Article

Nature Chemical Biology 4, 538 - 547 (2008)
Published online: 20 July 2008 | doi:10.1038/nchembio.103

Raft nanodomains contribute to Akt/PKB plasma membrane recruitment and activation

Rémi Lasserre1,2,3,14, Xiao-Jun Guo1,2,3,4,14, Fabien Conchonaud1,2,3,14, Yannick Hamon1,2,3, Omar Hawchar1,2,3, Anne-Marie Bernard1,2,3, Saïdi M'Homa Soudja1,2,3, Pierre-François Lenne5,6, Hervé Rigneault5,6, Daniel Olive7,8,9, Georges Bismuth10,11, Jacques A Nunès7,8,9, Bernard Payrastre12,13, Didier Marguet1,2,3 & Hai-Tao He1,2,3

Membrane rafts are thought to be sphingolipid- and cholesterol-dependent lateral assemblies involved in diverse cellular functions. Their biological roles and even their existence, however, remain controversial. Using an original fluorescence correlation spectroscopy strategy that recently enabled us to identify nanoscale membrane organizations in live cells, we report here that highly dynamic nanodomains exist in both the outer and inner leaflets of the plasma membrane. Through specific inhibition of biosynthesis, we show that sphingolipids and cholesterol are essential and act in concert for formation of nanodomains, thus corroborating their raft nature. Moreover, we find that nanodomains play a crucial role in triggering the phosphatidylinositol-3 kinase/Akt signaling pathway, by facilitating Akt recruitment and activation upon phosphatidylinositol-3,4,5-triphosphate accumulation in the plasma membrane. Thus, through direct monitoring and controlled alterations of rafts in living cells, we demonstrate that rafts are critically involved in the activation of a signaling axis that is essential for cell physiology.

Membrane rafts have been the focus of extensive research during the last few years and are thought to be involved in diverse cellular functions1, 2, 3, 4, 5, 6, particularly signal transduction7. However, the evidence clearly defining their biological roles or even their in vivo existence has remained elusive due to limitations in available experimental technologies3, 8. Nevertheless, a number of studies have produced compelling data indicating the presence of lipid-dependent lateral microheterogeneities in intact cell membranes and thus supporting the raft hypothesis9, 10, 11, 12, 13, 14, 15. Currently, it is thought that native rafts are sphingolipid- and sterol-dependent dynamic membrane assemblies with suboptical sizes (<200 nm), and that protein-lipid and protein-protein interactions participate in their formation. These small rafts sometimes coalesce into larger platforms through protein-based interactions and/or crosslinking3, 4, 6. We recently investigated the plasma membrane lateral organizations in living cells with fluorescence correlation spectroscopy (FCS)16, 17. This powerful technology quantifies diffusion parameters with high temporal resolution and statistical accuracy. In addition, low concentrations of probes and low light excitation make FCS measurements non-invasive. Using an original FCS strategy to determine constrained lateral diffusion, we unambiguously identified lipid-based nanoscaled domains in the plasma membrane of live COS-7 cells at 37 °C into which proteins and lipids dynamically partition or assemble with a timescale of tens to hundreds of milliseconds17.

The phosphoinositide-3 kinase (PI(3)K)/Akt pathway is essential for cellular physiology. It triggers a set of events in response to external cell stimuli, leading to cell growth, cell cycle entry, cell migration and cell survival, with abnormal PI(3)K/Akt signaling frequently causing major pathologies in humans and animals18. The serine/threonine kinase Akt, also known as protein kinase B (PKB), is a central component of this signaling axis, which has a multiple-step activation process19. Upon receptor-mediated activation, PI(3)K generates phosphatidylinositol-3,4,5-triphosphate (PIP3) through phosphorylation of phosphatidylinositol-4,5-bisphosphate (PIP2). The pleckstrin homology (PH) domain of Akt specifically recognizes PIP3, triggering Akt recruitment from the cytosol to the plasma membrane. Once in contact with the membrane, Akt undergoes an activation process involving phosphorylation by phosphoinositide-dependent kinase 1 (PDK1) and PDK2 (refs. 19,20). Akt is then released from the membrane and phosphorylates both cytosolic and nuclear targets. The interaction of Akt with PIP3 is therefore required for activation of Akt by triggering its translocation to the plasma membrane. Both Akt (refs. 21,22) and the PDKs (refs. 23,24,25) have been reported as partitioned into detergent-resistant membrane domains (DRMs). These observations suggest a possible involvement of rafts in PI(3)K/Akt pathway activation.

In this work, we show that highly dynamic nanodomains detected by FCS exist in both the outer and inner leaflets of the plasma membrane of living mouse and human T cells, as well as COS-7 fibroblasts. We demonstrate that sphingolipids and cholesterol (1Compound 1) play essential and concurrent roles in nanodomain formation, a result that fully corroborates the raft nature of nanodomains. Our study goes on to reveal a crucial role for nanodomains in the activation of the PI(3)K/Akt-dependent signaling pathway in different cell types. We show that raft nanodomains are crucial for efficient membrane recruitment and phosphorylation of Akt upon formation of PIP3 in the plasma membrane inner leaflet.

Top

Results

Sensing nanodomains in both plasma membrane leaflets

We first examined the presence of nanodomains in DWT6.11 mouse T cells26 expressing green fluorescent protein (GFP)-Thy-1 or Lck1–12-GFP by performing FCS measurements on various observation scales16, 17 (see Supplementary Methods online for the GFP chimeric constructs). GFP-Thy-1 partitions into nanodomains in COS-7 cells17 and attaches to the plasma membrane outer leaflet via a glycosylphosphatidylinositol (GPI)-anchor, whereas the putatively raft-associated Lck1-12-GFP consists of the GFP fused to the N-terminal 12-amino-acid sequence of Lck, allowing attachment to the plasma membrane inner leaflet via three saturated acyl chains (one myristate (2Compound 2) and two palmitates (3Compound 3)). We used these markers to probe constrained lateral diffusion in the two plasma membrane leaflets, respectively. We found a uniform distribution of GFP chimeras in the plasma membrane with very similar macroscopic diffusion parameters, as assessed by the fluorescence recovery after photobleaching (FRAP) method. The apparent diffusion coefficient (DFRAP) and mobile fraction (Mf) were respectively 0.5 plusminus 0.1 mum2 s-1 and 83 plusminus 5% for GFP-Thy-1, and 0.4 plusminus 0.1 mum2 s-1 and 90 plusminus 6% for Lck1–12-GFP.

For each molecule, we then established the FCS diffusion law at 37 °C by plotting the diffusion time taud (the average time a fluorescent molecule stays within the illuminated area) versus the observation spot area size (Fig. 1a). Our previous studies16, 17 showed that the diffusion law can be expressed in the form of a straight line, the slope of which provides the effective diffusion coefficient Deff, while the intercept extrapolated on the time axis provides the confinement index t0. A positive t0 value reflects confinement by individual domains (or molecular complexes), whereas a negative t0 value reflects confinement by the meshwork organization. Such t0 behavior has been shown to be caused by the significant change in the diffusion rates when the spot length scale goes from smaller to larger than the size of the domain or mesh16, 27; a prominent increase and decrease occurs for individual domains and the meshwork organization, respectively (see Supplementary Discussion online for a more detailed and qualitative explanation of our recent results using the FCS diffusion analysis to probe different types of membrane organizations). In the case of individual domains, depending on whether such domains consist of pre-existing organizations or self-promoting assemblies, t0 connects to either the confining time for the partitioning molecule or lifetime of the assemblies17.

Figure 1: Monitoring of dynamic nanodomains on both leaflets of the plasma membrane of living DWT6.11 T cells.

Figure 1 : Monitoring of dynamic nanodomains on both leaflets of the plasma membrane of living DWT6.11 T cells.

(a) Cells were either transfected with GFP-Thy-1 or Lck1–12-GFP, or loaded with FL-PE. We established FCS diffusion laws as previously described17. Data show mean plusminus s.d. from 8 to 12 independent measurements. (b) Cells transfected with Lck1–12-GFP were either mock-treated (C) or treated by COase, myriocin (M), zaragozic acid (Z) or myriocin and zaragozic acid (MZ). The lateral diffusion for Lck1–12-GFP in each condition was analyzed by FCS as in a to extrapolate the time intercept t0 (error bars give the s.d. of the means)17.

Full size image (24 KB)

In contrast to BODIPY-phosphatidylethanolamine (FL-PE, 4Compound 4), both GFP-Thy-1 and Lck1–12-GFP were found to exhibit constrained lateral diffusion with a t0 of 20.8 plusminus 1.4 ms and 20.1 plusminus 2.9 ms, respectively (Fig. 1a and Table 1) in DWT6.11 cells. These diffusion characteristics indicate the existence of nanodomains (average radius < 60 nm), to which molecules could be dynamically confined on timescales of more than tens of milliseconds16, 17. All together this provides evidence for the presence of highly dynamic nanodomains on the T-cell plasma membrane outer leaflet, confirming our previous observations in COS-7 cells17. More importantly, these results also establish a presence of highly dynamic nanodomains on the inner leaflet of the plasma membrane.


Nanodomains are sphingolipid- and cholesterol-dependent

In agreement with our previous observation of GFP chimeras attaching to the outer leaflet of the plasma membrane via a GPI-anchor in COS-7 cells17, cholesterol oxidase (COase) abolished the lateral confinement of Lck1–12-GFP (Fig. 1b and Table 1), whereas latrunculin A (5Compound 5)-induced depolymerization of the actin cytoskeleton had no effect (Table 1). The sensitivity to COase and sphingomyelinase (SMase) previously allowed us to conclude on a role of cholesterol and sphingomyelin in nanodomain formation, although we did not exclude a contribution of the reaction products to the impediment of nanodomains17. To more directly assess the involvement of sphingolipids and cholesterol in nanodomain formation, we inhibited their synthesis in DWT.6.11 cells using myriocin (6Compound 6) and zaragozic acid (7Compound 7), respectively. Myriocin potently inhibits the serine palmitoyltransferase that catalyzes the first step of sphingolipid metabolism, without accumulation of diverse sphingolipid metabolites28. Zaragozic acid, a specific inhibitor of squalene synthase, selectively blocks the sterol biosynthetic branch of the mevalonate(8Compound 8)/isoprenoid pathway29. At maximum doses that did not affect cell viability, myriocin strongly reduced sphingomyelin (the main cellular sphingolipid) content with no effect on cholesterol, whereas zaragozic acid partially reduced cholesterol content with no effect on sphingomyelin (Table 2). Concomitant treatment of myriocin and zaragozic acid reduced the sphingomyelin and cholesterol levels, respectively, to a similar extent as individual treatment with myriocin or zaragozic acid (Table 2). We found that individual treatments with myriocin and zaragozic acid had, respectively, only partial and no effect on the t0 of Lck1–12-GFP. The combined treatment, however, produced an almost null t0, which indicates an absence of those nanodomains laterally confining Lck1–12-GFP (Fig. 1b and Table 1). Very similar observations were made using Jurkat JA16 cells, a subclone of the Jurkat human leukemia cell line (Tables 1 and 2).


We also examined the effects of inhibiting sphingolipid and/or cholesterol metabolism on lateral diffusion of the outer leaflet–linked GFP-GPI in COS-7 cells. Treatment with myriocin and/or zaragozic acid induced reduction of sphingolipid and/or cholesterol content (Table 2) in a manner analogous to the two lymphoid cell lines examined above. More importantly, in similar fashion these treatments inhibited the lateral confinement of the lipid-anchored GFP probe in the plasma membrane (Table 1). Thus, our data collected from different cell types clearly show that sphingolipids and cholesterol play an essential role in and act concurrently promoting nanodomain formation in both leaflets of the plasma membrane. Furthermore, these results, together with those demonstrating the sensitivity toward cholesterol oxidase and sphingomyelinase, strongly indicate the raft nature of these nanodomains. It should be noted that myriocin and zaragozic acid treatment did not impair other modes of lateral diffusion in the plasma membrane, such as those monitored by FL-PE (Table 1) and the GFP-tagged transferrin receptor (TfR-GFP; Table 1 and Supplementary Data online).

Myriocin and zaragozic acid cause no general lipid changes

To check whether the inhibition of sphingolipid and cholesterol biosynthesis with myriocin and zaragozic acid treatment elicits modifications on other phospholipids, in turn contributing to the observed blockade of nanodomain formation, we performed a lipidomic analysis in Jurkat cells. In these cells, while the predominant forms of fatty acids from Jurkat glycerophospholipids (Supplementary Table 1 online) were either saturated or monounsaturated (42.9% and 43.6%, respectively), about 13.5% were polyunsaturated. Except for sphingomyelin, myriocin and zaragozic acid treatment did not significantly modify the phospholipid composition of Jurkat cells (Supplementary Table 1). Moreover, the same treatment did not cause significant changes in the fatty acid composition of the major glycerophospholipids (Supplementary Table 1). Altogether, these results indicate that, except for sphingolipids, the myriocin and zaragozic acid treatment does not significantly impinge on the phospholipid composition or on the proportion of the major molecular species, thereby substantiating our demonstration of the fundamental role of sphingolipids and cholesterol in nanodomain formation.

Raft nanodomains and PI(3)K/Akt signaling in T cells

Ligation of CD28, a key co-stimulatory receptor for T cells, triggers activation of PI(3)K/Akt and Ras/mitogen-activated protein kinase (MAPK) pathways in DWT6.11 cells26 (as revealed respectively by Akt and extracellular signal-regulated kinase (ERK) phosphorylation; Fig. 2a). We found that the myriocin and zaragozic acid treatment strongly inhibits the CD28-triggered PI(3)K/Akt pathway, but not the Ras/MAPK pathway (Fig. 2a). We confirmed the inhibition of Akt activation by analyzing the phosphorylation of GSK3beta, an endogenous substrate of Akt (Fig. 2a). These data suggest a selective raft involvement in the activation of the PI(3)K/Akt signaling triggered by CD28 stimulation. Such selectivity was further supported by the unaltered overall tyrosine phosphorylation in the CD28-stimulated DWT6.11 T cells upon myriocin and zaragozic acid treatment (Supplementary Fig. 1 online).

Figure 2: Analysis of Akt activation and membrane association in DWT6.11 and Jurkat JA16 cells.

Figure 2 : Analysis of Akt activation and membrane association in DWT6.11 and Jurkat JA16 cells.

Treatments are as follows: mock treatment (C); treatment with zaragozic acid (Z), myriocin (M) or myriocin and zaragozic acid (MZ). (a) Immunoblotting analysis of DWT6.11 cells. Left panel, Akt and ERK phosphorylation after cell stimulation or no stimulation with CD28 monoclonal antibody. The phosphorylation induction upon stimulation for Thr308 and Ser473 of Akt is expressed as a percentage relative to untreated cells: for M, 50.8 plusminus 10.5% and 51.2 plusminus 17.6%; for MZ, 9.2 plusminus 5.9% and 25.5 plusminus 2.2%, respectively. Data are presented as mean plusminus s.d.; n = 3. Upper right panel, induction of GSK3beta phosphorylation upon stimulation by CD28 monoclonal antibody decreased in the MZ-treated cells relative to untreated cells (25.5 plusminus 10.7%, mean plusminus s.d., n = 3). Lower right panel, immunoblotting analysis of the molecular association with the membrane fraction after cell stimulation or no stimulation with CD28 monoclonal antibody. The results are representative of three independent experiments. (b) Immunoblotting analysis of JA16 cells. Left panel, Akt phosphorylation. The phosphorylation for Thr308 and Ser473 is expressed as a percentage relative to untreated cells: for Z, 105.1 plusminus 12.1% and 94.6 plusminus 8.5%; for M, 46.1 plusminus 7.0% and 48.8 plusminus 8.5% and for MZ, 35.1 plusminus 13.3% and 37.2 plusminus 8.2%, respectively. Data are presented as mean plusminus s.d.; n greater than or equal to 4. Right panel, immunoblotting analysis of the molecular association with the membrane fraction from JA16 cells mock-treated or treated with wortmannin (W) or MZ. The results are representative of three independent experiments.

Full size image (35 KB)

We also examined the consequence of myriocin and zaragozic acid treatment on the PI(3)K/Akt signaling pathway in Jurkat cells. Owing to the lack of inositol phosphatases30, these cells display a constitutively high level of both PIP3 accumulation and Akt phosphorylation. Similar to that observed in DWT6.11 cells, myriocin and zaragozic acid treatment strongly impaired the activation of Akt in Jurkat cells, as judged by the level of phosphorylation (Fig. 2b). Importantly, in both cells, evidence of a correlation existed between the impediment of nanodomains and inhibition of PI(3)K/Akt signaling activation, since cotreatment with myriocin and zaragozic acid greatly affected both events, whereas individual treatment of myriocin and zaragozic acid had only partial and no effect at all on either, respectively (Fig. 2 and Table 1).

These observations thus demonstrate a contribution of raft nanodomains in PI(3)K/Akt signaling in cell systems whereby the pathway is activated either constitutively or upon receptor stimulation. Moreover, that COase treatment blocked Akt phosphorylation in the CD28-stimulated mouse splenic T lymphocytes (Supplementary Fig. 2 online) supports a similar raft contribution in primary T cells.

Nanodomain impediment blocks Akt membrane recruitment

A prerequisite for Akt activation is its membrane recruitment driven by binding of its PH domain to PIP3, which enables the phosphorylation by PDKs. We found that nanodomain impediment in the CD28-stimulated DWT6.11 cells following myriocin and zaragozic acid treatment strongly reduces membrane recruitment of Akt and Tec (Fig. 2a), a PH-domain containing kinase also activated following PIP3-dependent membrane relocalization31. This finding was fully confirmed when COase was used to disrupt rafts (Supplementary Fig. 3a online), which further revealed that nanodomain impediment, however, alters neither the recruitment of PI(3)K to CD28 nor the increase of PIP3 (Supplementary Fig. 3b,c). We made similar observations in Jurkat cells. Indeed, whereas the myriocin and zaragozic acid treatment did not modify the level of PIP3 (Supplementary Fig. 3d), it strongly blocked membrane recruitment of Akt (Fig. 2b). Furthermore, the same treatment also inhibited the membrane recruitment of PDK1 (Fig. 2b), which also has a PIP3-binding PH domain18. Altogether, these data suggest that nanodomains are involved in the efficient membrane relocalization of Akt, and other PH domain–containing signaling proteins, including those participating in its activation.

Nanodomains contribute to the membrane on-rate of Akt

To identify the molecular basis of nanodomain contribution to Akt membrane recruitment, we examined Akt dynamics in living cells using real-time fluorescent methods along with GFP-tagged Akt, PH domain from Akt or PH domain from GRP1 (general receptor for phosphoinositides 1) (GFP-Akt, GFP-Akt-PH and GFP-GRP1-PH, respectively). Both Akt-PH and GRP1-PH domains bind to PIP3, whereas Akt-PH domain also binds to phosphatidylinositol-3,4-bisphosphate19. All GFP chimeras were expressed in Jurkat cells, which allowed us to study their recruitment dynamics at stationary state, facilitating the analysis by different biophysical approaches (see below).

We first noticed a predominant and uniform localization of all three GFP chimeras at the plasma membrane in untreated cells, and then a partial redistribution into the cytosol following myriocin and zaragozic acid treatment (for example, see GFP-Akt in Fig. 3a). By performing fluorescence loss in photobleaching (FLIP) experiments32 with repetitive bleaching on the cytosolic regions, we then monitored the molecular exchange between the plasma membrane and the cytosol (Fig. 3a). We observed a rapid loss of the fluorescence signal at the plasma membrane for all three GFP chimeras, thereby revealing a dynamic membrane association-dissociation process. Treatment with myriocin and zaragozic acid considerably accelerated the loss of the membrane fluorescence signal in all cases, thus indicating a shift in equilibrium toward the unbound state (Fig. 3a).

Figure 3: Membrane binding dynamics and nanodomain partitioning of GFP-Akt, GFP-Akt-PH and GFP-GRP1-PH in JA16 cells.

Figure 3 : Membrane binding dynamics and nanodomain partitioning of GFP-Akt, GFP-Akt-PH and GFP-GRP1-PH in JA16 cells.

(a) FLIP measurements for different GFP-tagged proteins. Left panel, consecutive confocal fluorescent images of GFP-Akt–transfected cells repeatedly bleached in cytosolic regions (small circles). Scale bar, 5 mum. Right panel, relative fluorescence intensity (RFI) (mean plusminus s.e.m. of the percentage, n greater than or equal to 12) in the crescent box over time. (b) Large-scale FRAP diffusion measurements. Upper panel, representative sequential images for GFP-Akt. Scale bar, 1 mum. Bleach box, 2 mum times 0.4 mum. Lower left panel, widening of the Gaussian one-dimensional line intensity curves along the plasma membrane over time (interval 0.7 s) for GFP-Akt. The dotted lines demark the width of the initial bleached area. Lower right panel, mean plusminus s.d. (n greater than or equal to 8) of the apparent membrane dissociation time constant tau (s). (c) t0 values as determined by FCS in Figure 1 (error bars give the s.d. of the means).

Full size image (69 KB)

To determine whether myriocin and zaragozic acid treatment caused a decrease in the association or an increase in the dissociation mechanism, we performed FRAP measurements using an original analytical method allowing the simultaneous determination of the apparent membrane dissociation times and macroscopic lateral diffusion coefficients of dynamically translocated cytosolic proteins33 (Fig. 3b). Using this approach, we were able to characterize the dynamics of the GFP chimeras in the plasma membrane (Supplementary Table 2 online), which appeared to be in good agreement with previous studies (Supplementary Discussion). Importantly, we found that once recruited to the plasma membrane the GFP chimeras diffused laterally with second-scale membrane dissociation time constants tau, which were unchanged by myriocin and zaragozic acid treatment (Fig. 3b and Supplementary Table 2). This implies that a decrease in the membrane on-rate via the PH domain of Akt causes the reduced binding to the membrane upon blockade of nanodomain formation.

Raft nanodomain partitioning of Akt and PH domains

The results of the reduced membrane on-rate of GFP chimeras in Jurkat cells treated with myriocin and zaragozic acid prompted us to investigate the interaction of Akt with nanodomains using FCS. That these molecules, once recruited, diffused laterally in the plasma membrane for a period of seconds before relocation to the cytosol met the conditions for the FCS diffusion law analysis. Our analysis revealed that GFP-Akt, GFP-Akt-PH and GFP-GRP1-PH exhibit constrained lateral diffusion in the plasma membrane (t0 = 27.7 plusminus 4.2 ms, 18.6 plusminus 2.8 ms and 22.0 plusminus 2.9 ms, respectively; Fig. 3c and Table 1). These t0 values were within the range detected for Lck1–12-GFP (t0 = 22.6 plusminus 1.8 ms; Table 1). Treatment with myriocin and zaragozic acid strongly reduced the t0 value for GFP-Akt-PH (by 90%), similar to that observed for Lck1–12-GFP (by 84%), and partially reduced that for GFP-Akt and GFP-GRP1-PH (by 44% and 65%, respectively) (Fig. 3c and Table 1). Thus, GFP-Akt-PH exhibits a lateral confinement completely mediated by raft nanodomains, whereas GFP-Akt and GFP-GRP1-PH exhibit both raft nanodomain–dependent and raft nanodomain–independent lateral confinements. Interactions by the non-PH domains of Akt with other membrane components are presumably responsible for the nanodomain-independent lateral confinement of GFP-Akt. In the case of GFP-GRP1-PH, the nanodomain-independent lateral confinement could be due to specific interactions mediated by regions within the GRP1-PH domain not directly involved in PIP3 binding34. Such interactions could also account for the much bigger membrane dissociation time constant tau observed for GFP-GRP1-PH.

Nanodomain restoration with exogenous sphingomyelin

We next examined the ability of exogenous sphingolipids to reverse the impediment of nanodomain formation and Akt activation by myriocin and zaragozic acid. We found that C12-SM (9Compound 9), which consists of a sphingosine chain (16:0) and an acyl chain (12:0), added to the culture medium efficiently integrates into the cell membrane in a manner apparently analogous to that of endogenous sphingomyelin. This was first examined on diffusion of the GPI-anchored GFP-Thy-1 protein at the surface of Jurkat cells using FRAP. We observed that the mobile fraction of GFP-Thy-1 was not significantly different when cells were either untreated or treated with myriocin and zaragozic acid or with myriocin and zaragozic acid in the presence of C12-SM (Fig. 4a). In contrast, the diffusion coefficient of GFP-Thy-1 decreased substantially (approx60%) following myriocin and zaragozic acid treatment (Fig. 4a), which agrees fully with data from FCS diffusion analysis concerning raft-favoring molecules (Table 1). Addition of exogenous C12-SM to the myriocin- and zaragozic acid–treated cells enabled recovery of the GFP-Thy-1 diffusion coefficient to a level similar to that found in untreated cells (Fig. 4a). These results indicate that C12-SM efficiently integrated into the cell membrane rather than sticking to the cell surface as hydrophobic aggregates—a conclusion further supported by analyzing the susceptibility of the plasma membrane exofacial cholesterol to COase (Supplementary Data and Supplementary Fig. 4 online).

Figure 4: Restoration of raft nanodomain formation and Akt activation in myriocin- and zaragozic acid–treated JA16 cells upon addition of C12-SM.

Figure 4 : Restoration of raft nanodomain formation and Akt activation in myriocin- and zaragozic acid|[ndash]|treated JA16 cells upon addition of C12-SM.

Cells were mock-treated (C) or treated with zaragozic acid (Z), myriocin (M), myriocin and zaragozic acid (MZ) or myriocin and zaragozic acid in the presence of 12.5 muM C12-SM (MZ + C12-SM). (a) Spot FRAP measurements of the lateral diffusion for GFP-Thy-1 (beam waist, 470 nm). Data show mean plusminus s.d. of taud, the half-time of recovery (left panel), DFRAP, the apparent diffusion coefficient (left panel) and Mf, the mobile fraction (right panel). ***P < 0.001. (b,c) FCS analysis for GFP-Akt-PH under the indicated conditions, performed as in Figure 1. Data show mean plusminus s.d. from 8 to 12 independent measurements. (d) FLIP analysis of GFP-Akt-PH as performed in Figure 3a. Each curve shows the mean plusminus s.e.m. (n greater than or equal to 12). (e) Immunoblotting analysis of Akt phosphorylation as performed in Figure 2a. The phosphorylation for Thr308 and Ser473 is expressed as a percentage relative to untreated cells: for MZ, 35.1 plusminus 13.3% and 37.2 plusminus 8.2%; for MZ + C12-SM, 105.3 plusminus 17.6% and 106.5 plusminus 11.2%, respectively. Data are presented as means plusminus s.d.; n greater than or equal to 4.

Full size image (78 KB)

Notably, the integration of C12-SM re-established the same lateral confinement for GFP-Akt-PH in myriocin- and zaragozic acid–treated Jurkat cells as that found in untreated cells (Fig. 4b and Table 1). This suggested a restoration of normal nanodomain organization, observed with either similar or slightly higher levels of sphingomyelin compared to untreated cells (Tables 1 and 2). The ability of sphingomyelin alone to restore the normal diffusion of GFP-Akt-PH in the myriocin- and zaragozic acid–treated cells agrees with the fact that reduction of cholesterol alone with zaragozic acid has no effect on GFP-Akt-PH lateral diffusion (Fig. 4c). Importantly, we also observed a total recovery of both the GFP-Akt-PH membrane binding (Fig. 4d) and Akt phosphorylation efficiency (Fig. 4e) in the nanodomain-restored cells. Our data thus strongly corroborate a key role of raft nanodomains in the membrane recruitment and activation of Akt.

Raft nanodomains and PI(3)K/Akt signaling in COS-7 cells

To extend the conclusion made in T cells, we conducted experiments in COS-7 cells to examine the contribution of raft nanodomains toward Akt activation upon stimulation by insulin-like growth factor 1 (IGF1), one of the best characterized activators of the PI(3)K/Akt signaling pathway. We found that nanodomain impediment with myriocin and zaragozic acid treatment results in a strong inhibition of IGF1-triggered Akt phosphorylation. The same treatment did not, however, inhibit IGF1-triggered phosphorylation of insulin receptor substrate 1 (IRS1), which induces recruitment of PI(3)K to the IGF1 receptor (IGF1R)–IRS1 complex (Fig. 5a). Furthermore, we found a substantially reduced IGF1-triggered membrane association of Akt and PDK1 (Fig. 5b). These data thus substantiate the role of rafts in promoting Akt recruitment and activation.

Figure 5: Akt activation trigger by IGF1 in COS-7 cells.

Figure 5 : Akt activation trigger by IGF1 in COS-7 cells.

Treatments shown are mock treatment (C) and treatment with myriocin and zaragozic acid (MZ). (a) Immunoblotting analysis of Akt and IRS1 phosphorylation after cell stimulation or no stimulation with IGF1 (15 ng ml-1). The phosphorylation upon cell stimulation for Thr308 and Ser473 of Akt expressed as a percentage relative to untreated cells is 41.5 plusminus 10.6% and 48.1 plusminus 7.4%, respectively (presented as mean plusminus s.d., n = 3). Phosphorylation of IRS1 at Tyr612 was analyzed using an antibody directly against human phospho-IRS1 (Tyr612). (b) Immunoblotting analysis of the molecular association with the membrane fraction after cell stimulation or no stimulation with IGF1. The results are representative of three independent experiments. (c) Confocal microscopy of GFP-Akt and GFP-Akt-PH in p110-CAAX–expressing cells. Scale bar, 10 mum. (d) Apparent dissociation time constants (tau) from the plasma membrane as determined in Figure 3b. Data show the mean plusminus s.d. (n greater than or equal to 8). (e) t0 values as determined by FCS in Figure 1 (error bars give the s.d. of the means).

Full size image (42 KB)

To determine Akt association with nanodomains upon PIP3 accumulation in COS-7 cells, we made use of a constitutively activated, membrane-targeted form of PI(3)K (p110-CAAX), because our FCS diffusion analysis requires molecules to be at stationary state. p110-CAAX was transfected along with GFP-Akt or GFP-Akt-PH. Expression of p110-CAAX elicited PIP3 production and induced re-localization of GFP-Akt and GFP-Akt-PH, which are normally only found in the cytosol, to the plasma membrane (Fig. 5c). The FRAP analysis, as performed in Figure 3b, indicated an apparent dissociation time constant tau of the GFP chimeras of several seconds in the untreated cells (Fig. 5d and Supplementary Table 2). The confinement index t0 as determined by FCS was 10.5 plusminus 0.2 ms and 6.4 plusminus 2.1 ms, for GFP-Akt and GFP-Akt-PH (Fig. 5e and Table 1) respectively. These t0 values were very close to that for Lck1–12-GFP (t0 = 8.3 plusminus 0.9 ms; Table 1). Treatment with myriocin and zaragozic acid was found to strongly inhibit both GFP-Akt and GFP-Akt-PH localization to the plasma membrane, though more severely in the former (Fig. 5c). In fact, under this condition, the particularly low level of membrane association of GFP-Akt was not sufficient to adequately perform FRAP and FCS analyses (data not shown). For GFP-Akt-PH, whose membrane association was still detectable, myriocin and zaragozic acid treatment did not affect the dissociation time constant tau (Fig. 5d and Supplementary Table 2); however, much like that of Lck1–12-GFP, the t0 value became null (Fig. 5e and Table 1). These data fit nicely with the proposed model that binding of the Akt-PH domain to PIP3 drives Akt recruitment to raft nanodomains, which in turn enhances and/or stabilizes Akt membrane association. Moreover, since we examined here the Akt re-localization to raft nanodomains under conditions where activation of the PI(3)K pathway bypasses receptor stimulation, these results provide further evidence that the raft partitioning of Akt makes up part of the general activation mechanism of the PI(3)K/Akt signaling cascade.

In addition, we conducted studies on mouse embryonic fibroblasts (MEF) that express p110-CAAX in combination with GFP-Akt-PH or GFP-GRP1-PH. The results again confirm the dynamic partitioning of PIP3-interacting PH domains into nanodomains and that blockade of the nanodomain formation impairs their membrane association (Table 1, Supplementary Fig. 5 and Supplementary Table 2 online). Altogether, the data from real-time analysis with GFP chimeras in Jurkat, COS-7 and MEF cells suggest a common existence of PIP3-containing specific nanodomains, formed in the plasma membrane inner leaflet, which heighten the recruitment from the cytosol of PH domain–containing signaling proteins, such as Akt.

Top

Discussion

In this study, we investigated the role of lateral membrane organizations in the activation of the PI(3)K/Akt signaling pathway. Our data reveal that raft nanodomains play a crucial role in the initiation of this signaling axis by promoting efficient membrane recruitment and activation of Akt upon PIP3 accumulation in the plasma membrane inner leaflet.

We recently reported the identification of dynamic and nanoscaled membrane domains in live COS-7 cells by using a new FCS approach17. Here, we have extended the characterization of these nanodomains. We have provided evidence for their existence in the plasma membrane of various cell types, including mouse and human T cell lines and primary mouse embryonic fibroblasts. Moreover, we have demonstrated the presence of these nanodomains in both the outer and inner leaflets of the plasma membrane. We stress the point that these membrane nanodomains may correspond either to pre-existing structures into which individual molecules are dynamically partitioned on a timescale of tens to hundreds of milliseconds or to self-promoting lateral assemblies with a confinement time corresponding to the lifespan of individual nanodomains16, 17.

Using myriocin and zaragozic acid to specifically inhibit the metabolism of sphingolipids and cholesterol respectively, we established the essential role of these two lipids in the formation of nanodomains. The metabolic inhibition of sphingolipid and cholesterol synthesis in our experiments did not induce significant modifications or compensatory processes among other lipid species. Furthermore, the observed effect of sphingolipid and cholesterol metabolic inhibition on the nanodomain-based diffusion mode was selective, as it did not impair other modes of lateral diffusion in the plasma membrane, such as those probed with FL-PE and TfR-GFP.

Our data show that raft nanodomains are present in both leaflets of the plasma membrane. These findings do not, however, necessarily imply that sphingolipids and cholesterol promote nanodomain formation at the two leaflets via the same mechanism. The interaction between sphingomyelin and cholesterol is a well-established notion, thought to be the basis of the formation of raft domains. Considering the abundance of these two lipids in the outer leaflet, it is most probable that at this location they directly participate in nanodomain formation. However, the contribution of sphingolipids to nanodomain formation in the inner leaflet of the plasma membrane is noteworthy considering the relative paucity of this lipid species there2. Such contribution may be due to indirect transmembrane effects from the outer leaflet35. Indeed, future studies to investigate membrane leaflet "coupling" could be of particular interest with regards to the raft lipid-dependent lateral organization of the cell membrane. It is of note that the nanodomains probed by Lck1–12-GFP and GFP-Akt-PH in this study share structural similarities with the previously documented inner leaflet nanodomains probed by the minimal anchor of H-Ras attached to GFP (GFP-tH)4, 12 (see Supplementary Discussion for a more detailed discussion). Awaiting the direct comparison between the two types of nanodomains, the possibility exists that their formation is underpinned by some common biophysical principles.

Simultaneous reduction of sphingolipids and cholesterol could hence represent an outstanding way of altering rafts in cell membranes. It is devoid of various side effects associated with some existing approaches, such as that used with methyl-beta-cyclodextrin (Ms zligCD, 10Compound 10). For instance, and in contrast to Ms zligCD (refs. 36,37; also see ref. 17), myriocin and zaragozic acid treatment did not induce molecular immobilization in the plasma membrane nor did it significantly alter overall tyrosine phosphorylation and MAPK/ERK pathway activation in either resting or stimulated cells. Assuming a high sensitivity of these two latter signaling events to changes happening within the cell and in particular in the membrane38, this also argues against the occurrence of general cellular alterations with myriocin and zaragozic acid treatment.

We observed a clear correlation between the presence of raft nanodomains in living cells and efficient Akt membrane recruitment and activation under various experimental conditions. Such a correlation existed regardless of whether the upstream PIP3 accumulation was triggered by receptor (CD28 or IGF1R) stimulation or an imbalance between inositol kinases and phosphatases (as in Jurkat cells or the p110-CAAX–expressing COS-7 cells). Indeed, we verified this correlation by demonstrating that (i) depletion of sphingolipid content partially affected both events; (ii) mild reduction of cholesterol content had no effect on either; (iii) concomitant reduction of sphingolipid and cholesterol content strongly inhibited both events; (iv) cholesterol oxidation impaired both events; and finally (v) addition of exogenous sphingomyelin fully reversed the inhibition of both events caused by sphingolipid and cholesterol reduction.

This study further reveals that the PH domain of Akt drives its dynamic partition into raft nanodomains in the plasma membrane inner leaflet, and that the presence of these nanodomains contributes to Akt membrane on-rate. This suggests an ability of nanodomains to operate as functional hot spots for the recruitment of Akt upon accumulation of PIP3. Future studies are required to investigate the molecular mechanism by which the PIP3-containing nanodomains are formed. PIP3 exhibits structural features thought to be unfavorable for inclusion into raft-type membrane environments. Here, we observed the raft nanodomain partition of PIP3-bound PH domains, but we have no information on the "free" PIP3 molecules. In light of several previous reports on the protein-induced sequestration of PIP2 into membrane domains, we propose that the PIP3-containing nanodomains are formed only upon PH domain binding to PIP3 (see Supplementary Discussion for details). Indeed, it has previously been shown that different proteins enabled the PIP2 sequestration into cholesterol-dependent ordered domains39, 40, 41, 42. The documented mechanisms appeared to involve in particular (i) a combination of the acyl chains covalently linked to the PIP2 binding proteins as well as the penetration into the bilayer via clusters of basic and hydrophobic residues, resulting in the local recruitment and nucleation of raft lipids39, 41, 42 and (ii) the presence of two PIP2 binding modules40, which could give rise to clustering and spatial reorientation of saturated acyl chains. As a result, the overall raft-favoring interactions overcome the raft-disfavoring tendency of the 2-arachidonate chain of PIP2 (which is also present in PIP3), causing polyphosphoinositide sequestration into ordered membrane domains. In the case of Akt, this could involve PIP3 binding by the homo- or heterodimers of PH domains43, 44 or PH domains associated with other putative raft-preferring proteins, thereby allowing multivalent binding between PH domains and PIP3. Conversely, the multivalent membrane association of PH domains could be substantially enhanced and/or stabilized by protein-lipid interactions within nanodomains where PIP3 diffusion is laterally constrained. In fact, we propose the possibility that the PIP3-containing nanodomains are self-promoting and self-strengthening membrane assemblies whose formation and (multivalent) protein PH domain binding to PIP3 mutually impinge on each other.

In addition, raft nanodomains might also contribute to Akt phosphorylation by PDKs, either through promoting membrane recruitment of the latter, as has already been observed for PDK1, and/or confining the Akt within close proximity. It has been reported that PDK1 was recovered in DRMs (ref. 23). In the case of PDK2s, such as the mammalian target of rapamycin complex 2 (mTORC2)20 and DNA-dependent protein kinase (DNA-PK)45, both DNA-PK (ref. 25) and positive regulators of mTORC2 (ref. 46) were found associated with DRMs.

The contribution of rafts to Akt recruitment and activation may exist not only on a nanoscale but also on a microscale after they coalesce into larger membrane domains. As one most prominent structural potential, native rafts could undergo coalescence to form optically resolvable platforms or condensed membranes47. Upon antigen stimulation of T cells, GFP-Akt-PH was rapidly recruited to the plasma membrane and accumulated both inside and outside the immunological synapse48, 49, and it was sustained there for many hours48. It is thus tempting to suggest that rafts are involved in both the speedy and heightened first phase and the sustained second phase of PI(3)K/Akt signaling pathway activation.

Finally, our study also suggests that raft nanodomains may play a general role in those signaling processes involved in membrane recruitment and activation that depend on the binding of the protein PH domain to PIP3. Membrane nanodomains may therefore participate in many if not most of the PI(3)K-mediated signaling events having known implication in a wide variety of cellular responses in both physiological and pathological conditions.

Top

Methods

Cell treatments.

For sphingolipid biosynthesis inhibition, cells were cultured in serum-containing complete medium for 48 h in the presence of 25 muM (10 muM for COS-7) myriocin and then serum-starved for 14 to 18 h in the presence of 25 muM myriocin. For cholesterol biosynthesis inhibition, cells were cultured in complete medium for 48 h and then serum-starved for 14 to 18 h in the presence of 5 muM zaragozic acid. For combined sphingolipid and cholesterol biosynthesis inhibition, cells were cultured in complete medium for 48 h in the presence of 25 muM myriocin and then serum-starved for 14 to 18 h in the presence of 25 muM myriocin and 5 muM zaragozic acid. For sphingomyelin restoration, we added C12-SM alongside myriocin throughout the treatment at the indicated concentration. For cholesterol oxidation, cells were treated for 1 h at 37 °C with 10 U ml-1Streptomyces sp. cholesterol oxidase. For actin cytoskeleton depolymerization, cells were treated with latrunculin A as described17. For inhibition of PI(3)K, cells were treated for 4 h at 37 °C with 1 muM wortmannin (11Compound 11). In the present work, all analyses were conducted on serum-starved cells, except where otherwise indicated. Serum starving was found to not significantly affect the FCS measurements in the untreated cells.

Diffusion measurements by FCS and spot FRAP.

We performed FCS measurements at 37 °C in Hank's buffered saline solution (HBSS) with 10 mM HEPES pH 7.4, on a custom-made stage of an Axiovert 200M inverted confocal microscope as previously described16. For the T cell lines, coverslips were first coated with poly-L-lysine before cell seeding. All measurements were carried out in three to five different areas on a minimum of five different cells. Each measurement was obtained from 20 runs.

We performed spot FRAP measurements at 37 °C on the same setup by illuminating the sample with an excitation power of 3 mW for photobleaching measurements and 3 muW for pre-bleach and post-bleach measurements. All measurements were carried out in three to five different areas on a minimum of five different cells.

The FRAP measurement analysis was performed as previously described17

Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com

where alpha represents the fraction of the mobile species Mf, K is a parameter related to the degree of bleaching, tauD is the characteristic time of recovery, and Fp and F0 are the fluorescence intensities before and immediately after the bleach event, respectively.

FLIP experiments.

We performed FLIP imaging at 37 °C on a Zeiss LSM 510 confocal microscope using the 488 nm line of a 25 mW argon laser beam, a Zeiss plan-Apochromat 63times NA = 1.4 oil immersion objective and the filter set supplied by the manufacturer. Briefly, cells were repeatedly photobleached within cytosolic regions of interest (ROI) (Empty set variant = 1.24 mum). Between the bleaching periods, the cells were imaged with low light intensity. We recorded one image every 5 s for GFP-Akt and GFP-Akt-PH or every approx8 s for GFP-GRP1-PH, at 1 % laser transmission, in turn with 20 iteration photobleaching illuminations (approx1.8 s) at 100% laser transmission. Plasma membrane fluorescence decay was measured using the line scan function of MetaMorph software (Molecular Devices) and plotted versus time. Graphs showed the average fluorescence measurement of at least four individual recordings plusminus s.e.m. The experiments were reproduced three times independently.

Apparent membrane diffusion coefficient and dissociation analysis by FRAP measurements on large membrane regions.

We performed the measurements at 37 °C on a Zeiss LSM 510 confocal microscope. Briefly, a large area of plasma membrane (2.1 mum times 0.7 mum) was photobleached by 10 iterative scans at 100% laser output from the 488 nm, 25 mW Argon laser. Recovery was followed over approx25 s at a frequency of 0.5 s per frame. We quantified the apparent membrane dissociation times and macroscopic lateral diffusion coefficients of GFP-tagged proteins as previously described33. Briefly, the fluorescence recovery of GFP-tagged proteins is recorded over time after the photobleaching of a large area of the plasma membrane. We quantified the data from a series of images by analyzing the evolution of the shape of a one-dimensional line intensity curve along the plasma membrane. Each curve was fitted by a Gaussian function. A widening of the curves indicates a substantial contribution of lateral diffusion of the molecule to the recovery process, whereas the absence of widening indicates the predominance of dissociation during the recovery process. By computing the widening of the curves as a function of time, it is possible to derive an apparent macroscopic lateral diffusion coefficient and an apparent dissociation time constant from the plasma membrane. To determine the mobile fraction, we used a more classical fitting based on the Axelrod equation50.

Statistical analysis.

All statistical analyses were performed using GraphPad Prism 5.00. For Supplementary Table 1, a nonparametric two-tailed unpaired Mann-Whitney test was used; for sphingomyelin, P = 0.0022. For the FRAP data in Figure 4a, a one-way ANOVA test with Tukey's multiple comparison post-tests were used; the P values of the post-test were 1.34 times 10-5 for taud between C and MZ and 1.16 times 10-5 for taud between MZ and MZ + C12-SM.

Other methods.

See Supplementary Methods for cell culture, preparation, transfection and incorporation with fluorescent lipid analogs, and for antibodies and reagents, cell stimulation, SDS-PAGE, immunoblotting and quantification, GFP-conjugated chimeras, membrane fraction preparation and lipidomics studies (total phospholipids, glycerophospholipid fatty acid methyl esters and phosphoinositides).

Note: Supplementary information and chemical compound information is available on the Nature Chemical Biology website.



Top

Acknowledgments

This research project was supported by institutional grants from Institut National de la Santé et de la Recherche Médicale and Centre National de la Recherche Scientifique, and by specific grants from Agence Nationale de la Recherche, Association pour la Recherche sur le Cancer, European regional development fund, Fondation pour la Recherche Médicale, Institut National du Cancer (PL06-026, PL96-009 and PF108-05), Ligue Nationale Française contre le Cancer, Ministère de l'Éducation Nationale, de la Recherche et de la Technologie and Centre National de la Recherche Scientifique. Part of the work was supported by IFR30 Lipidomics of the Functional Exploration Platform of Toulouse Genopole. R.L. and F.C. were awarded fellowships from the Ministère de l'Éducation Nationale, de la Recherche et de la Technologie. We thank M. Fallet (PICsL imaging core facility) and F. Garçon for technical assistance, P. Chavrier (Institut Curie), J. Downward (Cancer Research UK London Research Institute), H. Lelouard (Centre d'Immunologie de Marseille-Luminy), S. Méresse (Centre d'Immunologie de Marseille-Luminy) and T. Meyer (Stanford University Medical School) for reagents, Y. Xia for chemical structure drawing, K. Simons, G. van Meer and C. Zhu for discussion, P. Golstein, J.P. Gorvel, A.-O. Hueber, L. Leserman, A. Pamidi and T. Reisine for critical reading of the manuscript and E. Witty (AngloScribe) for editing the English.

Author Contributions

R.L., X.-J.G., F.C., Y.H., B.P., D.M. and H.-T.H. conceived and designed the experiments. R.L., X.-J.G., F.C., Y.H., O.H., A.-M.B. and S.M.S. performed the experiments. R.L., X.-J.G., F.C., Y.H., B.P., D.M. and H.-T.H. analyzed the data. P.-F.L., H.R., D.O., G.B., J.A.N. and B.P. contributed reagents, materials and analysis tools. D.M. and H.-T.H. wrote the paper with contributions from R.L., X.-J.G., F.C., Y.H. and B.P.

Received 16 May 2008; Accepted 27 June 2008; Published online 20 July 2008.

Top

References

  1. Simons, K. & Ikonen, E. Functional rafts in cell membranes. Nature 387, 569–572 (1997). | Article | PubMed | ISI | ChemPort |
  2. Brown, D.A. & London, E. Functions of lipid rafts in biological membranes. Annu. Rev. Cell Dev. Biol. 14, 111–136 (1998). | Article | PubMed | ISI | ChemPort |
  3. Edidin, M. The state of lipid rafts: from model membranes to cells. Annu. Rev. Biophys. Biomol. Struct. 32, 257–283 (2003). | Article | PubMed | ISI | ChemPort |
  4. Hancock, J.F. Lipid rafts: contentious only from simplistic standpoints. Nat. Rev. Mol. Cell Biol. 7, 456–462 (2006). | Article | PubMed | ISI | ChemPort |
  5. Marguet, D., Lenne, P.F., Rigneault, H. & He, H.T. Dynamics in the plasma membrane: how to combine fluidity and order. EMBO J. 25, 3446–3457 (2006). | Article | PubMed | ChemPort |
  6. Jacobson, K., Mouritsen, O.G. & Anderson, R.G. Lipid rafts: at a crossroad between cell biology and physics. Nat. Cell Biol. 9, 7–14 (2007). | Article | PubMed | ISI | ChemPort |
  7. Simons, K. & Toomre, D. Lipid rafts and signal transduction. Nat. Rev. Mol. Cell Biol. 1, 31–39 (2000). | Article | PubMed | ISI | ChemPort |
  8. Munro, S. Lipid rafts: elusive or illusive? Cell 115, 377–388 (2003). | Article | PubMed | ISI | ChemPort |
  9. Pralle, A., Keller, P., Florin, E.L., Simons, K. & Horber, J.K. Sphingolipid-cholesterol rafts diffuse as small entities in the plasma membrane of mammalian cells. J. Cell Biol. 148, 997–1008 (2000). | Article | PubMed | ISI | ChemPort |
  10. Niv, H., Gutman, O., Kloog, Y. & Henis, Y.I. Activated K-Ras and H-Ras display different interactions with saturable nonraft sites at the surface of live cells. J. Cell Biol. 157, 865–872 (2002). | Article | PubMed | ISI | ChemPort |
  11. Zacharias, D.A., Violin, J.D., Newton, A.C. & Tsien, R.Y. Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells. Science 296, 913–916 (2002). | Article | PubMed | ISI | ChemPort |
  12. Prior, I.A., Muncke, C., Parton, R.G. & Hancock, J.F. Direct visualization of Ras proteins in spatially distinct cell surface microdomains. J. Cell Biol. 160, 165–170 (2003). | Article | PubMed | ISI | ChemPort |
  13. Gaus, K. et al. Visualizing lipid structure and raft domains in living cells with two-photon microscopy. Proc. Natl. Acad. Sci. USA 100, 15554–15559 (2003). | Article | PubMed | ChemPort |
  14. Sharma, P. et al. Nanoscale organization of multiple GPI-anchored proteins in living cell membranes. Cell 116, 577–589 (2004). | Article | PubMed | ISI | ChemPort |
  15. Ianoul, A. et al. Imaging nanometer domains of beta-adrenergic receptor complexes on the surface of cardiac myocytes. Nat. Chem. Biol. 1, 196–202 (2005). | Article | PubMed | ChemPort |
  16. Wawrezinieck, L., Rigneault, H., Marguet, D. & Lenne, P.F. Fluorescence correlation spectroscopy diffusion laws to probe the submicron cell membrane organization. Biophys. J. 89, 4029–4042 (2005). | Article | PubMed | ChemPort |
  17. Lenne, P.F. et al. Dynamic molecular confinement in the plasma membrane by microdomains and the cytoskeleton meshwork. EMBO J. 25, 3245–3256 (2006). | Article | PubMed | ChemPort |
  18. Hanada, M., Feng, J. & Hemmings, B.A. Structure, regulation and function of PKB/AKT-a major therapeutic target. Biochim. Biophys. Acta 1697, 3–16 (2004). | Article | PubMed | ISI | ChemPort |
  19. Scheid, M.P. & Woodgett, J.R. Unravelling the activation mechanisms of protein kinase B/Akt. FEBS Lett. 546, 108–112 (2003). | Article | PubMed | ISI | ChemPort |
  20. Sarbassov, D.D., Guertin, D.A., Ali, S.M. & Sabatini, D.M. Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex. Science 307, 1098–1101 (2005). | Article | PubMed | ISI | ChemPort |
  21. Zhuang, L., Lin, J., Lu, M.L., Solomon, K.R. & Freeman, M.R. Cholesterol-rich lipid rafts mediate akt-regulated survival in prostate cancer cells. Cancer Res. 62, 2227–2231 (2002). | PubMed | ISI | ChemPort |
  22. Adam, R.M. et al. Cholesterol sensitivity of endogenous and myristoylated Akt. Cancer Res. 67, 6238–6246 (2007). | Article | PubMed | ISI | ChemPort |
  23. Lee, K.Y., D'Acquisto, F., Hayden, M.S., Shim, J.H. & Ghosh, S. PDK1 nucleates T cell receptor-induced signaling complex for NF-kappaB activation. Science 308, 114–118 (2005). | Article | PubMed | ISI | ChemPort |
  24. Hill, M.M., Feng, J. & Hemmings, B.A. Identification of a plasma membrane Raft-associated PKB Ser473 kinase activity that is distinct from ILK and PDK1. Curr. Biol. 12, 1251–1255 (2002). | Article | PubMed | ISI | ChemPort |
  25. Lucero, H., Gae, D. & Taccioli, G.E. Novel localization of the DNA-PK complex in lipid rafts: a putative role in the signal transduction pathway of the ionizing radiation response. J. Biol. Chem. 278, 22136–22143 (2003). | Article | PubMed | ISI | ChemPort |
  26. Pages, F. et al. Binding of phosphatidylinositol-3-OH kinase to CD28 is required for T-cell signalling. Nature 369, 327–329 (1994). | Article | PubMed | ISI | ChemPort |
  27. Wenger, J. et al. Diffusion analysis within single nanometric apertures reveals the ultrafine cell membrane organization. Biophys. J. 92, 913–919 (2007). | Article | PubMed | ChemPort |
  28. Miyake, Y., Kozutsumi, Y., Nakamura, S., Fujita, T. & Kawasaki, T. Serine palmitoyltransferase is the primary target of a sphingosine-like immunosuppressant, ISP-1/myriocin. Biochem. Biophys. Res. Commun. 211, 396–403 (1995). | Article | PubMed | ISI | ChemPort |
  29. Bergstrom, J.D. et al. Zaragozic acids: a family of fungal metabolites that are picomolar competitive inhibitors of squalene synthase. Proc. Natl. Acad. Sci. USA 90, 80–84 (1993). | Article | PubMed | ChemPort |
  30. Freeburn, R.W. et al. Evidence that SHIP-1 contributes to phosphatidylinositol 3,4,5-trisphosphate metabolism in T lymphocytes and can regulate novel phosphoinositide 3-kinase effectors. J. Immunol. 169, 5441–5450 (2002). | PubMed | ISI | ChemPort |
  31. Berg, L.J., Finkelstein, L.D., Lucas, J.A. & Schwartzberg, P.L. Tec family kinases in T lymphocyte development and function. Annu. Rev. Immunol. 23, 549–600 (2005). | Article | PubMed | ISI | ChemPort |
  32. Lippincott-Schwartz, J. & Patterson, G.H. Development and use of fluorescent protein markers in living cells. Science 300, 87–91 (2003). | Article | PubMed | ISI | ChemPort |
  33. Oancea, E., Teruel, M.N., Quest, A.F. & Meyer, T. Green fluorescent protein (GFP)-tagged cysteine-rich domains from protein kinase C as fluorescent indicators for diacylglycerol signaling in living cells. J. Cell Biol. 140, 485–498 (1998). | Article | PubMed | ISI | ChemPort |
  34. Varnai, P. et al. Selective cellular effects of overexpressed pleckstrin-homology domains that recognize PtdIns(3,4,5)P3 suggest their interaction with protein binding partners. J. Cell Sci. 118, 4879–4888 (2005). | Article | PubMed | ChemPort |
  35. Kiessling, V., Crane, J.M. & Tamm, L.K. Transbilayer effects of raft-like lipid domains in asymmetric planar bilayers measured by single molecule tracking. Biophys. J. 91, 3313–3326 (2006). | Article | PubMed | ChemPort |
  36. Kwik, J. et al. Membrane cholesterol, lateral mobility, and the phosphatidylinositol 4,5-bisphosphate-dependent organization of cell actin. Proc. Natl. Acad. Sci. USA 100, 13964–13969 (2003). | Article | PubMed | ChemPort |
  37. Kabouridis, P.S., Janzen, J., Magee, A.L. & Ley, S.C. Cholesterol depletion disrupts lipid rafts and modulates the activity of multiple signaling pathways in T lymphocytes. Eur. J. Immunol. 30, 954–963 (2000). | Article | PubMed | ISI | ChemPort |
  38. Magee, A.I., Adler, J. & Parmryd, I. Cold-induced coalescence of T-cell plasma membrane microdomains activates signalling pathways. J. Cell Sci. 118, 3141–3151 (2005). | Article | PubMed | ChemPort |
  39. Tong, J. et al. Role of GAP-43 in sequestering phosphatidylinositol 4,5-bisphosphate to Raft bilayers. Biophys. J. 94, 125–133 (2008). | Article | PubMed | ChemPort |
  40. Gokhale, N.A., Abraham, A., Digman, M.A., Gratton, E. & Cho, W. Phosphoinositide specificity of and mechanism of lipid domain formation by annexin A2-p11 heterotetramer. J. Biol. Chem. 280, 42831–42840 (2005). | Article | PubMed | ChemPort |
  41. Epand, R.M., Vuong, P., Yip, C.M., Maekawa, S. & Epand, R.F. Cholesterol-dependent partitioning of PtdIns(4,5)P2 into membrane domains by the N-terminal fragment of NAP-22 (neuronal axonal myristoylated membrane protein of 22 kDa). Biochem. J. 379, 527–532 (2004). | Article | PubMed | ChemPort |
  42. Golebiewska, U. et al. Membrane-bound basic peptides sequester multivalent (PIP2), but not monovalent (PS), acidic lipids. Biophys. J. 91, 588–599 (2006). | Article | PubMed | ChemPort |
  43. Datta, K. et al. AH/PH domain-mediated interaction between Akt molecules and its potential role in Akt regulation. Mol. Cell. Biol. 15, 2304–2310 (1995). | PubMed | ISI | ChemPort |
  44. Calleja, V. et al. Intramolecular and intermolecular interactions of protein kinase B define its activation in vivo. PLoS Biol. 5, e95 (2007). | Article | PubMed | ChemPort |
  45. Feng, J., Park, J., Cron, P., Hess, D. & Hemmings, B.A. Identification of a PKB/Akt hydrophobic motif Ser-473 kinase as DNA-dependent protein kinase. J. Biol. Chem. 279, 41189–41196 (2004). | Article | PubMed | ISI | ChemPort |
  46. Jones, K.A., Jiang, X., Yamamoto, Y. & Yeung, R.S. Tuberin is a component of lipid rafts and mediates caveolin-1 localization: role of TSC2 in post-Golgi transport. Exp. Cell Res. 295, 512–524 (2004). | Article | PubMed | ChemPort |
  47. Gaus, K. et al. Condensation of the plasma membrane at the site of T lymphocyte activation. J. Cell Biol. 171, 121–131 (2005). | Article | PubMed | ISI | ChemPort |
  48. Costello, P.S., Gallagher, M. & Cantrell, D.A. Sustained and dynamic inositol lipid metabolism inside and outside the immunological synapse. Nat. Immunol. 3, 1082–1089 (2002). | Article | PubMed | ISI | ChemPort |
  49. Harriague, J. & Bismuth, G. Imaging antigen-induced PI3K activation in T cells. Nat. Immunol. 3, 1090–1096 (2002). | Article | PubMed | ISI | ChemPort |
  50. Axelrod, D., Koppel, D.E., Schlessinger, J., Elson, E. & Webb, W.W. Mobility measurement by analysis of fluorescence photobleaching recovery kinetics. Biophys. J. 16, 1055–1069 (1976). | Article | PubMed | ISI | ChemPort |
  1. Centre d'Immunologie de Marseille-Luminy, Université de la Méditerranée, Parc scientifique de Luminy, Case 906, F-13288 Marseille Cedex 09, France.
  2. Institut National de la Santé et de la Recherche Médicale, Unité Mixte de Recherche 631, Parc scientifique de Luminy, Case 906, F-13288 Marseille Cedex 09, France.
  3. Centre National de la Recherche Scientifique, Unité Mixte de Recherche 6102, Parc scientifique de Luminy, Case 906, F-13288 Marseille Cedex 09, France.
  4. Centre National de la Recherche Scientifique, Unité Mixte de Recherche 6231, Laboratoire des Interactions Moléculaires et Systèmes Membranaires, Université Paul Cézanne, avenue Escadrille Normandie-Nieman, F-13331 Marseille Cedex 20, France.
  5. Institut Fresnel, Université Paul Cézanne, Domaine Universitaire de Saint Jérôme, F-13397 Marseille Cedex 20, France.
  6. Centre National de la Recherche Scientifique, Unité Mixte de Recherche 6133, Domaine Universitaire de Saint Jérôme, F-13397 Marseille Cedex 20, France.
  7. Institut National de la Santé et de la Recherche Médicale, Centre de Recherche en Cancérologie de Marseille, Unité Mixte de Recherche 891, 27 Bd Lei Roure, F-13009 Marseille, France.
  8. Institut Paoli-Calmettes, 232 Bd Sainte Marguerite, F-13273 Marseille Cedex 09, France.
  9. Université de la Méditerranée, 58 Boulevard Charles Livon, F-13007 Marseille, France.
  10. Institut National de la Santé et de la Recherche Médicale, Unité 567, 22 rue Méchain, 75014 Paris, France.
  11. Centre National de la Recherche Scientifique, Unité Mixte de Recherche 8104, Université René Descartes, 22 rue Méchain, 75014 Paris, France.
  12. Institut National de la Santé et de la Recherche Médicale, Unité 563, Centre de Physiopathologie de Toulouse Purpan, Département d'Oncogenèse, Signalisation et Innovation thérapeutique, Place du Docteur Baylac, Toulouse F-31300, France.
  13. Université Toulouse III Paul-Sabatier, Place du Docteur Baylac, Toulouse F-31400, France.
  14. These authors contributed equally to this work.

Correspondence to: Hai-Tao He1,2,3 e-mail: he@ciml.univ-mrs.fr

Correspondence to: Bernard Payrastre12,13 e-mail: bernard.payrastre@toulouse.inserm.fr

MORE ARTICLES LIKE THIS

These links to content published by NPG are automatically generated.

RESEARCH

Dynamic molecular confinement in the plasma membrane by microdomains and the cytoskeleton meshwork

The EMBO Journal Article (26 Jul 2006)

Extra navigation

Subscribe to Nature Chemical Biology

Subscribe

Search PubMed for

Open Innovation Challenges

naturejobs

More science jobs
Post a job for free