Search

A variety of signal transduction pathways including PI turnover, MAP kinase activation, and PI 3-kinase activation have been shown to be affected by changes in cellular cholesterol content. However, no information is available regarding the locus (or loci) in the pathways that are susceptible to modulation by cholesterol. We report here that depletion of cholesterol with methyl-beta-cyclodextrin increases cell surface (125)I-EGF binding by approximately 40% via a mechanism that does not involve externalization of receptors from an internal pool. Cholesterol depletion also enhances in vivo EGF receptor autophosphorylation 2-5-fold without altering the rate of receptor dephosphorylation. In vitro kinase assays, which are done under conditions where phosphotyrosine phosphatases are inhibited and receptor trafficking cannot occur, demonstrate that treatment with methyl-beta-cyclodextrin leads to an increase in intrinsic EGF receptor tyrosine kinase activity. EGF receptors are localized in cholesterol-enriched lipid rafts but are released from this compartment upon treatment with methyl-beta-cyclodextrin. These data are consistent with the interpretation that localization to lipid rafts partially suppresses the binding and kinase functions of the EGF receptor and that depletion of cholesterol releases the receptor from lipid rafts, relieving the functional inhibition of the receptor. Cholesterol depletion also inhibits EGF internalization and down-regulation of the EGF receptor, and this likely contributes to the enhanced ability of EGF to stimulate downstream signaling pathways such as the activation of MAP kinase.

We have identified physical and functional interactions between the epidermal growth factor (EGF) receptor and phospholipid scramblase 1 (PLSCR1), an endofacial plasma membrane protein proposed to affect phospholipid organization. PLSCR1, a palmitoylated protein, was found to partition with the EGF receptor in membrane lipid rafts. Cell stimulation with EGF transiently elevated Tyr-phosphorylation of PLSCR1, peaking at 5 min. Although PLSCR1 is a known substrate of c-Abl [Sun, J., et al. (2001) J.Biol. Chem. 276, 28984-28990], the Abl inhibitor STI571 did not substantially affect its EGF-dependent phosphorylation, suggesting PLSCR1 is a substrate of the EGF receptor kinase, or another EGF-activated kinase. Coinciding with phosphorylation, there was a transient increase in physical association of PLSCR1 with both the EGF receptor and the adapter protein Shc, as determined by immunoprecipitation and Western blotting. Confocal immunofluorescence analysis revealed that EGF initiates rapid internalization of both the EGF receptor and PLSCR1, with trafficking into both distinct and common endosomal pools. These data also suggested that whereas the EGF receptor is ultimately degraded, much of the endocytosed PLSCR1 is recycled to the cell surface within 3 h after EGF treatment. Consistent with this interpretation, Western blotting revealed neither ubiquitination nor proteolysis of PLSCR1 under these conditions, whereas the ubiquitination and degradation of the EGF receptor were readily confirmed. Finally, stimulation with EGF was also found to markedly increase the total cellular expression of PLSCR1, suggesting that in addition to its initial interactions with activated EGF receptor, PLSCR1 may also contribute to posttranscriptional effector pathway(s) mediating the cellular response to EGF.

Lipid rafts are specialized cholesterol-enriched membrane domains that participate in cellular signaling processes. Caveolae are related domains that become invaginated due to the presence of the structural protein, caveolin-1. In this paper, we use electrospray ionization mass spectrometry (ESI/MS) to quantitatively compare the phospholipids present in plasma membranes and nondetergent lipid rafts from caveolin-1-expressing and nonexpressing cells. Lipid rafts are enriched in cholesterol and sphingomyelin as compared to the plasma membrane fraction. Expression of caveolin-1 increases the amount of cholesterol recovered in the lipid raft fraction but does not affect the relative proportions of the various phospholipid classes. Surprisingly, ESI/MS demonstrated that lipid rafts are enriched in plasmenylethanolamines, particularly those containing arachidonic acid. While the total content of anionic phospholipids was similar in plasma membranes and nondetergent lipid rafts, the latter were highly enriched in phosphatidylserine but relatively depleted in phosphatidylinositol. Detergent-resistant membranes made from the same cells showed a higher cholesterol content than nondetergent lipid rafts but were depleted in anionic phospholipids. In addition, these detergent-resistant membranes were not enriched in arachidonic acid-containing ethanolamine plasmalogens. These data provide insight into the structure of lipid rafts and identify potential new roles for these domains in signal transduction.

A phosphatidylinositol kinase from A431 cells has been purified to near homogeneity. Purification was achieved through the use of a combination of chromatography steps including affinity elution of the enzyme from a heparin-agarose column with PI. Characterization of the (/sup 32/P)PIP formed by the purified PI kinase indicates that the enzyme phosphorylates the inositol on the 4-position and is therefore a phosphatidylinositol 4-kinase. The enzyme has a subunit weight of 55,000 as estimated by SDS gel electrophoresis and appears to be active as a monomer. Studies of the hydrodynamic properties of the enzyme indicate that the PI kinase binds substantial amounts of Triton X-100 and is actually present in detergent-containing solutions as a complex with a molecular weight of approximately 120,000. The K/sub m/ of the enzyme for PI is 16 ..mu..M and for ATP is 74 ..mu..M. The enzyme is inhibited by adenosine with an IC/sub 50/ of 100 ..mu..M. These properties are essentially identical with those of the membrane-bound PI kinase in A431 cells which is stimulated by EGF. The data therefore suggest that the EGF-stimulated PI kinase is a 55,000-Da monomer.

The insulin receptor has been shown to be a protein kinase which phosphorylates its substrates on tyrosine residues. To examine the acceptor specificity of affinity-purified insulin receptor/kinase, hydroxyamino acid containing analogues of the synthetic peptide substrate Arg-Arg-Leu-Ile-Glu-Asp-Ala-Glu-Tyr-Ala-Ala-Arg-Gly were prepared. Substitution of serine, threonine, or D-tyrosine for L-tyrosine completely ablated the acceptor activity of the synthetic peptides. These peptides, along with a phenylalanine-containing analogue, did serve as competitive inhibitors of the insulin receptor/kinase with apparent Ki values in the range of 2-4 mM. These data suggest that the insulin receptor/kinase is specific for tyrosine residues in its acceptor substrate and imply that serine phosphate or threonine phosphate present in receptor is due to phosphorylation by other protein kinases. The kinetics of the phosphorylation of the L-tyrosine-containing peptide were examined by using prephosphorylated insulin receptor/kinase. Prephosphorylation of the receptor was necessary to maximally activate the kinase and to linearize the initial velocity of the peptide phosphorylation reaction. The data obtained rule out a ping-pong mechanism and are consistent with a random-order rapid-equilibrium mechanism for the phosphorylation of this peptide substrate. Additional experiments demonstrated that the autophosphorylated insulin receptor was not able to transfer the preincorporated phosphate to the synthetic peptide substrate. Thus, the insulin receptor/kinase catalyzes the reaction via a mechanism that does not involve transfer of phosphate from a phosphotyrosine-containing enzyme intermediate.

Transglutaminase activity was detected in lysates of A431 cells, a human epidermal carcinoma cell line. Enzyme activity was increased 1.5-2.5-fold in lysates prepared from cells pretreated with epidermal growth factor (EGF) relative to untreated control cells. Half-maximal activation of the transglutaminase activity occurred at 3-5 nM EGF, a concentration in good agreement with the Kd for EGF binding to its receptor in these cells. The increase in transglutaminase activity could be detected as early as 2 min after the addition of EGF, with the maximal response attained by 30 min. The activation was not blocked by pretreatment of the cells with cycloheximide, suggesting that the increased activity was not the result of an induction of transglutaminase synthesis. Fractionation of A431 cell lysates by centrifugation at 100000g for 30 min demonstrated that 90% of the transglutaminase activity was present in the soluble fraction and that this soluble transglutaminase activity was increased after treatment of the cells with EGF. The demonstration that EGF acutely increases the activity of a soluble, intracellular transglutaminase defines a novel pathway of growth factor action and provides a useful model system for identifying and comparing the mechanism(s) by which growth factors activate soluble enzymes.