Requires Phosphatidylinositol 3-kinase Activity Research Articles

Hepatocyte growth factor switches orientation of polarity and mode of movement during morphogenesis of multicellular epithelial structures.

Epithelial cells form monolayers of polarized cells with apical and basolateral surfaces. Madin-Darby canine kidney epithelial cells transiently lose their apico-basolateral polarity and become motile by treatment with hepatocyte growth factor (HGF), which causes the monolayer to remodel into tubules. HGF induces cells to produce basolateral extensions. Cells then migrate out of the monolayer to produce chains of cells, which go on to form tubules. Herein, we have analyzed the molecular mechanisms underlying the production of extensions and chains. We find that cells switch from an apico-basolateral polarization in the extension stage to a migratory cell polarization when in chains. Extension formation requires phosphatidyl-inositol 3-kinase activity, whereas Rho kinase controls their number and length. Microtubule dynamics and cell division are required for the formation of chains, but not for extension formation. Cells in the monolayer divide with their spindle axis parallel to the monolayer. HGF causes the spindle axis to undergo a variable "seesaw" motion, so that a daughter cells can apparently leave the monolayer to initiate a chain. Our results demonstrate the power of direct observation in investigating how individual cell behaviors, such as polarization, movement, and division are coordinated in the very complex process of producing multicellular structures.

FcγR-Mediated Phagocytosis Stimulates Localized Pinocytosis in Human Neutrophils

Engulfment of IgG-coated particles by neutrophils and macrophages is an essential component of the innate immune response. This process, known as phagocytosis, is triggered by clustering of FcgammaR at sites where leukocytes make contact with the opsonized particles. We found that phagocytosis is accompanied by a burst of fluid phase pinocytosis, which is largely restricted to the immediate vicinity of the phagosomal cup. FcgammaR-induced pinocytosis preceded and appeared to be independent of phagosomal sealing. Accordingly, fluid phase uptake was accentuated by actin depolymerization, which precludes phagocytosis. Stimulation of pinocytosis required phosphatidylinositol 3-kinase activity and was eliminated when changes in the cytosolic free Ca(2+) concentration were prevented. Because stimulation of FcgammaR also induces secretion, which is similarly calcium and phosphatidylinositol 3-kinase dependent, we studied the possible relationship between these events. Neutrophil fragments devoid of secretory granules (cytoplasts) were prepared by sedimentation through Ficoll gradients. Cytoplasts could perform FcgammaR-mediated phagocytosis, which was not accompanied by activation of pinocytosis. This observation suggests that granule exocytosis is required for stimulation of pinocytosis. Analysis of the cytosolic Ca(2+) dependence of secretion and pinocytosis suggests that primary (lysosomal) granule exocytosis is the main determinant of pinocytosis during FcgammaR stimulation. Importantly, primary granules are secreted in a polarized fashion near forming phagosomes. Focal pinocytosis during particle engulfment may contribute to Ag processing and presentation and/or to retrieval of components of the secretory machinery. Alternatively, it may represent an early event in the remodeling of the phagosomal membrane, leading to phagosomal maturation.

Pitx2a expression alters actin-myosin cytoskeleton and migration of HeLa cells through Rho GTPase signaling.

We ectopically expressed the transcription factor Pitx2a, one of the Pitx2 isoforms, in HeLa cells by using a tetracycline-inducible expression system and examined whether Pitx2a was capable of modulating Rho GTPase signaling and altering the cell's cytoskeleton. Ectopic expression of Pitx2a induced actin-myosin reorganization, leading to increased cell spreading, suppression of cell migration, and the strengthening of cell-cell adhesion, marked by the accumulation and localization of beta-catenin and N-cadherin to the sites of cell-cell contacts. Moreover, Pitx2a expression resulted in activation of the Rho GTPases Rac1 and RhoA, and the dominant negative Rac1 mutant N17Rac1 inhibited cell spreading and disrupted localization of beta-catenin to the sites of cell-cell contacts. Both reorganization of actin-myosin and cell spreading require phosphatidylinositol 3-kinase activity, which is also necessary for activation of the Rho GTPase proteins. Pitx2a induced the expression of Trio, a guanine nucleotide exchange factor for Rac1 and RhoA, which preceded cell spreading, and the expression of Trio protein was down-regulated after the changes in cell spreading and cell morphology were initiated. In addition, Pitx2a also induces cell cycle arrest at G0/G1, most likely due to the accumulation of the tumor suppressor proteins p53 and p21. Our data indicate that the transcriptional activities initiated in the nucleus by Pitx2a result in profound changes in HeLa cell morphology, migration, and proliferation.

Epstein-Barr virus immediate-early protein BRLF1 induces the lytic form of viral replication through a mechanism involving phosphatidylinositol-3 kinase activation.

Expression of the Epstein-Barr virus (EBV) immediate-early (IE) protein BRLF1 induces the lytic form of viral replication in most EBV-positive cell lines. BRLF1 is a transcriptional activator that binds directly to a GC-rich motif present in some EBV lytic gene promoters. However, BRLF1 activates transcription of the other IE protein, BZLF1, through an indirect mechanism which we previously showed to require activation of the stress mitogen-activated protein kinases. Here we demonstrate that BRLF1 activates phosphatidylinositol-3 (PI3) kinase signaling in host cells. We show that the specific PI3 kinase inhibitor, LY294002, completely abrogates the ability of a BRLF1 adenovirus vector to induce the lytic form of EBV infection, while not affecting lytic infection induced by a BZLF1 adenovirus vector. Furthermore, we demonstrate that the requirement for PI3 kinase activation in BRLF1-induced transcriptional activation is promoter dependent. BRLF1 activation of the SM early promoter (which occurs through a direct binding mechanism) does not require PI3 kinase activation, whereas activation of the IE BZLF1 and early BMRF1 promoters requires PI3 kinase activation. Thus, there are clearly two separate mechanisms by which BRLF1 induces transcriptional activation.

Endosomal Localization and Receptor Dynamics Determine Tyrosine Phosphorylation of Hepatocyte Growth Factor-Regulated Tyrosine Kinase Substrate

Hepatocyte growth factor-regulated tyrosine kinase substrate (Hrs) is a prominent substrate for activated tyrosine kinase receptors that has been proposed to play a role in endosomal membrane trafficking. The protein contains a FYVE domain, which specifically binds to the lipid phosphatidylinositol (PI) 3-phosphate (PI 3-P). We show that this interaction is required both for correct localization of the protein to endosomes that only partially coincides with early endosomal autoantigen 1 and for efficient tyrosine phosphorylation of the protein in response to epidermal growth factor stimulation. Treatment with wortmannin reveals that Hrs phosphorylation also requires PI 3-kinase activity, which is necessary to generate the PI 3-P required for localization. We have used both hypertonic media and expression of a dominant-negative form of dynamin (K44A) to inhibit endocytosis; under which conditions, receptor stimulation fails to elicit phosphorylation of Hrs. Our results provide a clear example of the coupling of a signal transduction pathway to endocytosis, from which we propose that activated receptor (or associated factor) must be delivered to the appropriate endocytic compartment in order for Hrs phosphorylation to occur.

Neurotrophic factor protection against ethanol toxicity in rat cerebellar granule cell cultures requires phosphatidylinositol 3-kinase activation

Neonatal rat cerebellar granule cells were used to assess the possible role of the phosphatidylinositol 3-kinase (PI3-K) signaling pathway in the neuroprotective effects of neurotrophic factors against ethanol toxicity. Culture conditions included medium with ethanol (400 and 600 mg/dl), nerve growth factor (NGF) or brain-derived neurotrophic factor (BDNF), ethanol+NGF or BDNF, the PI3-K inhibitor wortmannin (10 or 100 μM), and wortmannin+ethanol+NGF or BDNF. Neuronal survival was determined via the MTT assay. The results indicated that both NGF and BDNF ameliorate ethanol neurotoxicity, and wortmannin abolished this effect, except at the higher ethanol concentration combined with the lower wortmannin level. These data strongly implicate the PI3-K pathway in growth factor protection against ethanol neurotoxicity.

Inhibition of phosphatidylinositol 3-kinase activity blocks cellular differentiation mediated by glial cell line-derived neurotrophic factor in dopaminergic neurons.

Glial cell line-derived neurotrophic factor (GDNF) is a potent survival factor for midbrain dopaminergic neurons. To begin to understand the intracellular signaling pathways used by GDNF, we investigated the role of phosphatidylinositol 3-kinase activity in GDNF-stimulated cellular function and differentiation of dopaminergic neurons. We found that treatment of dopaminergic neuron cultures with 10 ng/ml GDNF induced maximal levels of Ret phosphorylation and produced a profound increase in phosphatidylinositol 3-kinase activity, as measured by western blot analysis and lipid kinase assays. Treatment with 1 microM 2-(4-morpholinyl)-8-phenylchromone (LY294002) or 100 nM wortmannin, two distinct and potent inhibitors of phosphatidylinositol 3-kinase activity, completely inhibited GDNF-induced phosphatidylinositol 3-kinase activation, but did not affect Ret phosphorylation. Furthermore, we examined specific biological functions of dopaminergic neurons: dopamine uptake activity and morphological differentiation of tyrosine hydroxylase-immunoreactive neurons. GDNF significantly increased dopamine uptake activity and promoted robust morphological differentiation. Treatment with LY294002 completely abolished the GDNF-induced increases of dopamine uptake and morphological differentiation of tyrosine hydroxylase-immunoreactive neurons. Our findings show that GDNF-induced differentiation of dopaminergic neurons requires phosphatidylinositol 3-kinase activation.

Src-induced activation of inducible T cell kinase (ITK) requires phosphatidylinositol 3-kinase activity and the Pleckstrin homology domain of inducible T cell kinase.

The Tec family of tyrosine kinases are involved in signals emanating from cytokine receptors, antigen receptors, and other lymphoid cell surface receptors. One family member, ITK (inducible T cell kinase), is involved in T cell activation and can be activated by the T cell receptor and the CD28 cell surface receptor. This stimulation of tyrosine phosphorylation and activation of ITK can be mimicked by the Src family kinase Lck. We have explored the mechanism of this requirement for Src family kinases in the activation of ITK. We found that coexpression of ITK and Src results in increased membrane association, tyrosine phosphorylation and activation of ITK, which could be blocked by inhibitors of the lipid kinase phosphatidylinositol 3-kinase (PI 3-kinase) as well as overexpression of the p85 subunit of PI 3-kinase. Removal of the Pleckstrin homology domain (PH) of ITK resulted in a kinase that could no longer be induced to localize to the membrane or be activated by Src. The PH of ITK was also able to bind inositol phosphates phosphorylated at the D3 position. Membrane targeting of ITK without the PH recovered its ability to be activated by Src. These results suggest that ITK can be activated by a combination of Src and PI 3-kinase.

Rab7 Regulates Transport from Early to Late Endocytic Compartments in Xenopus Oocytes

Rab7 has been shown to localize to late endosomes and to mediate transport from early to late endosome/lysosome in mammalian cells and in yeast. We developed a novel assay to quantify transport from early to late endosomes using the Xenopus oocyte. Oocytes were pulsed with avidin after which the oocytes were incubated to allow avidin transport to a late compartment. The oocytes were then allowed to internalize biotin-horseradish peroxidase (HRP). The oocytes were then injected with test proteins and incubated further to allow transport of biotin-HRP from early endosomes to late endosomal/lysosomal compartments. Transport was quantified by assessing the formation of HRP-biotin-avidin complexes. Injection of Rab7:wild-type (WT) and Rab7:Q67L, a GTPase defective mutant, stimulated transport. Rab5:WT had no effect. Rab7:WT-stimulated transport was inhibited by nocodazole, suggesting a role for intact microtubules. Wortmannin, a phosphatidylinositol 3-kinase inhibitor, blocked Rab7:WT-stimulated transport, but Rab7:Q67L-stimulated transport was unaffected by the drug. Rab7:Q67L is constitutively activated and may not require phosphatidylinositol 3-kinase activity for activation. Rab7-stimulated transport requires N-ethylmaleimide-sensitive factor (NSF) activity as transport was blocked by N-ethylmaleimide and ATPase defective NSF mutants. Our results indicate that sequentially acting endocytic Rab GTPases utilize similar factors although their modes of action may be different.

Insulin induces a change in Rab5 subcellular localization in adipocytes independently of phosphatidylinositol 3-kinase activation.

We investigated whether Rab5, a small guanosine triphosphatase that regulates early endocytic transport in different cell types is involved in the insulin-regulated endocytic pathways in adipocytes. Rab5 was detected in freshly isolated adipocytes and 3T3-L1 adipocytes, but its expression level was not markedly increased with adipocyte differentiation. After subcellular fractionation of adipocytes incubated in the absence of insulin, Rab5 was found to be abundant in plasma membrane and cytosol, but was also present in high and low density microsomes. This subcellular distribution was compatible with a role in early endocytosis. When cells were incubated with insulin, the concentration of Rab5 decreased by about 50% in the internal compartments. In contrast to Rab4, which also leaves the low density microsomes in response to insulin, Rab5 was not found in Glut4-containing vesicles purified by immunoadsorption on antibodies to Glut4. When adipocytes were treated with wortmannin, an inhibitor of phosphatidylinositol 3-kinase, the effect of insulin on Rab5 movement was not affected, whereas the insulin-induced movements of Rab4 and Glut4 were abolished. In parallel, wortmannin inhibited the increase in horseradish peroxidase uptake induced by insulin, an index of fluid phase endocytosis, but did not prevent the endocytosis of the glucose transporters. As a whole, our results suggest that Rab5 is not involved in insulin-stimulated Glut4 exocytosis. These results are compatible with the postulated role of Rab5 in the endocytotic pathway, at a step that does not require phosphatidyl-inositol 3-kinase activation.

Insulin induces a change in Rab5 subcellular localization in adipocytes independently of phosphatidylinositol 3-kinase activation

We investigated whether Rab5, a small guanosine triphosphatase that regulates early endocytic transport in different cell types is involved in the insulin-regulated endocytic pathways in adipocytes. Rab5 was detected in freshly isolated adipocytes and 3T3-L1 adipocytes, but its expression level was not markedly increased with adipocyte differentiation. After subcellular fractionation of adipocytes incubated in the absence of insulin, Rab5 was found to be abundant in plasma membrane and cytosol, but was also present in high and low density microsomes. This subcellular distribution was compatible with a role in early endocytosis. When cells were incubated with insulin, the concentration of Rab5 decreased by about 50% in the internal compartments. In contrast to Rab4, which also leaves the low density microsomes in response to insulin, Rab5 was not found in Glut4-containing vesicles purified by immunoadsorption on antibodies to Glut4. When adipocytes were treated with wortmannin, an inhibitor of phosphatidylinositol 3-kinase, the effect of insulin on Rab5 movement was not affected, whereas the insulin-induced movements of Rab4 and Glut4 were abolished. In parallel, wortmannin inhibited the increase in horseradish peroxidase uptake induced by insulin, an index of fluid phase endocytosis, but did not prevent the endocytosis of the glucose transporters. As a whole, our results suggest that Rab5 is not involved in insulin-stimulated Glut4 exocytosis. These results are compatible with the postulated role of Rab5 in the endocytotic pathway, at a step that does not require phosphatidyl-inositol 3-kinase activation.

The glucose transporters of skeletal muscle

Abstract Glucose transport into skeletal muscle occurs through the GLUT1 and GLUT4 glucose transporters. Muscle cells in culture also express the GLUT3 fetal muscle/neuronal type transporter. In skeletal muscle, the GLUT1 transporter is restricted to the cell surface, while the more abundant GLUT4 transporter is largely sequestered intracellularly from where it is rapidly translocated to the cell surface in response to insulin, exercise or hypoxia. The insulin effect has been documented by subcellular fractionation of rat, mouse and human muscle, and has been confirmed quantitatively by photolabelling of the surface transporters and qualitatively by immunoelectron microscopy. In L6 myotubes in culture, the GLUT1 and GLUT3 transporters are mostly located at the cell surface but a fraction resides intracellularly, whereas the GLUT4 transporter is distributed evenly between the surface and the intracellular location. Immunopurified intracellular GLUT4 vesicles from these cells do not contain appreciable amounts of GLUT1 or GLUT3 transporters, although all three transporters respond to insulin by translocating to the plasma membrane. The glucose transporter translocation induced by insulin in skeletal muscle and L6 myotubes requires phosphatidylinositol 3-kinase activity, as does the maintenance of the basal amount of transporters at the plasma membrane. Two different phosphatidylinositol 3-kinase activities may control basal and insulin-dependent transport. In contrast, the stimulation of glucose transport induced by exercise or hypoxia is independent of this enzymatic activity. In both L6 myotubes and mature skeletal muscle, the GLUT4-containing vesicle contains synaptobrevin II/VAMP-2 and cellubrevin. These proteins also redistribute in response to insulin, and may be required for correct vesicle docking and/or fusion with the plasma membrane.