 5 min. Control cells received an equivalent volume of DMSO. Cells were then lysed, and immunoprecipitates (prepared using the anti-Tyr(P) monoclonal antibody Py72) were assayed for in vitro protein kinase activity. The position of the β-chain of the IGF-I receptor is indicated by an arrowhead. The positions of molecular weight markers (× 10⁻³) are shown on the left. C, Cells were stimulated as described in panel B, immunoprecipitated with anti-αp85 antibody, and analyzed by Western blotting with the anti-Tyr(P) antibody (4G10). The position of IRS (either IRS-1 and/or IRS-2) is indicated by an arrowhead. The positions of molecular weight markers (× 10⁻³) are shown on the left. The results shown in panels B and C are representative of at least three independent experiments.

To substantiate the results obtained with wortmannin, we examined whether a structurally unrelated compound, LY294002 (a flavonoid related to quercetin), which has been identified as a specific inhibitor of PI 3-kinase (26), also inhibits IGF-I-stimulated PI 3-kinase activity in brown adipocytes. As shown in Fig. 1A (lower panel) LY294002 inhibited IGF-I-stimulated PI 3-kinase activity in a dose-dependent manner: half-maximal effect was elicited at 2.5 μm concentration and its maximal effect at 10 μm.

To determine whether the effect of PI 3-kinase inhibitors on IGF-I-stimulated PI 3-kinase activation was due to alterations in IGF-I-stimulated receptor phosphorylation and/or phosphotransferase activity, cells were either stimulated for 5 min with 10 nm IGF-I or pretreated for 15 min with 20 nm wortmannin or 10μ m LY294002 and subsequently stimulated with 10 nm IGF-I for a further 5 min. Then, lysates were subjected to immunoprecipitation with the Py72 anti-Tyr(P) antibody and assayed for in vitro protein kinase activity as described in Materials and Methods. Figure 1B is a representative autoradiogram showing the tyrosine- phosphorylated proteins in the immunoprecipitates after separation by SDS-PAGE. The presence of 10 nm IGF-I caused a marked increase in the tyrosine phosphorylation of the 95-kDa band, which corresponds with the M_r of the β-subunit of the IGF-I receptor (20), no phosphorylation being observed in the control cells. The level of tyrosine phosphorylation of the 95-kDa band did not change significantly when cells were pretreated with 10 μm LY294002, this band being even higher upon 20 nm wortmannin pretreatment.

In the case of the IGF-I receptor, PI 3-kinase is stimulated by interaction of the p85-regulatory subunit of PI 3-kinase with tyrosine-phosphorylated IRS-1. To study whether this interaction is affected by wortmannin and LY294002, we prepared soluble cell lysates after incubation of cells with both PI 3-kinase inhibitors and IGF-I as described above. They were next immunoprecipitated with the αp85 antibody and analyzed by Western blotting with the anti-Tyr(P) antibody (4G10) (Fig. 1C). After IGF-I stimulation of brown adipocytes, p85 was associated with tyrosine-phosphorylated IRS (either IRS-1 and/or IRS-2). Wortmannin (20 nm) increased and LY294002 (10μ m) did not affect IGF-I-stimulated association of the p85 subunit of PI 3-kinase with IRS-1/IRS-2. Since what is being measured in this experiment (Fig. 1C) is both tyrosine phosphorylation on IRS-1/IRS-2 and its interaction with p85, a possible explanation is that IRS-1/IRS-2 is phosphorylated to a greater extent in the presence of wortmannin. This might be associated with the greater tyrosine kinase activity of the receptor in the presence of wortmannin seen in Fig. 1B.

Effect of PI 3-Kinase Inhibitors on IGF-I-Stimulated Glucose Transport in Brown Adipocytes

Recently it has been demonstrated, in newborn brown fat precursor cells, that PI 3-kinase is involved in the mechanism of insulin-induced glucose transport (27). The fact that glucose transport is induced in fetal brown adipocytes upon IGF-I stimulation (20) prompted us to investigate whether this effect could be blocked by PI 3-kinase inhibitors in our fetal primary cells. Quiescent cells were treated for 10 min with 10 nm IGF-I and then incubated for a further 5 min in the presence of 2-deoxy-D(1-³H)glucose. At the same time, another set of cells were pretreated for 15 min with various doses of wortmannin or LY294002 and subsequently stimulated with IGF-I as described above. As shown in Fig. 2A, IGF-I-induced glucose transport was supressed by wortmannin in a dose-dependent manner at the same doses that inhibited PI 3-kinase activity (Fig. 1A). This inhibition was statistically significant at 10 and 20 nm concentration of wortmannin, total inhibition being observed at 20 nm. When cells were pretreated with LY294002, under the same experimental conditions (Fig. 2B), IGF-I-induced glucose transport was also blocked at the same doses that inhibited PI 3-kinase activity (significant inhibition at 5 and 10μ m concentration and total inhibition at 10μ m).

Figure 2.

Effect of PI 3-Kinase Inhibitors on IGF-I-Induced Glucose Transport in Fetal Brown Adipocytes Cells (after 20 h of serum deprivation) were treated with 10 nm IGF-I for 10 min or preincubated for 15 min with various doses of wortmannin (A) or LY294002 (B) and subsequently treated for an additional 10 min with 10 nm IGF-I. Control cells were incubated with the corresponding volume of DMSO. Deoxyglucose (dGlc) transport was measured as described in Materials and Methods. Results are expressed as disintegrations per min/μg of protein and are means ± sem from four independent experiments. Statistical analysis by Student’s paired t test between values in the presence vs. in the absence of IGF-I is represented by (*) or between IGF-I plus wortmannin/LY294002 vs. those in the presence of IGF-I by (ns, •); *, •P < 0.01; ns, not significant.

PI 3-Kinase Inhibitors Down-Regulate IGF-I-Induced Expression of Adipogenic Genes in Fetal Brown Adipocytes

IGF-I is a growth factor also involved in adipogenic differentiation of fetal brown adipocytes (17–19, 23). To investigate whether PI 3-kinase plays a role in fetal brown adipocyte differentiation, we analyzed the effect of PI 3-kinase inhibitors on the expression of a set of IGF-I-induced adipogenic-related genes. Fetal brown adipocytes (after 4 h of attachment followed by culture for 20 h in a serum-free medium) were cultured for a further 24 h with 10 nm IGF-I both in the absence and presence of wortmannin (20 nm) or LY294002 (10μ m). As it has been shown that wortmannin is highly unstable in aqueous solutions (28), we replaced the medium from cells treated with wortmannin every 6 h. At the end of the culture time the expression of a set of genes involved in adipogenesis, such as fatty acid synthase (FAS) (the main adipogenic marker), acetylcoenzyme A carboxylase (ACC) (the rate-limiting enzyme for long chain fatty acid synthesis), malic enzyme (ME, a NADPH provider), and glycerol-3-phosphate dehydrogenase (G3PD, a sterification marker) was studied by Northern blot, as depicted in the representative experiment shown in Fig. 3. In the absence of IGF-I all the mRNA levels analyzed remained very low, regardless of the presence of PI 3-kinase inhibitors in the culture medium. Upon treatment of cells for 24 h with 10 nm IGF-I, a significant accumulation of FAS, ME, G3PD, and ACC mRNAs occurred, relative to untreated cells. Interestingly, when brown adipocytes were cultured in the presence of IGF-I together with wortmannin or LY294002 (at doses which completely inhibited PI 3-kinase enzymatic activity), the induction of all the adipogenic-related mRNAs analyzed was completely prevented, their expression remaining at the basal levels observed in control cells.

Figure 3.

PI 3-Kinase Inhibitors Down-Regulate IGF-I-Induced Expression of Adipogenic Genes in Fetal Brown Adipocytes Brown adipocytes after 20 h of serum deprivation were cultured for a further 24 h with 10 nm IGF-I, both in the absence and presence of 20 nm wortmannin or 10 μm LY294002. Control cells were cultured without IGF-I in the absence or presence of the same concentrations of PI 3-kinase inhibitors. Total RNA (10 μg) was submitted to Northern blot analysis and hybridized with labeled FAS, ME, G3PD, ACC, and β-actin cDNAs. Autoradiograms from a representative experiment of six are shown.

Effect of PI 3-Kinase Inhibitors on IGF-I-Induced ME Protein Content and ME Activity in Brown Adipocytes

We have reported that ME can be induced by hormonal stimuli in differentiating brown adipocyte primary cultures (16). Based on this, we next examined whether PI 3-kinase plays a role in the induction of ME in brown adipocytes treated with IGF-I. Fetal brown adipocytes were serum-starved for 20 h and subsequently cultured for a further 24 h with 10 nm IGF-I, both in the absence and presence of wortmannin (20 nm) or LY294002 (10μ m). At the end of the culture period, cells were lysed and equal amounts of protein (20 μg) were submitted to Western blot analysis with the anti-ME antibody. As shown in Fig. 4A, upon treatment with IGF-I, a significant increase in the ME protein content relative to untreated cells (6- to 7-fold increase) occurred, and this effect was totally precluded in the presence of wortmannin (20 nm) or LY294002 (10 μm) together with IGF-I. Furthermore, the changes observed in ME protein content in response to IGF-I and PI 3-kinase inhibitors were parallel to changes in its enzymatic activity, as shown in Fig. 4B.

Figure 4.

IGF-I-Induced ME Protein Content and ME Activity in Fetal Brown Adipocytes: Inhibition by Wortmannin and LY294002 A, Brown adipocytes were cultured in both the absence and presence of IGF-I and PI 3-kinase inhibitors as described in Fig. 3. At the end of the culture time, cells were lysed and total protein (20 μg) was submitted to SDS-PAGE, blotted to nylon membranes, and immunodetected with the anti-ME antibody. A representative experiment of three is shown. B, Cells were cultured as described above. At the end of the culture time, ME activity was determined as described in Materials and Methods. Enzyme activity is expressed as milliunits/mg protein and is the mean ± sem (n = 4–6).

PI 3-Kinase Inhibitors Decreased Lipid Content in Fetal Brown Adipocytes

Because wortmannin and LY294002 down-regulate the expression of adipogenic genes and also ME activity and protein content, we studied Nile Red fluorescence (a sensitive detector of cytoplasmic lipid content) (29), to assess the role of PI 3-kinase in the overall adipogenic program. Brown adipocytes were cultured as described in Fig. 3, and at the end of the culture period Nile Red fluorescence was analyzed in the Flow cytometer and quantitated in arbitrary units as shown in Fig. 5. The presence of 10 nm IGF-I in the culture medium for 24 h significantly doubled the lipid content detected in control cells. However, when cells were cultured with IGF-I together with 20 nm wortmannin or 10μ m LY294002, lipid content reached a level similar to that observed in control cells.

Figure 5.

Role of PI 3-Kinase in IGF-I-Increased Cytoplasmatic Lipid Content in Fetal Brown Adipocytes Cells were serum-deprived for 20 h and further cultured for 24 h with 10 nm IGF-I in both the absence and presence of various doses of wortmannin and LY294002. Cytoplasmatic lipid content was determined by Nile Red fluorescence at the end of the culture time. Mean intensities of Nile Red fluorescence (expressed in arbitrary units) ± sem from three independent experiments are shown. Statistical analysis by Student’s paired t test between values in the presence vs. in the absence of IGF-I is represented by (*) or between IGF-I plus wortmannin/LY294002 vs. those in the presence of IGF-I by (•); *, •P < 0.01.

Wortmannin and LY294002 Inhibited UP mRNA Content in Fetal Brown Adipocytes

Apart from its role in the adipogenic program, IGF-I is involved in the thermogenic differentiation of brown adipocytes by inducing the expression of the UP (17, 19). To investigate this further we examined whether PI 3-kinase could also mediate IGF-I-induced thermogenic-related gene expression. Accordingly, cells were cultured for 24 h with 10 nm IGF-I in both the absence and presence of 20 nm wortmannin and 10 μm LY294002, and total RNA was submitted to Northern blot analysis as shown in the representative experiment depicted in Fig. 6. Upon treatment with IGF-I, a significant accumulation in UP mRNA relative to control cells (cultured in the absence of IGF-I) occurred. However, UP mRNA expression induced by IGF-I was totally prevented when either wortmannin or LY294002 was added to the culture medium.

Figure 6.

Wortmannin and LY294002 Down-Regulate IGF-I-Induced UP mRNA Expression in Fetal Brown Adipocytes Brown adipocytes were cultured in both the absence and presence of IGF-I and PI 3-kinase inhibitors as described in Fig. 3. At the end of the culture time, total RNA (10 μg) was submitted to Northern blot analysis and hybridized with labeled UP and β-actin cDNAs. A representative experiment of six is shown.

Effect of Rapamycin on IGF-I-Induced Differentiation-Related Gene Expression in Brown Adipocytes

p70^S6K is a ubiquitous serine/threonine kinase that is activated by many mitogens through distinct signaling pathways (for review see Ref.30). p70^S6K has been reported to be a molecular downstream target of PI 3-kinase (31–33). In view of the results presented here, it was important to clarify whether p70^S6K lies in the signaling pathway that mediated IGF-I-induced differentiation of brown adipocytes. Because the immunosuppressant rapamycin is a selective inhibitor of p70^S6K activation in many cell types (34), we examined the effect of rapamycin on the expression of both adipogenic and thermogenic genes induced by IGF-I. Figure 7 is a representative Northern blot of brown adipocytes cultured for 24 h with 10 nm IGF-I in both the absence and presence of 25 ng/ml rapamycin, which has been proved to completely block p70^S6K activity. As shown in Fig. 7, the expression of both adipogenic (FAS, ME, G3PD, ACC) and thermogenic (UP) markers was partly precluded in the presence of rapamycin compared with IGF-I-treated brown adipocytes.

Figure 7.

Effect of Rapamycin on IGF-I-Induced Adipogenic and Thermogenic mRNA Expression in Fetal Brown Adipocytes Brown adipocytes (after 20 h of serum starvation) were cultured for a further 24 h with 10 nm IGF-I in both the absence and presence of 25 ng/ml rapamycin. Control cells were cultured without IGF-I in both the absence and presence of 25 ng/ml rapamycin. At the end of the culture time, total RNA (10 μg) was submitted to Northern blot analysis and hybridized with labeled FAS, ME, G3PD, ACC, UP, and β-actin cDNAs. A representative experiment of four is shown.

IGF-I-Induced PKCζ Activity Is Inhibited by Wortmannin and LY294002 in Brown Adipocytes

Recent data support the view that PKCζ, which has recently found to be involved in adipogenesis (35), is activated by phosphatidylinositol 3,4,5-trisphosphate (PIP₃) (36). Furthermore PKCζ is expressed in brown adipocytes, and its activity is induced by IGF-I (20). These previous findings prompted us to investigate whether the blockade of PI 3-kinase could interfere with IGF-I-induced PKCζ activity in brown adipocytes. Quiescent cells (20 h serum-starved) were either stimulated with IGF-I (10 nm) for 5 min or pretreated for 15 min with 20 nm wortmannin or 10 μm LY294002 before the addition of IGF-I for a further 5 min. Then, cells were lysed and assayed for PKCζ activity in the immune complexes as described in Materials and Methods. As shown in Fig. 8A, myelin basic protein (MBP) phosphorylation induced in the presence of IGF-I was totally prevented by both wortmannin and LY294002 pretreatments and reached the basal levels observed in control cells.

Figure 8.

Effect of PI 3-Kinase and p70^S6k Inhibitors on IGF-I-Induced PKCζ Activity in Fetal Brown Adipocytes A, 20-h serum-starved fetal brown adipocytes were incubated for 5 min at 37 C with 10 nm IGF-I or preincubated for 15 min with wortmannin (20 nm) and LY294002 (10 μm) followed by treatment for a further 5 min with 10 nm IGF-I. Control cells received an equivalent volume of DMSO. Cells were lysed and immunoprecipitated with an anti-PKCζ antibody. The immune complexes were washed and immediately used for MBP phosphorylation as described in Materials and Methods. Results are representative of four independent experiments. B, 20-h serum-starved fetal brown adipocytes were incubated for 5 min at 37 C with 10 nm IGF-I or preincubated for 15 min with rapamycin (25 ng/ml) followed by treatment for a further 5 min with 10 nm IGF-I. Cells were lysed and assayed for PKCζ activity. Results are representative of four independent experiments.

Next, we examined whether IGF-I-induced PKCζ activity was affected by the blockade of p70^S6K activity. Quiescent cells were either stimulated with IGF-I (10 nm) for 5 min or pretreated for 15 min with 25 ng/ml rapamycin before the addition of IGF-I for a further 5 min. Then, cells were lysed and assayed for PKCζ activity in the immune complexes. As shown in Fig. 8B, when fetal brown adipocytes were pretreated with the immunosuppressant rapamycin before the addition of IGF-I, PKCζ activity remained unchanged relative to that observed in IGF-I-induced cells.

PI 3-Kinase Inhibitors Did Not Block IGF-I-Induced Mitogenesis or Proliferating Cellular Nuclear Antigen (PCNA) Expression in Fetal Brown Adipocytes

Based on the fact that IGF-I is a complete mitogen in fetal brown adipocyte primary cultures (15, 17), we proceeded finally to investigate whether PI 3-kinase inhibitors could also block IGF-I-induced mitogenesis in our cells. Quiescent (20 h serum-starved) brown adipocytes were cultured for 24 h with 1.4 nm IGF-I (which maximally stimulates brown adipocyte growth) in both the absence and presence of 20 nm wortmannin and 10μ m LY204002, and [³H]thymidine incorporation was measured during the last 4 h of culture. As shown in Fig. 9A, fetal brown adipocytes cultured in the presence of 1.4 nm IGF-I increased[ ³H]thymidine incorporation by 3-fold relative to control cells, as previously described (21). The presence of wortmannin or LY294002 in the culture medium, together with IGF-I, did not significantly modify the levels of [³H]thymidine incorporation relative to those observed in IGF-I-treated cells. Furthermore, no significant effect of PI 3-kinase inhibitors was observed in the percentage of cells in S+G2+M phases of the cell cycle as compared with cells stimulated with IGF-I alone (Fig. 9B).

Figure 9.

PI 3-Kinase Inhibitors Did Not Suppress IGF-I-Induced Mitogenesis in Fetal Brown Adipocytes A, Quiescent brown adipocytes were cultured for 24 h with 1.4 nm IGF-I in both the absence and presence of 20 nm wortmannin and 10 μm LY294002. Control cells were cultured without IGF-I in both the absence or presence of the same doses of inhibitors. [³H]Thymidine incorporation into acid-insoluble material was determined during the last 4 h of culture. Results are expressed as disintegrations per min/dish and are means ± sem from six independent experiments. Statistical analysis by Student’s paired t test between values in the presence of IGF-I plus wortmannin/LY294002 vs. those in the presence of IGF-I did not reveal significance (ns). B, Cells were cultured for 24 h as described in panel A. At the end of the culture time, the percentage of cells in S+G2+M phases was determined as described in Materials and Methods. Results are means ± sem from three independent experiments. Statistical analysis by Student’s paired t test between values in the presence of IGF-I plus wortmannin/LY294002 vs. those in the presence of IGF-I did not reveal significance (ns). C, Brown adipocytes were cultured as described above. At the end of the culture time, cells were lysed and total protein (50 μg) was submitted to SDS-PAGE, blotted to a nylon membrane, and immunodetected with the anti-PCNA monoclonal antibody. A representative experiment of three is shown.

PCNA is a nuclear protein required for cell cycle progression and cellular proliferation (37). In brown adipocytes treated with IGF-I, there is an important increase of PCNA content as has recently been described (21). Figure 9C shows that PCNA levels induced by IGF-I remained unchanged regardless of the presence of PI 3-kinase inhibitors in the culture medium.

DISCUSSION

Despite recent advances in the understanding of both the function and structure of PI 3-kinase, the precise mechanism of its involvement in mediating IGF-I/insulin signaling is still not well understood. In this study we have shown that two structurally different inhibitors of PI 3-kinase, i.e. wortmannin and LY294002, blocked both the adipogenic and thermogenic differentiation programs induced by IGF-I in primary fetal rat brown adipocytes.

In previous reports, we have suggested that IGF-I might have a role leading brown adipose tissue to adipogenic and thermogenic differentiation before birth (19, 23). Furthermore, fetal brown adipose cells bear a high number of high-affinity IGF-I receptors (17), and this fact allowed us to characterize the very early events of the brown adipocyte IGF-I-signaling cascade from the receptor toward the nucleus (20). Among all of these events, PI 3-kinase enzymatic activity was significantly activated subsequent to its association with phosphorylated IRS-1. The results presented here demonstrate that wortmannin, at nanomolar concentrations, dramatically inhibits IGF-I-stimulated PI 3-kinase activity in brown adipocytes. Furthermore, the PI 3-kinase inhibitor LY294002, which is structurally unrelated to wortmannin, also inhibited IGF-I-stimulated PI 3-kinase activity in a concentration-dependent manner. However, neither of these two inhibitors disrupted other cellular events of the IGF-I-signaling cascade upstream from PI 3-kinase activation, such as β-chain receptor autophosphorylation, IRS-1 tyrosine phosphorylation, and its association with the p85 subunit of the PI 3-kinase.

Recent findings indicate that PI 3-kinase is required for the movement of glucose transporters to the cell membrane in both white and brown adipose tissues and in muscle cells (27, 38, 39). Moreover, overexpression of the catalytic subunit p110α of PI 3-kinase increases glucose transport with translocation of glucose transporters in 3T3L1 adipocytes (40, 41). Although in fetal brown adipocyte primary cultures we have previously described an induction of Glut4 mRNA levels following 24 h of treatment with insulin and IGF-I (23), in this paper we found that glucose transport increased significantly after 10 min treatment with IGF-I (20, 23), probably due to Glut4 translocation. Both wortmannin and LY294002 impaired IGF-I-induced glucose transport at the same concentrations at which they inhibited PI 3-kinase activity. These results provide evidence that IGF-I-induced glucose transport in brown adipocytes during fetal development is dependent on PI 3-kinase activation.

It is known that IGF-I plays a dual role by inducing both mitogenesis and differentiation in fetal brown adipocytes (17–19, 23). With regard to differentiation, our results show that the expression of several adipogenic genes induced by IGF-I is completely prevented by treatment of brown adipocytes with PI 3-kinase inhibitors. The induction of ME activity was also impaired according to its protein content. As a result, the fat droplet content of brown adipocytes newly synthesized in the presence of IGF-I measured by Nile red fluorescence was significantly reduced. Because the expression of the thermogenic differentiation marker UP is also inhibited in the presence of wortmannin/LY294002, the results presented here implicate PI 3-kinase activation as a crucial step in the brown adipocyte adipogenic and thermogenic differentiation-signaling pathways.

Previous findings in confluent 3T3L1 mouse fibroblasts undergoing insulin-induced adipogenic differentiation demonstrate that insulin increases the ras.GTP/ras.GDP ratio (42). In addition, transforming ras transfection induced 3T3L1 adipogenic differentiation in the absence of IGF-I (43). On the other hand, insulin-stimulated glucose uptake has been shown by a number of groups to be dependent on activation of PI 3-kinase (39, 44, 45). It is not known to what extent IRS-1/PI 3-kinase and ras.GTP reside in distinct signaling branches or whether significant cross-talk occurs. Recently, a reciprocal relationship between PI 3-kinase and p21-ras has been demonstrated to exist (46–48). The fact that in brown adipocytes IGF-I increased the expression of differentiation-related genes in parallel with an increase in the amount of p21-ras.GTP active form (18), together with the data presented here, gives rise to the possibility that IRS-1-associated PI 3-kinase in conjunction with ras.GTP is required for the signaling involved in inducing and/or maintaining the differentiation state of brown adipocytes before birth. However, the relative contribution of these two pathways leading to PI 3-kinase activation and hence brown adipocyte differentiation in response to IGF-I deserves further experimental work.

p70^S6K has been identified as a molecular downstream target of PI 3-kinase (30, 31–33). Our results demonstrate that inhibition of the phosphorylation and activation of p70^S6K with the immunosuppressant rapamycin partly, but not totally, inhibited IGF-I-induced adipogenic and thermogenic related gene expression. We can therefore suggest that in addition to p70^S6K, other molecules that are activated downstream from PI 3-kinase might participate in the molecular cascade leading to the nucleus where gene expression is regulated. In this regard, PKCζ has been shown to be activated by the PI 3-kinase product PIP₃ (36), and its activity is induced in brown adipocytes upon IGF-I stimulation, in parallel with cell proliferation (20). The fact that in our cells IGF-I-induced PKCζ activation is wortmannin/LY294002-sensitive indicate that this PKC isoenzyme could be involved in the protein network downstream from PI 3-kinase, which leads or maintains the onset of brown adipocyte differentiation.

In addition to its role in inducing differentiation, IGF-I is a complete mitogen in fetal brown adipocyte primary cultures by inducing DNA synthesis, cell number increase, the entry of cells into the cell cycle, and PCNA expression (15, 17, 21). Finally, we have demonstrated that IGF-I-induced brown adipocyte proliferation is not inhibited by the presence of wortmannin/LY294002 in the medium. The exact role of PI 3-kinase in regulating cell proliferation has been the subject of controversy. It has been reported that microinjection of neutralizing antibodies against PI 3-kinase blocks the ability of a number of growth factors to induce DNA synthesis in fibroblasts (49). In 3T3L1 adipocytes, PI 3-kinase inhibits GTPase-activating protein, allowing insulin to fully activate p21-ras (50). On the other hand, experiments using wortmannin in Chinese hamster ovary (CHO) cells indicated that PI 3-kinase activity is not required for ras activation (45). In our fetal cells IGF-I-induced mitogenesis, which has been shown to be p21 ras-dependent (20, 21), seems to be PI 3-kinase-independent. Interestingly, it has been reported that SHC (SRC homology domain and collagen-like) is the predominant signaling molecule that activates ras in the insulin-signaling cascade (51, 52). SHC is tyrosine phosphorylated following IGF-I stimulation of brown adipocytes (20). Thus, the exact contribution of IRS-1/IRS-2, SHC, and perhaps other docking proteins to IGF-I-induced proliferation and/or differentiation remains to be established.

In conclusion, our results indicate that PI 3-kinase is a requirement for the IGF-I-induced adipogenic and thermogenic differentiation signaling pathways in fetal brown adipocytes, partly diverging through p70^S6k. However, the IGF-I-induced mitogenesis-signaling pathway is PI 3-kinase-independent. In addition, PKCζ seems to be a signaling molecule also involved in the IGF-I-induced differentiation pathways downstream from PI 3-kinase.

MATERIALS AND METHODS

Materials

FCS and culture media were from Imperial Laboratories (Hampshire, UK). IGF-I, LY294002, and rapamycin were purchased from Calbiochem (Calbiochem-Novabiochem Intl, La Jolla, CA). Wortmannin and anti-mouse IgG-agarose were from Sigma Chemical Co. (St. Louis, MO). Protein A-agarose was from Boehringer Mannheim (Mannheim, Germany). The Py72 anti-Tyr(P) and the anti-αp85 subunit of PI 3-kinase mouse monoclonal antibodies were the generous gifts of Dr. E. Rozengurt and J. Sinnet-Smith and Drs. J. Downward and P. Rodriguez-Viciana (Imperial Cancer Research Fundation, London), respectively. For IRS-1 immunoprecipitations, a rabbit polyclonal antibody was the generous gift of Dr. R. Kahn (Joslin Diabetes Center, Boston, MA). The 4G10 anti-Tyr(P) monoclonal antibody was purchased from Upstate Biotechnology (Lake Placid, NY). The anti-PCNA mouse monoclonal antibody was purchased from Boehringer. The anti-ME polyclonal antibody was obtained as previously described (16). For PKCζ immunoprecipitations the rabbit polyclonal antiserum used was a gift of Dr. J. Moscat (Centro de Biologia Molecular, Madrid).[γ ³²P]ATP (3000 Ci/mmol), [³²P]dCTP (3000 Ci/mmol), 2-deoxy-d[1-³H]glucose (11.0 Ci/mmol), and [³H]thymidine (0.2 mCi/ml) were purchased from Amersham (Aylesbury, UK). All other reagents were of the purest grade available.

Cell Culture

Brown adipocytes were obtained from interscapular brown adipose tissue of 20-day Wistar rat fetuses and isolated by collagenase dispersion as previously described (17, 23). Cells were plated at 5 × 10⁶ cells/100 mm or 1–1.2 × 10⁶ cells/60-mm tissue culture plates in MEM supplemented with 10% FCS to allow cell attachment to the plastic surface of the plates. After 4–6 h of culture at 37 C, cells were rinsed twice with PBS, and 80% of the initial cells were attached. Cells were maintained for 20 h in a serum-free medium supplemented with 0.2% (wt/vol) BSA. At this time, cells were treated for 5 min with IGF-I (10 nm) or preincubated for 15 min with several doses of PI 3-kinase inhibitors (wortmannin or LY294002) and subsequently stimulated with IGF-I for a further 5 min. Both inhibitors were initially disolved in dimethyl sulfoxide and in all experimental series control cells were treated with the corresponding volumes of dimethyl sulfoxide.

To analyze the effect of PI 3-kinase inhibitors on IGF-I-induced brown adipocyte differentiation, 20 h serum-deprived cells were cultured for a further 24 h in the presence of IGF-I either in the absence or presence of wortmannin or LY294002 at the doses indicated in Results and in the figure legends. Due to the instability of wortmannin in aqueous solutions, the medium from cells treated with wortmannin was replaced every 6 h.

Immunoprecipitations

Quiescent fetal brown adipocytes (5 × 10⁶ cells/100-mm tissue culture dish) were treated with IGF-I for 5 min or preincubated for 15 min with wortmannin and LY294002 and subsequently stimulated with IGF-I for a further 5 min as indicated, and lysed at 4°C in 1 ml of a solution containing 10 mm Tris-HCl, 5 mm EDTA, 50 mm NaCl, 30 mm sodium pyrophosphate, 50 mm NaF, 100 μm Na₃VO₄, 1% Triton X-100, and 1 mm phenylmethylsulfonyl fluoride, pH 7.6 (lysis buffer). Lysates were clarified by centrifugation at 15,000 × g for 10 min, and the supernatants were transferred to a fresh tube. After protein content determination, equal amounts of protein were immunoprecipitated at 4°C either with the monoclonal antibodies anti-Tyr(P) (Py72) andα p85, or with a polyclonal antibody against IRS-I. The immune complexes were collected on anti-mouse IgG-agarose beads or, in the case of the IRS-I antibody, on Protein A-agarose beads. Immunoprecipitates were washed three times with lysis buffer and extracted for 10 min at 95 C in 2 × SDS-PAGE sample buffer (200 mm Tris-HCl, 6% SDS, 2 mm EDTA, 4% 2-mercaptoethanol, 10% glycerol, pH 6.8) and analyzed by SDS-PAGE and as described in Results and in the figure legends.

Western Blotting

After SDS-PAGE, proteins were transferred to Immobilon membranes and were blocked using 5% nonfat dried milk in 10 mm Tris-HCl and 150 mm NaCl, pH 7.5, and incubated overnight with several antibodies as indicated in 0.05% Tween-20, 1% non-fat dried milk in 10 mm Tris-HCl, and 150 mm NaCl, pH 7.5. Immunoreactive bands were visualized using the enhanced chemiluminescence (ECL) Western blotting protocol (Amersham).

In Vitro Kinase Assay

The protein kinase activity of the immunoprecipitates was measured as described (53). The immune complexes were incubated in 20μ l of buffer containing 20 mm HEPES, 3 mm MnCl₂, 10 mm MgCl₂, and 20 μCi of[γ ³²P]ATP (in a final concentration of 5μ m) for 15 min at room temperature. The complexes were washed twice with cold PBS and then resuspended in 2 × SDS-PAGE sample buffer and analyzed by SDS-PAGE. The separated proteins were dried in the gel, and the incorporation of [³²P]phosphate into protein was visualized by autoradiography and quantitated by scanning laser densitometry (Molecular Dynamics densitometer, Sunnyvale, CA).

PI 3-Kinase Activity

PI 3-Kinase activity was measured by in vitro phosphorylation of phosphatidylinositol as described (54). Fetal brown adipocytes were incubated with IGF-I in the absence or presence of PI 3-kinase inhibitors as indicated in the figure legends. After washing with ice-cold PBS, cells were solubilized in lysis buffer containing leupeptin (10 μg/ml), aprotinin (10 μg/ml), and 1 mm phenylmethylsulfonyl fluoride. Lysates were clarified by centrifugation at 15,000 × g for 10 min at 4 C, and proteins were immunoprecipitated with the anti-IRS-1 polyclonal antibody. The immunoprecipitates were washed successively in PBS containing 1% Triton X-100 and 100 μm Na₃VO₄ (twice), 100 mm Tris (pH 7.5) containing 0.5 m LiCl, 1 mm EDTA and 100 μm Na₃VO₄ (two times), and 25 mm Tris (pH 7.5) containing 100 mm NaCl and 1 mm EDTA (twice). To each pellet were added 25 μl of 1 mg/ml l-α-phosphatidylinositol/l-α-phosphatidyl-l-serine sonicated in 25 mm HEPES (pH 7.5) and 1 mm EDTA.

The PI 3-kinase reaction was started by the addition of 100 nm [γ³²P]ATP (10 μCi) and 300μ m ATP in 25 μl of 25 mm HEPES, pH 7.4, 10 mm MgCl₂, and 0.5 mm EGTA. After 15 min at room temperature, the reaction was stopped by the addition of 500 μl CHCl₃-methanol (1:2) in a 1% concentration of HCl plus 125 μl chloroform and 125 μl HCl (10 mm). The samples were centrifuged, and the lower organic phase was removed and washed once with 480 μl methanol-100 mm HCl plus 2 mm EDTA (1:1). The organic phase was extracted, dried in vacuo, and resuspended in chloroform. Samples were applied to a a silica gel TLC plate (Whatman, Clifton, NJ). TLC plates were developed in propanol-1-acetic acid (2 n; 65:35 vol/vol), dried, visualized by autoradiography, and quantitated by scanning laser densitometry (Molecular Dynamics personal densitometer).

PKCζ Activity

Fetal brown adipocytes either untreated or stimulated were extracted with lysis buffer (50 mm Tris, pH 7.5, 150 mm NaCl, 1% Triton X-100, 2 mm EDTA, 1 mm EGTA, 1 mm phenylmethylsulfonyl fluoride, 25 μg/ml leupeptin, and 25 μg/ml aprotinin) and immunoprecipitated with an anti-PKCζ antibody as previously described (20). Immune complexes were washed seven times with ice-cold lysis buffer with 0.5 m NaCl and twice with kinase buffer (35 mm Tris, pH 7.5, 10 mm MgCl₂, 0.5 mm EGTA, and 1 mm Na₃VO₄). The kinase reaction was performed in 20 μl kinase buffer containing 1 μCi [γ³²P]ATP, 60μ m ATP, and 1 μg MBP as a substrate for 30 min at 30 C and was terminated by the addition of 4 × SDS-PAGE sample buffer followed by boiling for 5 min at 95 C. Samples were resolved in 12% SDS-PAGE, and gels were dried out and subjected to autoradiography.

Measurement of the 2-Deoxyglucose

Transport 2-Deoxyglucose transport was measured as described (55). After culture, quiescent brown adipocytes (1–1.2 × 10⁶ cells/60-mm plate) were washed three times with Krebs-Ringer-phosphate buffer (KRP) containing 135 mm NaCl, 5.4 mm KCl, 1.4 mm CaCl₂, 1.4 mm MgSO₄, and 10 mm sodium pyrophosphate, pH 7.4, and then incubated with 1 ml KRP buffer with IGF-I for 10 min at 37 C or preincubated with PI 3-kinase inhibitors for 15 min and subsequently stimulated with IGF-I for another 10 min. 2-Deoxy-d[1-³H]glucose was added to this solution to a final concentration of 0.1 mm and 250 nCi/ml, and the incubation was continued for 5 min at 37 C. The cells were then washed three times with ice-cold KRP buffer and solubilized in 1 ml 1% SDS. The radioactivity of a 200-μl aliquot was determined in a scintillation counter.

Determination of ME Activity

At the end of the culture period, ME activity was measured in the cytosolic supernatants as previosly described (16). Enzyme activity was expressed as milliunits/mg of protein. A milliunit is nanomoles NADPH formed/min.

Flow Cytometric Analysis of Nile Red Fluorescence

Cytoplasmatic lipid content was determined by Nile Red fluorescence emission 530 (BP 530/30 nm) in a FACScan flow cytometer (Becton-Dickinson, San Jose, CA). Cells were detached from dishes by addition of 0.05% trypsin-0.02% EDTA, and lipid content was determined in aliquots of 2 × 10⁵ cells after the addition of Nile Red (0.1 μg/ml) (56). Results represent mean intensities of fluorescence (obtained from the histograms of numbers of cells vs. intensity of fluorescence) and are expressed in arbitrary units.

RNA Extraction and Analysis

At the end of the culture time, cells were washed twice with ice-cold PBS, and RNA was isolated with RNazol B (Biotecx Lab, Dallas, TX) following the protocol supplied by the manufacturer for total RNA isolation (57). Total cellular RNA (10 μg) was submitted to Northern blot analysis, i.e . electrophoresed on 0.9% agarose gels containing 0.66 m formaldehyde, transferred to GeneScreen (NEN Research Products, Boston, MA) membranes using a VacuGene blotting apparatus (LKB-Pharmacia, Upsala, Sweden). Hybridization was in 0.25 mm NaHPO₄, pH 7.2, 0.25 m NaCl, 100μ g/ml denatured salmon sperm DNA, 7% SDS, and 50% deionized formamide, containing denatured ³²P-labeled cDNA (10⁶ cpm/ml) for 24 h at 42 C. Complementary DNA labeling was carried out with [³²P]dCTP to a specific activity of 10⁹ cpm/μg of DNA by using multiprimer DNA-labeling system kit. For serial hybridization with different probes, the blots were stripped and rehybridized subsequently as needed in each case. The cDNAs used as probes were FAS (58), ME (59), G3PD (60), ACC (61), UP (62), and β-actin (63). Membranes were subjected to autoradiography, and relative densities of the hybridization signals were determined by densitometric scanning of the autoradiograms.

Determination of [³H]Thymidine Incorporation into Acid-Insoluble Material

DNA synthesis was determined after 24 h of cell culture in the presence of IGF-I (1.4 nm) in the absence or presence of PI 3-kinase inhibitors by [³H]thymidine incorporation (0.2 mCi/ml) into acid-insoluble material during the last 4 h of culture (15). Results are expressed as disintegrations per min/dish.

Cell Cycle Analysis by Flow Cytometry

After culture of cells for 24 h in the presence of IGF-I without or with PI 3-kinase inhibitors, cells were detached from plates by addition of 0.05% trypsin-0.02% EDTA. Trypsinization was stopped by addition of 10% FCS to the culture medium. The percentages of cells in G0/G1 and in S+G2+M phases of the cell cycle were determined after nuclei were stained with propidium iodine by using the Cycle test DNA reagent kit (Becton-Dickinson, San Jose, CA), measured in a Double Discriminator Module and computer analyzed. All measurements were performed in a FACScan flow cytometer (Becton-Dickinson).

Protein Determination

Protein determination was performed by the Bradford dye method (64), using the Bio-Rad reagent (Bio-Rad, Richmond, CA) and BSA as the standard.

Experimental Animals

The animals used for the required experiments in this report were treated in accord with the “Guidelines for Care and Use of Experimental Animals.”

Acknowledgments

We are grateful for valuable reagents provided by Drs. E. Rozengurt, J. Downward, P. Rodriguez-Viciana, J. Sinnet-Smith (Imperial Cancer Research Foundation, London); Dr. R. Kahn (Joslin Diabetes Center, Boston); and Dr. J. Moscat (Centro de Biologia Molecular, Madrid). We thank Dr. A. Alvarez for his expert technical assistance with the flow cytometer.

This work was supported by a SAF 96/0115 Grant from the Comision Interministerial de Ciencia y Tecnologia, Spain. P. Navarro was a recipient of a fellowship from the Ministerio de Educacion y Ciencia, Spain.