Differentiated 3T3L1 adipocytes are composed of heterogenous cell populations with distinct receptor tyrosine kinase signaling properties

Diverse repertoire of human adipocyte subtypes develops from transcriptionally distinct mesenchymal progenitor cells

Single-cell sequencing technologies have revealed an unexpectedly broad repertoire of cells required to mediate complex functions in multicellular organisms. Despite the multiple roles of adipose tissue in maintaining systemic metabolic homeostasis, adipocytes are thought to be largely homogenous with only 2 major subtypes recognized in humans so far. Here we report the existence and characteristics of 4 distinct human adipocyte subtypes, and of their respective mesenchymal progenitors. The phenotypes of these distinct adipocyte subtypes are differentially associated with key adipose tissue functions, including thermogenesis, lipid storage, and adipokine secretion. The transcriptomic signature of "brite/beige" thermogenic adipocytes reveals mechanisms for iron accumulation and protection from oxidative stress, necessary for mitochondrial biogenesis and respiration upon activation. Importantly, this signature is enriched in human supraclavicular adipose tissue, confirming that these cells comprise thermogenic depots in vivo, and explain previous findings of a rate-limiting role of iron in adipose tissue browning. The mesenchymal progenitors that give rise to beige/brite adipocytes express a unique set of cytokines and transcriptional regulators involved in immune cell modulation of adipose tissue browning. Unexpectedly, we also find adipocyte subtypes specialized for high-level expression of the adipokines adiponectin or leptin, associated with distinct transcription factors previously implicated in adipocyte differentiation. The finding of a broad adipocyte repertoire derived from a distinct set of mesenchymal progenitors, and of the transcriptional regulators that can control their development, provides a framework for understanding human adipose tissue function and role in metabolic disease.

Subcutaneous adipose tissue imaging of human obesity reveals two types of adipocyte membranes: Insulin-responsive and -nonresponsive

In adipose tissue, resistance to insulin's ability to increase glucose uptake can be induced by multiple factors, including obesity. Impaired insulin action may take place at different spatial loci at the cellular or subcellular level. To begin to understand the spatial response to insulin in human subcutaneous adipose tissue (hSAT), we developed a quantitative imaging method for activation of a major signaling node in the glucoregulatory insulin signaling pathway. After treatment with insulin or control media, biopsied tissues were immunostained for Akt phosphorylation at Thr-308/9 (pAkt) and then imaged by confocal fluorescence microscopy automated to collect a large grid of high resolution fields. In hSAT from 40 men and women with obesity, substantial heterogeneity of pAkt densities in adipocyte membranes were quantified in each image mosaic using a spatial unit of at least twice the size of the point spread function. Statistical analysis of the distribution of pAkt spatial units was best fit as the weighted sum of two separate distributions, corresponding to either a low or high pAkt density. A "high pAkt fraction" metric was calculated from the fraction of high pAkt distributed units over the total units. Importantly, upon insulin stimulation, tissues from the same biopsy showed either a minimal or a substantial change in the high pAkt fraction. Further supporting a two-state response to insulin stimulation, subjects with similar insulin sensitivity indices are also segregated into either of two clusters identified by the amount of membrane-localized pAkt.

Cell-autonomous heterogeneity of nutrient uptake in white adipose tissue of rhesus macaques

Phenotypic diversity may play an adaptive role by providing graded biological responses to fluctuations in environmental stimuli. We used single-cell imaging of the metabolizable fluorescent fatty acid analog 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY)-C12 and fluorescent 2-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino]-2-deoxy-D-glucose (2-NBDG) to explore cellular heterogeneity in nutrient uptake in white adipose tissue (WAT) explants of rhesus macaques. Surprisingly, WAT displayed a striking cell size-independent mosaic pattern, in that adjacent adipocytes varied with respect to insulin-stimulated BODIPY-C12 and 2-NBDG uptake. Relative free fatty acid (FFA) transport activity correlated with the cellular levels of FFA transporter protein-1 and the scavenger receptor CD36 in individual adipocytes. In vitro incubation of WAT explants for 24 hours caused partial desynchronization of cellular responses, suggesting that adipocytes may slowly alter their differential nutrient uptake activity. In vitro-differentiated human adipocytes also exhibited a mosaic pattern of BODIPY-C12 uptake. WAT from animals containing a homogeneous population of large adipocytes was nonmosaic, in that every adipocyte exhibited a similar level of BODIPY-C12 fluorescence, suggesting that the development of obesity is associated with the loss of heterogeneity in WAT. Hence, for the first time, we demonstrate an intrinsic heterogeneity in FFA and glucose transport activity in WAT.

A novel RNAi lethality rescue screen to identify regulators of adipogenesis

Adipogenesis, the differentiation of fibroblast-like mesenchymal stem cells into mature adipocytes, is tightly regulated by a complex cascade of transcription factors, including the nuclear receptor Peroxisome proliferator activator receptor γ (PPARγ). RNAi-mediated knock down libraries may present an atractive method for the identification of additional adipogenic factors. However, using in vitro adipogenesis model systems for high-throughput screening with siRNA libraries is limited since (i) differentiation is not homogeneous, but results in mixed cell populations, and (ii) the expression levels (and activity) of adipogenic regulators is highly dynamic during differentiation, indicating that the timing of RNAi-mediated knock down during differentiation may be extremely critical. Here we report a proof-of-principle for a novel RNAi screening method to identify regulators of adipogenesis that is based on lethality rescue rather than differentiation, using microRNA expression driven by a PPARγ responsive RNA polymerase II promoter. We validated this novel method through screening of a dedicated deubiquitinase knock down library, resulting in the identification of UCHL3 as an essential deubiquitinase in adipogenesis. This system therefore enables the identification of novel genes regulating PPARγ-mediated adipogenesis in a high-throughput setting.

Heterogeneity in the physiological states and pharmacological responses of differentiating 3T3-L1 preadipocytes

A systems biology–based analysis shows that differentiating adipocytes look very different at the single-cell level and form distinct cellular subpopulations.

Increases in key components of adipogenesis and lipolysis pathways correlate at the population-averaged level during adipogenesis. However, differentiating preadipocytes are highly heterogeneous in cellular and lipid droplet (LD) morphologies, and the degree to which individual cells follow population-averaged trends is unclear. In this study, we analyze the molecular heterogeneity of differentiating 3T3-L1 preadipocytes using immunofluorescence microscopy. Unexpectedly, we only observe a small percentage of cells with high simultaneous expression of markers for adipogenesis (peroxisome proliferator-activated receptor γ [PPARγ], CCAAT/enhancer-binding protein α, and adiponectin) and lipid accumulation (hormone-sensitive lipase, perilipin A, and LDs). Instead, we identify subpopulations of cells with negatively correlated expressions of these readouts. Acute perturbation of adipocyte differentiation with PPARγ agonists, forskolin, and fatty acids induced subpopulation-specific effects, including redistribution of the percentage of cells in observed subpopulations and differential expression levels of PPARγ. Collectively, our results suggested that heterogeneity observed during 3T3-L1 adipogenesis reflects a dynamic mixture of subpopulations with distinct physiological states.

Single-cell profiling reveals the origin of phenotypic variability in adipogenesis

Phenotypic heterogeneity in a clonal cell population is a well-observed but poorly understood phenomenon. Here, a single-cell approach is employed to investigate non-mutative causes of phenotypic heterogeneity during the differentiation of 3T3-L1 cells into fat cells. Using coherent anti-Stokes Raman scattering microscopy and flow cytometry, adipogenic gene expression, insulin signaling, and glucose import are visualized simultaneously with lipid droplet accumulation in single cells. Expression of adipogenic genes PPARγ, C/EBPα, aP2, LP2 suggests a commitment to fat cell differentiation in all cells. However, the lack of lipid droplet in many differentiating cells suggests adipogenic gene expression is insufficient for lipid droplet formation. Instead, cell-to-cell variability in lipid droplet formation is dependent on the cascade responses of an insulin signaling pathway which includes insulin sensitivity, kinase activity, glucose import, expression of an insulin degradation enzyme, and insulin degradation rate. Increased and prolonged insulin stimulation promotes lipid droplet accumulation in all differentiating cells. Single-cell profiling reveals the kinetics of an insulin signaling cascade as the origin of phenotypic variability in drug-inducible adipogenesis.

The Rab GTPase-activating protein AS160 as a common regulator of insulin- and Galphaq-mediated intracellular GLUT4 vesicle distribution

Akt substrate of 160kDa (AS160) is a Rab GTPase activating protein (GAP) and was recently identified as a component of the insulin signaling pathway of glucose transporter type 4 (GLUT4) translocation. We and others, previously reported that the activation of Galphaq protein-coupled receptors (GalphaqPCRs) also stimulated GLUT4 translocation and glucose uptake in several cell lines. Here, we report that the activation of GalphaqPCRs also promoted phosphorylation of AS160 by the 5'-AMP activated protein kinase (AMPK). The suppression of AS160 phosphorylation by the siRNA mediated AMPKalpha1 subunit knockdown promoted GLUT4 vesicle retention in intracellular compartments. This suppression did not affect the ratio of non-induced cell surface GLUT4 to Galphaq-induced it. Rat 3Y1 cells lacking AS160 did not show insulin-induced GLUT4 translocation. The cells stably expressing GLUT4 revealed GLUT4 vesicles that were mainly localized in the perinuclear region and less frequently on the cell surface. After expression of exogenous AS160, GLUT4 on the cell surface decreased and GLUT4 vesicles were redistributed throughout the cytoplasm. Although PMA-induced or sodium fluoride-induced GLUT4 translocation was significantly increased in these cells, insulin did not affect GLUT4 translocation. These results suggest that AS160 is a common regulator of insulin- and GalphaqPCR activation-mediated GLUT4 distribution in the cells.

Tyrosine phosphorylation of Munc18c regulates platelet-derived growth factor-stimulated glucose transporter 4 translocation in 3T3L1 adipocytes

Platelet-derived growth factor (PDGF) stimulation of skeletal muscle, cultured myotubes, and 3T3L1 adipocytes results in glucose transporter 4 (Glut4) translocation, albeit to a reduced level compared with insulin. To address the mechanism of PDGF action, we have determined that the Syntaxin 4 negative regulatory protein, Munc18c, undergoes PDGF-stimulated phosphorylation on tyrosine residue 521. The tyrosine phosphorylation of Munc18c on Y521 occurred concomitant with the dissociation of the Munc18c protein from Syntaxin 4 in a time frame consistent with Glut4 translocation. Moreover, expression of the wild-type Munc18c protein did not inhibit PDGF-induced Glut4 translocation, whereas expression of Y521A-Munc18c mutant was inhibitory and failed to dissociate from Syntaxin 4. In contrast, expression of either wild-type Munc18c or the Y521A-Munc18c mutant both resulted in a marked inhibition of insulin-stimulated Glut4 translocation. Together, these data demonstrate that one mechanism accounting for the PDGF induction of Glut4 translocation is the suppression of the Munc18c negative regulation of Syntaxin 4 function.

Disruption of microtubules ablates the specificity of insulin signaling to GLUT4 translocation in 3T3-L1 adipocytes

Although the cytoskeletal network is important for insulin-induced glucose uptake, several studies have assessed the effects of microtubule disruption on glucose transport with divergent results. Here, we investigated the effects of microtubule-depolymerizing reagent, nocodazole and colchicine, on GLUT4 translocation in 3T3-L1 adipocytes. After nocodazole treatment to disrupt microtubules, GLUT4 vesicles were dispersed from the perinuclear region in the basal state, and insulin-induced GLUT4 translocation was partially inhibited by 20-30%, consistent with other reports. We found that platelet-derived growth factor (PDGF), which did not stimulate GLUT4 translocation in intact cells, was surprisingly able to enhance GLUT4 translocation to approximately 50% of the maximal insulin response, in nocodazole-treated cells with disrupted microtubules. This effect of PDGF was blocked by pretreatment with wortmannin and attenuated in cells pretreated with cytochalasin D. Using confocal microscopy, we found an increased co-localization of GLUT4 and F-actin in nocodazole-treated cells upon PDGF stimulation compared with control cells. Furthermore, microinjection of small interfering RNA targeting the actin-based motor Myo1c, but not the microtubule-based motor KIF3, significantly inhibited both insulin- and PDGF-stimulated GLUT4 translocation after nocodazole treatment. In summary, our data suggest that 1) proper perinuclear localization of GLUT4 vesicles is a requirement for insulin-specific stimulation of GLUT4 translocation, and 2) nocodazole treatment disperses GLUT4 vesicles from the perinuclear region allowing them to engage insulin and PDGF-sensitive actin filaments, which can participate in GLUT4 translocation in a phosphatidylinositol 3-kinase-dependent manner.

Platelet-derived growth factor stimulates glucose transport in skeletal muscles of transgenic mice specifically expressing platelet-derived growth factor receptor in the muscle, but it does not affect blood glucose levels

Insulin stimulates the disposal of blood glucose into skeletal muscle and adipose tissues by the translocation of GLUT4 from intracellular pools to the plasma membrane, and consequently the concentration of blood glucose levels decreases rapidly in vivo. Phosphatidylinositol (PI) 3-kinase and Akt play a pivotal role in the stimulation of glucose transport by insulin, but detailed mechanisms are unknown. We and others reported that not only insulin but also platelet-derived growth factor (PDGF) and epidermal growth factor facilitate glucose uptake through GLUT4 translocation by activation of PI 3-kinase and Akt in cultured cells. However, opposite results were also reported. We generated transgenic mice that specifically express the PDGF receptor in skeletal muscle. In these mice, PDGF stimulated glucose transport into skeletal muscle in vitro and in vivo. Thus, PDGF apparently shares with insulin some of the signaling molecules needed for the stimulation of glucose transport. The degree of glucose uptake in vivo reached approximately 60% of that by insulin injection in skeletal muscle, but blood glucose levels were not decreased by PDGF in these mice. Therefore, PDGF-induced disposal of blood glucose into skeletal muscle is insufficient for rapid decrease of blood glucose levels.

Insulin but not PDGF relies on actin remodeling and on VAMP2 for GLUT4 translocation in myoblasts

Insulin promotes the translocation of glucose transporter 4 (GLUT4) from intracellular pools to the surface of muscle and fat cells via a mechanism dependent on phosphatidylinositol (PtdIns) 3-kinase, actin cytoskeletal remodeling and the v-SNARE VAMP2. The growth factor PDGF-BB also robustly activates PtdIns 3-kinase and induces actin remodeling, raising the question of whether it uses similar mechanisms to insulin in mobilizing GLUT4. In L6 myoblasts stably expressing Myc-tagged GLUT4, neither stimulus affected the rate of GLUT4 endocytosis, confirming that they act primarily by enhancing exocytosis to increase GLUT4 at the cell surface. Although surface GLUT4myc in response to insulin peaked at 10 minutes and remained steady for 30 minutes, PDGF action was transient, peaking at 5 minutes and disappearing by 20 minutes. These GLUT4myc translocation time courses mirrored that of phosphorylation of Akt by the two stimuli. Interestingly, insulin and PDGF caused distinct manifestations of actin remodeling. Insulin induced discrete, long (>5 μm) dorsal actin structures at the cell periphery, whereas PDGF induced multiple short (<5 μm) dorsal structures throughout the cell, including above the nucleus. Latrunculin B, cytochalasin D and jasplakinolide, which disrupt actin dynamics, prevented insulin- and PDGF-induced actin remodeling but significantly inhibited GLUT4myc translocation only in response to insulin (75-85%, P<0.05), not to PDGF (20-30% inhibition). Moreover, transfection of tetanus toxin light chain, which cleaves the v-SNAREs VAMP2 and VAMP3, reduced insulin-induced GLUT4myc translocation by >70% but did not affect the PDGF response. These results suggest that insulin and PDGF rely differently on the actin cytoskeleton and on tetanus-toxin-sensitive VAMPs for mobilizing GLUT4.

Expression of a mutant IRS inhibits metabolic and mitogenic signalling of insulin in human adipocytes

Adipose tissue is a primary target of insulin, but knowledge about insulin signalling in human adipocytes is limited. We developed an electroporation technique for transfection of primary human adipocytes with a transfection efficiency of 15% +/- 5 (mean +/- S.D.). Human adipocytes were co-transfected with a mutant of IRS-3 (all four potential PI3-kinase binding motifs mutated: IRS-3F4) and HA-tagged protein kinase B (HA-PKB/Akt). HA-PKB/Akt was immunoprecipitated from cell lysates with anti-HA antibodies, resolved with SDS-PAGE, and immunoblotted with phospho-specific antibodies. We found that IRS-3F4 blocked insulin stimulation of HA-PKB/Akt phosphorylation and in further analyses also translocation of recombinant HA-tagged glucose transporter to the plasma membrane. IRS-3F4 also blocked insulin-induced activation of the transcription factor Elk-1. Our results demonstrate the critical importance of IRS for metabolic as well as mitogenic signalling by insulin. This method for transfection of primary human adipocytes will be useful for studying insulin signalling in human adipocytes with molecular biological techniques.

Nelfinavir-induced insulin resistance is associated with impaired plasma membrane recruitment of the PI 3-kinase effectors Akt/PKB and PKC-zeta

AIMS/HYPOTHESIS

Chronic exposure of 3T3-L1 adipocytes to the HIV protease inhibitor nelfinavir induces insulin resistance, recapitulating key metabolic alterations of adipose tissue in the lipodystrophy syndrome induced by these agents. Our goal was to identify the defect in the insulin signal transduction cascade leading to nelfinavir-induced insulin resistance.

METHODS

Fully differentiated 3T3-L1 adipocytes were exposed to 30 micro mol/l nelfinavir for 18 h, after which the amount, the phosphorylation and the localisation of key proteins in the insulin signalling cascade were evaluated.

RESULTS

Insulin-induced interaction of phosphatidylinositol 3'-kinase (PI 3-kinase) with IRS proteins was normal in cells treated with nelfinavir, as was IRS-1-associated PI 3-kinase activity. Yet insulin-induced phosphorylation of Akt/protein kinase B (PKB), p70S6 kinase and extracellular signal-regulated kinase 1/2 was significantly impaired. This could not be attributed to increased protein phosphatase 2A activity or to increased expression of phosphoinositide phosphatases (SHIP2 or PTEN). However, insulin failed to induce translocation of the PI 3-kinase effectors Akt/PKB and protein kinase C-zeta (PKC-zeta) to plasma membrane fractions of nelfinavir-treated adipocytes.

CONCLUSIONS/INTERPRETATION

We therefore conclude that nelfinavir induces a defect in the insulin signalling cascade downstream of the activation of PI 3-kinase. This defect manifests itself by impaired insulin-mediated recruitment of Akt/PKB and PKC-zeta to the plasma membrane.

Simple flow cytometric method used to assess lipid accumulation in fat cells

Adipogenesis of preadipocytes in culture has been frequently used to study the molecular basis and effect of drugs on fat cell conversion. However, after adipogenic induction, cells respond to the inducing agent with various speeds of conversion and fat accumulation, which complicates direct molecular and biochemical analyses. Here we present a simple and sensitive method to detect and quantify fat accumulation inside cells by flow cytometry. Using this method, we detected elevated levels of cytoplasmic granularity that correlated well with an increased level of fat accumulated inside cells after adipogenic conversion. We further demonstrated the ability of this method to monitor and quantify fat cell maturation within a complex population of cells and to identify and collect the fat cells with similar fat storage for further analysis. Flow cytometry offers distinct advantages over existing detection systems for cytoplasmic lipid staining and lipid extraction and could represent a powerful analytical tool to monitor the effect of chemicals and biological molecules on fat cell conversion and maturation. Moreover, in combination with a cell sorting facility, our method offers a simple and efficient means of collecting fat cells of specific status for further analysis.

Platelet-derived growth factor (PDGF) stimulates glucose transport in 3T3-L1 adipocytes overexpressing PDGF receptor by a pathway independent of insulin receptor substrates

Insulin is unique among growth factors and hormones in its ability to control metabolic functions such as the stimulation of glucose uptake and glucose transporter (GLUT4) translocation in physiological target tissues, such as muscle and adipose cells. Nonetheless, the mechanisms underlying this specificity have remained incompletely understood, particularly in view of the ability of some growth factors to mimic insulin-dependent early signaling events. In this study, we have probed the basis of insulin specificity by overexpressing in hormone-responsive 3T3-L1 adipocytes wild-type platelet-derived growth factor (PDGF) receptor (PDGFR)-beta and selected, informative mutant receptor proteins. We show that such adipocytes overexpressing wild-type PDGFR on exposure to cognate growth factor activate glucose transport, GLUT4 translocation, and the serine-threonine protein kinase Akt/protein kinase B to a degree comparable with that produced in response to insulin. In addition, PDGF elicits the robust generation of phosphatidylinositol-3,4,5-trisphosphate in vivo in PDGFR-overexpressing 3T3-L1 adipocytes. Expression of PDGFR-beta mutant proteins demonstrates that these responses require the presence of an intact phosphatidylinositol 3-kinase (PI3K)-binding site on the overexpressed PDGF receptor. Furthermore, PDGF stimulates these effects independent of insulin receptor substrate(IRS)-1 or IRS-2 tyrosine phosphorylation or docking to activated PI3K. These data demonstrate that 1) the basis of insulin-specific glucose transport in cultured adipocytes is the low level of receptors for other growth factors and 2) in the presence of adequate receptors, PDGF is fully capable of activating glucose transport in a manner requiring PI3K and subsequent phosphatidylinositol-3,4,5-trisphosphate accumulation but independent of insulin, insulin receptor, and IRS proteins.

A PI3-kinase signaling code for insulin-triggered insertion of glucose transporters into the plasma membrane

Activation of phosphatidyl-inositol-3'-OH-kinase (PI3K) and the resulting production of phosphatidyl-inositol-3,4,5-trisphosphate (PIP3) are ubiquitous signaling steps that link various cell surface receptors to multiple intracellular targets. In fat and muscle cells, the same PI3K pathway that regulates metabolic enzymes, proliferation, and differentiation has also been shown to be involved in insulin-triggered insertion of glucose transporter GLUT4 into the plasma membrane. The multiple PI3K functions raise the question of how the same PI3K pathway can be selectively used for different cell functions. Here we developed a dual-color evanescent wave microscopy method to simultaneously measure PIP3 production and GLUT4 insertion in individual 3T3L1 adipocytes. Activation of PI3K was found to be both necessary and sufficient for triggering GLUT4 insertion, but transporter insertion was markedly suppressed for small-amplitude, persistent PIP3 signals and for large-amplitude, short PIP3 signals. The rejection of these common PI3K signaling responses may explain the selective advantage of insulin over platelet-derived growth factor and other stimuli for inducing GLUT4 insertion. Our study suggests that the same PI3K pathway can control specific cell functions by relying on effector systems that respond to particular receptor-encoded time courses and amplitudes of PIP3 signals.

Role of mTOR in the degradation of IRS-1: regulation of PP2A activity

We have investigated the role of PI 3-kinase and mTOR in the degradation of IRS-1 induced by insulin. Inhibition of mTOR with rapamycin resulted in approximately 50% inhibition of the insulin-induced degradation of IRS-1. In contrast, inhibition of PI-3 kinase, an upstream activator of mTOR, leads to a complete block of the insulin-induced degradation. Inhibition of either PI-3 kinase or mTOR prevented the mobility shift in IRS-1 in response to insulin, a shift that is caused by Ser/Thr phosphorylation. These results indicate that insulin stimulates PI 3-kinase-mediated degradation of IRS-1 via both mTOR-dependent and -independent pathways. Platelet-derived growth factor (PDGF) stimulation leads to a lower level of degradation, but significant phosphorylation of IRS-1. Both the degradation and phosphorylation of IRS-1 in response to PDGF are completely inhibited by rapamycin, suggesting that PDGF stimulates IRS-1 degradation principally via the mTOR-dependent pathway. Inhibition of the serine/threonine phosphatase PP2A with okadaic acid also induced the phosphorylation and degradation of IRS-1. IRS-1 phosphorylation and degradation in response to okadaic acid were not inhibited by rapamycin, suggesting that the action of mTOR in the degradation of IRS-1 results from inhibition of PP2A. Consistent with this, treatment of cells with rapamycin stimulated PP2A activity. While the role of mTOR in the phosphorylation of IRS-1 appears to proceed primarily through the regulation of PP2A, we also provide evidence that the regulation of p70S6 kinase phosphorylation requires the direct activity of mTOR.

Lipid rafts and insulin signaling

Lipid rafts are domains within the plasma membrane that are enriched in cholesterol and lipids with saturated acyl chains. Specific proteins, including many signaling proteins, segregate into lipid rafts, and this process is important for certain signal transduction events in a variety of cell types. Within the past decade, data have emerged from many laboratories that implicate lipid rafts as critical for proper compartmentalization of insulin signaling in adipocytes. A subset of lipid rafts, caveolae, are coated with membrane proteins of the caveolin family. Direct interactions between resident raft proteins (caveolins and flotillin-1) and insulin-signaling molecules may organize these molecules in space and time to ensure faithful transduction of the insulin signal, at least with respect to the glucose-dependent actions of insulin in adipocytes. The in vivo relevance of this model remains to be determined.