| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLE |
Department of Physiology, Wayne State University School of Medicine, Detroit, Michigan 48201
Address all correspondence and requests for reprints to: Assia Shisheva, Department of Physiology, Wayne State University School of Medicine, 540 East Canfield, Detroit, Michigan 48201. E-mail: ashishev{at}med.wayne.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
K, introduced into the preadipocyte stage profoundly delayed the hormone-induced adipogenesis. Next, insulin-induced mitogenesis was markedly inhibited in HEK293 stable cell lines, inducibly expressing the dominant-negative kinase-dead PIKfyveK1831E mutant but not in cells expressing PIKfyveWT. Similarly, expression of the dominant negative mutants PIKfyveK1831E or PIKfyveK strongly inhibited insulin-stimulated translocation of GLUT4 in 3T3-L1 adipocytes, or GLUT1-mediated glucose uptake in Chinese hamster ovary T cells expressing the human insulin receptor. Expression of PIKfyveK and PIKfyveWT in Chinese hamster ovary T cells decreased or increased, respectively, insulin-stimulated Akt phosphorylation at Ser473 but not at Thr308. Furthermore, a powerful inhibition of PIKfyve was documented at a very low concentration (ID50 = 6 µM) of the cell-permeable kinase inhibitor curcumin. When introduced into 3T3-L1 adipocytes, curcumin markedly inhibited insulin-induced GLUT4 translocation and glucose transport. Together these data indicate that PIKfyve enzymatic activity functions as a positive regulatory intermediate in insulin acute and long-term biological responses and identify Ser473 in Akt as one potential PIKfyve downstream target. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mouse PIKfyve is a mammalian protein of 2052 amino acids, cloned by our laboratory in a search for transcripts that, similar to the fat/muscle-specific glucose transporter GLUT4, is expressed in a tissue specific manner (10, 11). It is encoded by a single-copy gene, transcribed in at least two close-in-size splice variants, with an mRNA size of 9 kb. PIKfyve harbors several evolutionarily conserved domains, including a Zn2+/PtdIns 3-P binding FYVE finger, a DEP domain (with unknown function found in signaling proteins containing PH domains Disheveled, Egl-10, and Pleckstrin), a chaperonin-like region, and a catalytic domain found in PI5Ks and PIP4Ks. PIKfyve is an active 5'-PI kinase, synthesizing PtdIns 5-P and PtdIns 3,5-P2 in vitro and in vivo, and its activity is likely regulated intracellularly by autophosphorylation (12, 13, 14, 15). Although PIKfyve RNA was found to be abundant in insulin sensitive cells, all mammalian cells tested thus far express the protein to some degree, as judged by detecting both the protein and the enzyme activity in lysates of different cell types. In fact, PIKfyve belongs to an evolutionarily ancient gene family represented by a single-copy gene, present possibly in all eukaryotic cells (15). The cellular functions of PIKfyve have begun to be unraveled by the observation that kinase-defective mutants exert a dominant- negative effect. These mutants drastically altered endomembrane cell morphology and induced a progressive dilation of the PIKfyve-containing vesicles concomitant with a massive vacuolation of membranes of endocytic origin (14). Because these phenotypic changes were corrected on subsequent coexpression of PIKfyveWT or exogenously added PtdIns 3,5-P2, the central role of PIKfyve enzymatic activity, and particularly PtdIns 3.5-P2 production, in maintaining cell morphology and endocytic membrane homeostasis was directly demonstrated (14, 16).
Several observations point to the role of PIKfyve in insulin action. Thus, PIKfyve mRNA increases with the differentiation of 3T3-L1 or L6 fibroblasts in insulin-sensitive cells (11). Furthermore, acute insulin action into 3T3-L1 adipocytes causes both PIKfyve membrane recruitment and its phosphorylation, possibly as a way to tether cytosolic PIKfyve to membranes (17). Next, in 3T3-L1 adipocytes a population of class IA PI3Ks associates with PIKfyve and is activated by cell stimulation with insulin but not with platelet-derived or epithelial growth factors (18). Finally, p85/p110 PI3K (class IA) was found to not only coimmunoprecipitate with PIKfyve but also to cofractionate with it upon equilibrium density sedimentation of 3T3-L1 adipocyte intracellular membranes, indicating that the two kinases likely associate in the cellular context, possibly for concerted PtdIns 3,5-P2 production (17). Together, these data are consistent with the hypothesis that PIKfyve or the products of its enzymatic activity may be part of the insulin signal transduction network and, thus, may play a role in insulin-regulated cellular responses in insulin-sensitive cells. Therefore, we have tested several classical insulin-regulated responses, both acute and chronic, in cells expressing high levels of either active PIKfyve enzyme or kinase-dead mutants, shown to act in a dominant-negative fashion (14, 16). We have observed in all cases that, whereas PIKfyveWT exerts no significant to slightly stimulatory effect, the kinase-dead PIKfyve mutants markedly abrogated insulin-regulated actions. These results indicate that PIKfyve enzymatic activity acts as a positive, stimulatory signaling intermediate in several insulin-regulated biological responses.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Generation of stable cell lines
Stably transfected doxycycline-inducible (Tet-On) cell lines were generated following the Tet-Off/Tet-On gene Expression Systems manual (CLONTECH Laboratories, Inc., Palo Alto, CA). Briefly, PIKfyveWT or PIKfyveK1831E cDNAs (released by XbaI-SalI from the respective pBluescript IISK+ plasmids (11, 13) together with a hemagglutinin (HA)-encoding adapter (flanked with BamHI and XbaI restriction sites) were cloned into the BamHI-SalI site of the pTRE2hyg vector. The expected organization of the constructs was confirmed by restriction mapping. The pTRE2hygHA-PIKfyveWT and pTRE2hygHA-PIKfyveK1831E cDNAs, were linearized by SalI and used to transfect an HEK293 Tet-On cell line (CLONTECH Laboratories, Inc.) by Lipofectamine as a transfection reagent. The transfected cells were selected by hygromycin treatment at 125 µg/ml, a concentration that was optimized to eliminate all susceptible cells after 5–7 d. Individual cell clonal lines were isolated by the help of cloning cylinders, propagated, and then probed for a doxycycline-inducible expression of recombinant PIKfyve proteins by Western blot analysis with anti-HA polyclonal antibodies (a kind gift of Mike Czech, University of Massachusetts Medical Center).
Preparation of recombinant adenovirus and cell infection
Generation of recombinant adenoviruses, expressing HA-tagged PIKfyvesWTand green fluorescent protein (GFP) or GFP alone by the AdEasy system (19), was previously described (14). Recombinant adenovirus expressing a kinase-dead deletion mutant of PIKfyve (PIKfyveK), in which the entire catalytic domain (amino acids 1812–2052) has been eliminated, was generated in a similar manner. Briefly, the C-terminal portion of the vector pAdTrackCMV-PIKfyveWT (14) was released with Kpn1-Sal1 and replaced with the Kpn1-Sal1 fragment of pBluescript HA-PIKfyvesK (12). The resultant construct was linearized with PmeI and cotransformed with pAdEasy-1 adenoviral backbone plasmid into Escherichia coli BJ5183 cells. Selected recombinants were confirmed by restriction mapping and then linearized with PacI and used to transfect an HEK293 adenovirus packaging cell line by Lipofectamine (Invitrogen, Carlsbad, CA). Two weeks after transfection the cells were harvested. The viruses were extracted by freeze-thaw and subsequent centrifugation. This viral extract was used for further viral propagation in HEK293 cells. Viral stocks were purified by ultracentrifugation in two discontinuous CsCl2 gradients and subsequent passage through a Nap 10 column (Sephadex G25, Amersham Biosciences, Piscataway, NJ). Purified viral stocks were titrated and the lowest dilution resulting in 100% infection of HEK293 cells 18 h after infection (monitored by the GFP signals) was defined as multiplicity of infection (MOI) = 1. The adenovirus-mediated protein expression and lack of enzymatic activity of PIKfyvesK were confirmed by immunoblotting and lipid kinase assay. For adenovirus-mediated gene transfer, 3T3-L1 fibroblasts or CHO-T cells were infected with various adenoviral vectors at MOI indicated in the figure legends.
Transient expression by electroporation
Differentiated 3T3-L1 adipocytes were transfected with the indicated GFP-based or HA-based PIKfyve cDNA constructs reported elsewhere (11, 12, 13, 14), separately or in combination with GFP-GLUT4 (a kind gift of Jeff Pessin, University of Iowa, Iowa City, IA) using the electroporation method described previously (20). Briefly, cells at d 7 of the differentiation program, harvested from a 150-mm dish were electroporated with the indicated cDNA, banded three times on CsCl2 gradients. Forty-eight hours after transfection, the cells were serum starved and then stimulated or not stimulated with insulin (100 nM) for 20 min and processed further for fluorescence microscopy analysis.
Fluorescence microscopy analysis of GLUT4 translocation
Subsequent to fixation in 4% formaldehyde and washings, cotransfected cells were processed for immunofluorescence microscopy analysis using anti-HA antibodies as described elsewhere (14, 16). Coverslips were mounted on slides using the Slow Fade antifade kit (Molecular Probes, Inc., Eugene, OR). Fluorescence analysis was performed in an Eclipse TE 200 inverted fluorescence microscope (Nikon, Melville, NY) using a Hoffman modulation contrast system (Modulation Optics, New York, NY) with a 40x objective and a standard green fluorescence filter for GFP. Images were captured with a SPOT RT slider charge-coupled device camera (Diagnostic Instruments, Sterling Heights, MI) mounted on the microscope. Science Lab software (Fuji Photo Film Co. Ltd., Tokyo, Japan) was used to quantify the fluorescence intensity.
3T3-L1 adipocyte subcellular fractionation and GLUT4 translocation
The 3T3-L1 adipocytes were serum starved for 16 h in DMEM supplemented with 0.5% BSA. The cells were preincubated for 30 min at 37 C with or without 100 µM curcumin, then stimulated or not, with insulin (100 nM) for an additional 20 min and subjected to subcellular fractionation as described previously (17). Briefly, cells were washed twice with PBS, once in the homogenization buffer [HES buffer: 20 mM HEPES-HCl, pH 7.5; 1 mM EDTA; 255 mM sucrose containing 1x protease inhibitor cocktail (1 mM phenylmethylsulfonylfluoride, 5 µg/ml leupeptin, 5 µg/ml aprotinin, 1 µg/ml pepstatin, and 1 mM benzamidine)] at 25 C and then scraped at 4 C in HES buffer, supplemented with the inhibitor cocktail. Cells were homogenized by a motor-driven homogenizer with 10 strokes at 225 rpm and 5 strokes at 800 rpm. Subcellular fractionation was performed with an SS-34 rotor (Sorvall Instrument Division, Newton, CT) in the first spin and a TLA 100.3 rotor (Beckman Instruments Inc., Fullerton, CA) in the next spins to obtain intracellular membrane (IM), plasma membrane (PM), and cytosolic fractions. Pellets were resuspended in HES buffer supplemented with the inhibitor cocktail to a protein concentration of approximately 2 mg/ml (bicinchoninic acid protein assay kit, Piece, Rockford, IL). GLUT4 levels in the fractions were detected by immunoblotting with polyclonal anti-GLUT4 antibodies (kind gift of Mike Czech, University of Massachusetts Medical Center).
Immunoblotting
Immunoblotting with the indicated polyclonal antibodies was performed subsequent to protein separation by SDS-PAGE and electrotransfer onto nitrocellulose membranes as previously described (12). Renaissance chemiluminescence kit (NEN Life Science Products DuPont, Boston, MA) was used to detect the horseradish-peroxidase-bound secondary antibodies. Protein levels or the extent of protein phosphorylation on the blots were quantified with a laser densitometer (Molecular Dynamics, Inc., Sunnyvale, CA) by area integration scanning. Several exposures of each blot were quantified to assure that the chemiluminescence exposures were within the linear range of the film.
Akt activation
Akt activation was measured by the induction of phosphorylation on amino acids Thr308 or Ser473 as determined by immunoblotting with selective antiphopshoThr308 or antiphosphoSer473 antibodies (New England Biolabs, Inc., Beverly, MA), respectively. Briefly, CHO-T cells stably expressing human IR, seeded onto 60-mm dishes were transduced with recombinant adenoviruses expressing GFP alone (empty virus) or in combination with kinase-dead PIKfyveK or PIKfyveWT. Forty-eight hours after infection, the cells were serum starved for 3 h and then left untreated or stimulated with insulin (100 nM) for 10 min at 37 C, washed three times in ice-cold PBS and homogenized in HES buffer containing 1x protease inhibitors and 1x phosphatase inhibitors (50 mM NaF, 2 mM sodium metavanadate, 10 mM sodium pyrophosphate, and 25 mM ß-glycerophosphate) by passing six times through a 22-gauge needle. Nuclei-free total membranes and cytosols were obtained as described previously (21) and the protein contents determined as described above. In some experiments, total cell lysates, prepared in RIPA buffer (50 mM Tris HCl, pH 8.0; 150 mM NaCl; 1% Nonidet P-40; and 0.5% sodium deoxycholate) containing the above inhibitors, were used instead of the subcellular fractions as specified in the figure legends. Equal amounts of protein were resolved on a gradient SDS-PAGE gel (3–15%, duplicate gels). After electrotransfer, the bottom parts of the duplicate nitrocellulose membrane were probed with either antiphosphoSer473 or antiphosphoThr308, and the upper parts were probed for expression of the indicated PIKfyve constructs with anti-HA antibodies.
DNA synthesis
For 3'-bromo-5'-deoxyuridine (BrdU) incorporation, parental Tet-On HEK293 cells and the selected stably transfected cell lines inducibly expressing PIKfyveWT (clone 9) or PIKfyveK1831E (clone 3) were seeded onto cover slips in 35-mm dishes at 50–60% confluency in the presence or absence of doxycycline (1 µg/ml). Cells were serum deprived for 36 h. Mitogenesis was then stimulated by the addition of insulin (100 nM) or FBS (20%) for 16 h. BrdU (Sigma, St. Louis, MO) was added at 20 µM during the last 6 h. At the end of the incubation, the cells were washed twice with PBS and fixed in 3.7% formaldehyde for 10 min. DNA was denatured with 1 N HCl for 10 min and the medium neutralized by two washings in 0.1 M sodium borate, pH 8.3, for 15 min. Cells were then permeabilized with 0.5% Triton X-100 in PBS-1% FBS, and incubated with a monoclonal anti-BrdU antibody (1:100 dilution, BD PharMingen, San Diego, CA) for 90 min. After four washes in permeabilization buffer, the cells were exposed to fluorescein isothiocyanate-conjugated antimouse secondary antibody (1:1000 dilution, BioSource International, Camarillo, CA) for 30 min. Washed cells were postfixed in formaldehyde, rewashed in PBS, and observed by fluorescence and phase contrast microscopy (Eclipse TE 200 inverted fluorescence microscope, Nikon). To quantify the results, at least 200 cells/dish were typically counted and the percent of nuclei incorporating BrdU calculated.
[3H]Thymidine incorporation was performed essentially as described previously (22). Briefly, the parental Tet-On HEK293 or the stable clonal lines inducibly expressing PIKfyveWT or PIKfyveK1831E were seeded on 12-well collagen IV precoated plates to promote attachment. Subsequent to serum starvation in DMEM medium, containing 0.1% BSA (20 h) and protein induction with doxycycline, cells were stimulated or not with insulin (100 nM) for an additional 18 h. [Methyl-3H]thymidine (NEN Life Science Products DuPont) was present during the last 5 h of incubation. Cells were then washed two times with ice-cold PBS, precipitated with 10% trichloroacetic acid for 1 h at 4 C, and rewashed in PBS. Cells were then solubilized in 0.2 N NaOH-0.1% sodium dodecyl sulfate for 1 h at 37 C. Incorporated radioactivity was quantified by liquid scintillation subsequent to neutralization of the cell lysates with 2 M Tris-HCl, pH 6.8, and the addition of scintillation cocktail.
Oil red O staining of 3T3-L1 adipocytes
At the indicated day of the differentiation program, cells were washed twice with PBS and then fixed for 1 h in 10% formaldehyde in PBS. After washes in water (three times, 5 min each) cells were incubated for 2 h at 25 C with oil red O solution, prepared as described previously (23). Cells were then washed three times in water (1 h each) and photographed with Zeiss (Oberkochen, Germany) confocal microscope using a 63/1.4 oil immersion lens.
Glucose transport
Glucose transport was determined by measuring 2-deoxyglucose (2DG) uptake as previously described (24). Briefly, following several washes in PBS, cells were incubated in Krebs Ringers Henseleit buffer (NaCl, 120 mM; KCl, 6 mM; Mg2SO4, 1.2 mM; CaCl2.2H20, 1 mM; Na2HPO4, 0.6 mM; NaH2PO4, 0.4 mM; and HEPES, 30 mM, pH 7.4), containing 0.5% BSA and 2 mM sodium pyruvate for 3 h. Cells were then washed and incubated for 20 min with or without insulin (100 nM). 2-[3H]deoxy-D-glucose (NEN Life Science Products DuPont) was added to a final concentration of 100 µM (25 µCi per 35-mm-diameter well) for 5 min at 37 C. After extensive washing in Krebs Ringers Henseleit, cells were lysed in 1% Triton X-100. Aliquots of the solubilized cells were used for liquid scintillation counting and determination of the protein concentration. Nonspecific glucose uptake was measured in the presence of 20 mM cytochalasin B and was subtracted for each determination. All values were normalized for protein content.
Lipid kinase assay and thin-layer chromatography (TLC) resolution of lipid products
PIKfyve lipid kinase activity was measured in the anti-HA or anti-PIKfyve immunoprecipitates as described previously (13). Briefly, cell lysates from treated or nontreated cells were immunoprecipitated with anti-HA or anti-PIKfyve antibodies (R7069, characterized elsewhere) (12, 13) as specified in the figure legends. Immunoprecipitates immobilized on protein A-Sepharose beads were washed once with RIPA buffer; twice with 50 mM HEPES (pH 7.4), 1 mM EDTA, and 150 mM NaCl; thrice with 100 mM Tris-HCl (pH 7.5), 500 mM LiCl; twice with 10 mM Tris-HCl (pH 7.5), 100 mM NaCl, 1 mM EDTA; and twice with assay buffer [25 mM HEPES (pH 7.4), 120 mM NaCl, 5 mM ß-glycerophosphate, 1 mM dithiothreitol, 2.5 mM MgCl2 and 2.5 mM MnCl2]. Beads were preincubated for 10 min at 37 C with the indicated concentration of curcumin, and then the kinase reaction was carried out for 15 min at 37 C in the assay buffer supplemented with 50 µM ATP, [
-32P]ATP (12.5 µCi), and 100 µM PtdIns. Lipids were extracted, applied on TLC plates (PE SIL G, 250 µm, Whatman, Clifton, NJ) and separated by a chromatographic solvent system. The radioactive products were detected by autoradiography and quantified by laser densitometry (Molecular Dynamics, Inc.) and radioactive counting of the silica scrapings.
Data analysis
Statistical analysis was performed using one-way ANOVA or t test with P less than 0.05 considered statistically significant. Data are presented as mean ± SE, the number of experiments being indicated in the figure legend.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, PIKfyve protein levels progressively increased after induction of differentiation, the increment being evident as early as d 2 following initiation of the differentiation program. When normalized for equal protein amounts, 5 ± 0.6-fold higher levels of the immunoreactive PIKfyve were estimated in lysates derived from adipocytes at d 8 of the differentiation program vs. those obtained from preadipocytes (Fig. 1B). Analysis of PIKfyve lipid kinase activity revealed a commensurate increase of the PIKfyve lipid products with the increase of PIKfyve protein levels (data not shown). Together these results indicate that the transition into the adipocyte phenotype is associated with the up-regulation of an active PIKfyve enzyme.
|
, CCAAT/enhancer-binding proteins, and 
-adducin/sterol regulatory element-binding protein 1 (25, 26). These transcription factors coordinate the expression of genes, including those of activator protein-2 and adipsin, which also contribute to the creation and maintenance of the adipocyte phenotype. Because PIKfyve gene expression was also up-regulated during adipocyte differentiation, we examined whether PIKfyve enzymatic activity is among the factors regulating the complex mechanisms that control adipocyte differentiation. To test this, we transduced confluent monolayers of 3T3-L1 fibroblasts with the empty virus or adenoviruses encoding PIKfyveWT or a kinase-dead PIKfyve truncated mutant (PIKfyveK) and monitored the appearance of the adipocyte phenotype induced in response to adipogenic hormones. Several kinase-dead point or double mutants have been shown previously to act in a dominant-negative fashion (14, 16), implying that PIKfyveK will have a similar dominant-negative effect. Two days after infection, cells were exposed to the standard differentiation protocol consisting of 3-d exposure to FBS media supplemented with insulin/3-isobutyl-1-methylxanthine/dexamethasone, followed by 2-d exposure to FBS media containing insulin alone. The differentiation was evaluated by the accumulation of lipid droplets, as determined by live light microscopy observations or oil red O staining after fixation. As illustrated in Fig. 2A, differentiation of cells infected with the empty virus was indistinguishable from that of control noninfected cells. Intriguingly, cells infected with the PIKfyveK virus delayed the accumulation of lipid droplets. Thus, although lipid droplets were documented in 30% of the control cells on d 4 of the differentiation program, they were not detectable in cells expressing PIKfyveK (Fig. 2A). These cells, however, retracted long processes and acquired a round shape as the control cells infected with the empty virus at the same MOI as exemplified on the fluorescence microscopy image presented in Fig. 2B. At d 8, the lipid accumulation in cells expressing PIKfyveK was only 50% of that observed in the control cells (Fig. 2A). In contrast, cells infected with the PIKfyveWT showed a reproducible tendency to accelerate the lipid accumulation, compared with cells infected with the empty virus (Fig. 2A). These cells also appeared to accumulate more lipids, as could be seen by the oil red O staining presented in Fig. 2B. However, although the presence of high-level PIKfyveWT promoted lipid accumulation, PIKfyve alone was unable to induce spontaneous differentiation in the absence of adipogenic hormones. It should be pointed out that expressed PIKfyveK negatively affected adipogenesis at the relatively early stages of the differentiation program. Twelve to 14 d following hormonal induction, the percentage of PIKfyveK-infected cells that accumulated lipid droplets was indistinguishable from that of cells infected with the control virus (>90%). Together these results indicate that PIKfyve enzymatic activity exerts a positive effect on the conversion from the fibroblast to the adipocyte phenotype induced in response to adipogenic hormones such as insulin and glucocorticoids.
|
) that was highly reminiscent to the localization of the endogenous protein in this cell type (Ref. 17 and this study, data not shown). Intriguingly, as opposed to other cell types such as COS7) and HEK293 (14, 16), expression of the kinase-dead PIKfyveK1831E mutant in 3T3-L1 adipocyte did not noticeably change the endomembrane morphology as seen on the phase-contrast images presented in Fig. 3A, panels b and d, and Fig. 3B, panels c and f. This result was important because it gave us the opportunity to follow up GLUT4 membrane dynamics in the absence of gross aberrations in cell morphology.
|
). Likewise, EGFP-GLUT4 coexpressed with PIKfyveWT displayed this typical distribution, i.e. perinuclearly in resting conditions and as a cell surface rim upon insulin stimulation (Fig. 3C, panels g–i) with the number of responding cells (45
60% of the cotransfected cells, Fig. 3D) being within the reported range for this cell type (27, 28, 29, 30). In contrast, GLUT4 fluorescence appearance in cells expressing kinase-dead PIKfyveK1831E demonstrated a different pattern. Thus, although PIKfyveK1831E did not affect the basal EGFP-GLUT4 localization, it prevented insulin-induced cell surface appearance of EGFP-GLUT4 (Fig. 3C, panels a–f). This phenomenon was observed in almost all cotransfected cells (Fig. 3D). Similar arrest of insulin-regulated EGFP-GLUT4 translocation in 3T3-L1 adipocytes was observed on expression of PIKfyveK1999/2000E (not shown), a kinase-dead dominant-negative mutant shown previously to have a selective defect in the lipid kinase activity but not in the protein kinase activity (16). Together these results suggest that PIKfyve lipid kinase activity positively regulates insulin-induced GLUT4 translocation.
Pharmacological inhibition of PIKfyve activity coincides with an arrest of insulin-stimulated GLUT4 translocation and glucose uptake in 3T3-L1 adipocytes
In 3T3-L1 adipocytes, insulin-regulated increase of glucose transport is thought to be due mainly to the GLUT4 glucose transporter. Because GLUT4 translocation is inhibited in the presence of dominant-negative kinase-dead PIKfyve mutants, it would be expected that these mutants will arrest the insulin-induced increase of glucose uptake. Unfortunately, the expression levels of all PIKfyve constructs in 3T3-L1 adipocytes were too low (1–30%), precluding biochemical tests with the transfected or transduced cells. To circumvent this technical problem, we sought to identify a selective cell-permeable inhibitor of PIKfyve enzymatic activity. Although PIKfyve activity was relatively resistant to wortmannin (11), a powerful inhibitor of PI3Ks, we have found that curcumin, a naturally occurring plant-derived polyphenol, is a potent PIKfyve blocker. In a cell-free system, a half-maximal inhibition of PIKfyve lipid kinase activity immunopurified from lysates of 3T3-L1 adipocytes was observed at 6 µM, and nearly full inhibition was evident at 40 µM (Fig. 4). At this relatively low concentration, curcumin is shown to inhibit PKA, whereas other enzymes are either inhibited at higher doses or insensitive (31, 32, 33). PI3K activity, as measured in P-Tyr immunoprecipitates, was almost unaffected by the above concentrations of curcumin (data not shown). Interestingly, we found no reports in the literature related to the effect of this inhibitor on insulin-regulated glucose transport and GLUT4 translocation.
|
). Intriguingly, short cell pretreatment with 100 µM curcumin resulted in a marked inhibition of insulin-induced GLUT4 translocation. Thus, the immunoreactive GLUT4 found on IM and PM fractions of curcumin-treated cells was approximately 1.6-fold higher and approximately 6-fold lower, respectively, vs. insulin-stimulated control 3T3-L1 adipocytes (Fig. 5A).
|
, curcumin treatment of 3T3-L1 adipocytes drastically inhibited the insulin-induced increase of glucose transport as measured by radiolabeled 2DG uptake. As illustrated in Fig. 5B, insulin stimulation of cells pretreated with curcumin resulted in an only 36% ± 8% increase of 2DG uptake, whereas in the nontreated cells, this effect was typically approximately 5-fold. In fact, we have observed a slight inhibition of the basal 2DG uptake in curcumin-treated cells that corresponded to 40% of the basal 2DG uptake of nontreated cells (Fig. 5B). This was consistent with the somewhat less GLUT4 on the PM fraction of curcumin-treated vs. nontreated basal cells, detected on overexposure of the immunoblots presented in Fig. 5A. These results show for the first time that the naturally occurring cell permeable compound curcumin drastically inhibits the insulin-regulated GLUT4 translocation and glucose transport. Thus, the results with the dominant-negative kinase-dead PIKfyve mutants combined with the pharmacological inhibition of PIKfyve enzymatic activity are consistent with the idea that PIKfyve enzymatic activity is essential for insulin’s effect on GLUT4 translocation and glucose uptake.
Effect of PIKfyve inactivation on glucose uptake in cells expressing only GLUT1
Insulin evokes the redistribution of both GLUT4 and the ubiquitous GLUT1 glucose transporter from the cell interior to the plasma membrane, but the increased cell capacity to transport glucose is believed to be mainly on account of GLUT4 in fat and muscle cells. To test whether PIKfyve affects insulin-induced glucose transport mediated by GLUT1 glucose transporter, we have examined a CHO-T cell line that stably expresses human IR but not GLUT4. The insulin-regulated increase of glucose transport in this cell type results exclusively from cell surface appearance of GLUT1 (35). Therefore, CHO-T cells transduced with adenoviruses encoding PIKfyveWT, the kinase-dead PIKfyveK mutant, or empty virus at approximately 100% efficiency were subjected to 2DG glucose uptake in the presence or absence of insulin. As demonstrated in Fig. 6, insulin stimulated 2DG uptake by approximately 1.8-fold in PIKfyveWT-infected cells. Essentially identical stimulation was observed in control cells infected with the empty virus (Fig. 6) or cells not infected (data not shown). By contrast, in PIKfyveK-infected cells, the insulin-induced stimulation of 2DG was substantially reduced and was not significantly different from the basal 2DG transport (Fig. 6). Together, these data indicate that PIKfyve enzymatic activity positively regulates insulin-induced stimulation of glucose uptake executed not only by GLUT4 but also by the GLUT1 glucose transporter.
|
K mutant, or empty virus at approximately 100% infection efficiency. The cells were then serum deprived, and, subsequent to acute insulin treatment, the phosphorylation status at Ser473 and Thr308 was examined by immunoblotting with specific antiphosphoSer473 or antiphosphoThr308 antibodies. As illustrated in Fig. 7A, expression of both PIKfyveWT and PIKfyveK resulted in neither a detection of phosphoThr308-Akt nor a change in the levels of Thr308 phosphorylation induced by insulin treatment. However, the antiphosphoSer473-Akt blots clearly demonstrated a different picture. Thus, PIKfyveWT expression induced a small but significant phosphorylation at Ser473 in the absence of insulin. This corresponded to 15% of the effect observed by insulin. Next, insulin stimulation of these cells resulted in a substantial rise in phosphorylation at Ser473, compared with insulin-stimulated cells transduced with the empty virus. This was detected in both membrane fraction (25% increase) and cytosol (50% increase) (Fig. 7, B and D). Conversely, insulin-induced Ser473 phosphorylation was reduced by 30% in cells infected with PIKfyveK adenovirus that was clearly detectable in the cytosolic fraction (Fig. 7, B and D). The more dramatic Ser473 phosphorylation in the cytosol vs. the membrane fraction of CHO-T cells likely represents the activated Akt form in transit to the downstream effectors located in the cytosol and/or nucleus, as also proposed by others (36, 37). It also should be emphasized that similar up-regulation of Akt phosphorylation at Ser473, both in the absence or presence of insulin, was reproducibly documented in HEK293 cells on transduction with PIKfyveWT adenovirus (not shown). Together these results are consistent with higher PIKfyve enzymatic activity leading and contributing to the insulin-induced Akt phosphorylation at the Ser473 site.
|
). Western blot analysis with anti-PIKfyve antibodies documented approximately 7- and 3-fold increase of PIKfyveWT and PIKfyveK1831E, respectively, over the endogenous PIKfyve upon doxycycline induction (Fig. 8).
|
, serum and insulin equally stimulated BrdU incorporation in the parental HEK293 cells with or without doxycycline induction, as judged by the similar increase of the number of fluorescence-positive cells. This remained unaffected on turning on the PIKfyveWT expression (Fig. 9A). However, the induction of PIKfyveK1831E expression decreased both the insulin- or serum-stimulated BrdU incorporation by 50% of that observed with the insulin-stimulated parental line (Fig. 9A). Similar inhibition of DNA synthesis by PIKfyveK1831E was obtained on quantitation of the [3H]thymidine accumulation. Thus, as illustrated in Fig. 9B, induction of PIKfyveK1831E expression abrogated the increase in DNA synthesis after insulin treatment. In contrast, insulin stimulated DNA synthesis by approximately 2-fold in the parental or PIKfyveWT-HEK293 cells with or without doxycycline induction (Fig. 9B and data not shown). It should be noted that insulin-stimulated DNA synthesis in the noninduced PIKfyveK1831E-transfected HEK293 cell line was slightly lower vs. the other two cell lines, likely caused by a protein leakage. Combined results of both strategies demonstrated that the dominant-negative kinase-dead PIKfyve mutant inhibited the ligand-induced DNA synthesis, suggesting that cellular PIKfyve enzymatic activity functions as a positive, stimulatory mitogenic signaling molecule.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The notion that PIKfyve enzymatic activity is an important intermediate in insulin-regulated metabolic and mitogenic functions was supported here by both mutagenesis and pharmacological studies. Thus, kinase-defective PIKfyve mutants with a dominant-negative effect down-regulated insulin stimulation of adipogenesis, cell growth and proliferation, and glucose metabolism. These effects have been demonstrated in: 1) 3T3-L1 fibroblasts that delayed in the early stages of the hormone-induced differentiation into the adipocytic phenotype a if transduced with kinase-dead PIKfyveK (Fig. 2); 2) a HEK293 stable cell line that attenuated insulin’s simulation of DNA synthesis if induced to express kinase-dead PIKfyveK1831E (Fig. 9); 3) 3T3-L1 adipocytes that failed to translocate GLUT4 onto the cell surface in response to insulin if transfected with kinase-dead PIKfyveK1831E (Fig. 3) or PIKfyveK1999/2000E (not shown); and4) CHO-T cells, in which the insulin-dependent increase of GLUT1-mediated glucose transport was dramatically inhibited upon transduction with kinase-dead PIKfyveK (Fig. 6). Furthermore, pharmacological inhibition of PIKfyve enzymatic activity with low concentrations of curcumin in 3T3-L1 adipocytes coincided with a dramatic decrease of insulin-stimulated glucose transport (Figs. 4 and 5). Although the meaning of the results with curcumin could be questioned because of the broad specificity of this inhibitor, it should be pointed out that several essential events in the insulin-signaling circuit leading to GLUT4 translocation, such as IR autophosphorylation or IRS-1 tyrosine phosphorylation (1, 2, 3, 4, 5) remain unaffected by the inhibitor in both 3T3-L1 adipocytes and CHO-T cells (Deeb, R., and A. Shisheva, unpublished data). Finally, it should be emphasized that COS cells treated with curcumin at a concentration (100 µM) and a time interval (30 min) that inhibited GLUT4 translocation to the 3T3-L1 adipocyte cell surface, produced a vacuolation phenotype resembling that induced upon expression of kinase-dead PIKfyve mutants in COS cells (data not shown) (14, 16). Thus, although curcumin may interact with and block proteins other than PIKfyve (31, 32, 33), it is conceivable that in the cellular context, PIKfyve is a major target for the inhibitor at the concentrations used. The dramatic arrest of insulin-induced GLUT4 translocation and 2DG uptake in 3T3-L1 adipocytes, therefore, may well be due to the inhibition of PIKfyve enzymatic activity.
The kinase-dead PIKfyve mutants used in this study were defined as dominant-negative because of their ability to induce abnormal morphology in the form of endomembrane swelling and vacuolation upon expression in cells of epithelial origin, including COS and HEK293 cells (14, 16). It should be pointed out that this abnormal phenotype was not noticeable in 3T3-L1 adipocytes (Fig. 3A), the reason for that being presently unknown. This observation is important because it renders the defects observed on expression of the mutant proteins in this cell type as primary, rather than secondary ones caused by altered endomembrane morphology. Therefore, the inhibition of insulin effect on GLUT4 translocation by PIKfyveK1831E (defective protein and lipid kinase activity) (13, 14) and PIKfyveK1999/2000E (selective defect in the lipid kinase activity) (16) documented here define the PIKfyve lipid products as important intermediates in the pathway whereby insulin regulates GLUT4 membrane dynamics and glucose homeostasis. Further insight into this issue should definitely await studies in transgenic animal models with targeted expression of PIKfyve dominant- negative mutants.
One intriguing observation in the present study was the fact that although the kinase-dead mutants and pharmacological inhibition of PIKfyve enzymatic activity were found to negatively control insulin-regulated cellular responses, higher levels of active PIKfyve enzyme were unable to mimic hormone’s effects. Taken together with the inability of acute insulin to further increase the PIKfyve enzymatic activity, these data may suggest that PIKfyve-derived signaling mediators are redundant. Alternatively, it is conceivable that to continue the signaling circuit leading to end point responses, PIKfyve may require an additional cofactor that is limited in the absence of insulin.
Whereas PIKfyve’s role as a positive regulator of several insulin-dependent bioresponses was unequivocally documented, the molecular mechanism(s) whereby this action is achieved was not resolved in the current study. Because the kinase-inactive mutants were shown to be inhibitory, it is conceivable that PIKfyve function is mediated via its enzymatic activity. PIKfyve, however, displays a remarkable diversity of enzymatic activities and synthesizes PtdIns 5-P, PtdIns 3,5-P2, and phosphoprotein(s) (12, 13, 14, 38). Because all activities were eliminated in PIKfyveK or PIKfyveK1831E mutants, it is particularly difficult to pinpoint the downstream mediator that relays the PIKfyve phosphorylation signals in every single case. Although we were able to identify the PIKfyve lipid kinase as essential in the insulin effect on GLUT4 translocation, our data do not imply that all of the cellular effects studied here are mediated by the PIKfyve lipid products. Further studies are necessary to clarify the relative impact of PtdIns 3,5-P2 vs. PtdIns 5-P as well as the contribution of the PIKfyve protein kinase activity and yet-to-be-identified phosphorylated proteins in the signal delivery to the PIKfyve downstream targets.
Although downstream effectors and upstream regulators of PIKfyve enzymatic activity are still unknown, one potential downstream target that emerges from the present work is Akt. Akt is believed to be downstream of PI3K and important in insulin metabolic effects such as glucose transport and protein synthesis in 3T3-L1 adipocytes (5, 6). According to current models, Akt is activated upon translocation to the plasma membrane and binding to PtdIns 3,4-P2 and PtdIns 3,4,5-P3, generated on PI3K activation. This binding alters Akt conformation and eases the accessibility and phosphorylation by two upstream kinases, PDK1 and an unidentified kinase referred to as PDK2 (7). PDK1, which is also activated by PtdIns 3,4,5-P3 binding, phosphorylates Thr308, but the phosphorylation at Ser473 is presently elusive. The documented decrease in insulin-dependent Akt phosphorylation at Ser473 in the presence of dominant-negative kinase-dead PIKfyveK and the increase by high levels of PIKfyveWT (Fig. 7) imply that PIKfyve enzymatic activity may also play a positive role in the activation of Akt. Although Ser473 phosphorylation is under the control of PI3Ks, the fact that the kinase for this site is still not identified makes these data more complicated to interpret. It is still premature to distinguish between a direct Akt activation, activation through PDK2 or inactivation through a phosphatase responsible for the turnover at Ser473. However, it is tempting to speculate that PIKfyve-dependent production of PtdIns 3,5-P2 in these cells (14) could contribute to the Akt activation; in fact, upon expression of high PIKfyve levels, we have observed a small but reproducible increase of Akt phosphorylation at Ser473 in the absence of insulin. In any case, in vitro binding of Akt to PtdIns 3,5-P2 lipids that has not been tested in the present study may shed light on the plausible role of PIKfyve in Akt activation. It is worth emphasizing that populations of PIKfyve and class IA PI3Ks are found in a complex and their concerted action for PtdIns 3,5-P2 production in an insulin-regulated manner has been suggested (18). Thus, it is conceivable that PIKfyve enzymatic activity, likely through acute PtdIns 3,5-P2 production, may contribute to the full Akt phosphorylation at Ser473 observed upon insulin treatment, although the physiological meaning of this is still unknown.
Emerging evidence indicates that proteins involved in insulin-signaling transduction mechanisms are also crucial in adipocyte differentiation. Thus, IR, IRS-1 and IRS-2, PI3K, or Akt have been separately shown to be essential in the ability of fibroblasts to differentiate into adipocytes (39, 40, 41, 42, 43). Here we present data that PIKfyve enzymatic activity is also positively regulating adipogenesis. One intriguing observation in the present study was the fact that PIKfyve kinase-dead mutants attenuated not only the hormone-induced adipogenesis but also the DNA synthesis. These data are consistent with the currently accepted model whereby active cell cycle machinery is required for the adipocyte differentiation process (44). Although detailed molecular mechanisms of this effect is outside the scope of this study, this result indicates that an inhibition of PIKfyve should both arrest a cell cycle and promote apoptosis. Thus, besides its role as a positive intermediate in insulin-regulated acute responses, PIKfyve enzymatic activity is also a positive regulator of insulin’s mitogenic effects.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
BrdU, 3'-Bromo-5'-deoxyuridine; CHO, Chinese hamster ovary; 2DG, 2-deoxyglucose; EGFP, enhanced GFP; FBS, fetal bovine serum; GFP, green fluorescent protein; HA, hemagglutinin; HEK, human embryonic kidney; HES, HEPES-HCl, EDTA, and sucrose; IM, intracellular membrane; IR, insulin receptor; IRS, insulin receptor substrate; MOI, multiplicity of infection; PDK, phosphoinositide-dependent kinase; PI3K, phosphatidylinositol 3-kinase; PM, plasma membrane; PtdIns, phosphatidylinositol; TLC, thin-layer chromatography.
Received June 12, 2002.
Accepted for publication August 27, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  O. C. Ikonomov, J. Fligger, D. Sbrissa, R. Dondapati, K. Mlak, R. Deeb, and A. Shisheva Kinesin Adapter JLP Links PIKfyve to Microtubule-based Endosome-to-Trans-Golgi Network Traffic of Furin J. Biol. Chem., February 6, 2009; 284(6): 3750 - 3761. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Shisheva Phosphoinositides in insulin action on GLUT4 dynamics: not just PtdIns(3,4,5)P3 Am J Physiol Endocrinol Metab, September 1, 2008; 295(3): E536 - E544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Osborne, P. J. Wen, C. Boucheron, H. N. Nguyen, M. Hayakawa, H. Kaizawa, P. J. Parker, N. Vitale, and F. A. Meunier PIKfyve Negatively Regulates Exocytosis in Neurosecretory Cells J. Biol. Chem., February 1, 2008; 283(5): 2804 - 2813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yu, P.-O. Berggren, and C. J. Barker An Autocrine Insulin Feedback Loop Maintains Pancreatic {beta}-Cell 3-Phosphorylated Inositol Lipids Mol. Endocrinol., November 1, 2007; 21(11): 2775 - 2784. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sbrissa, O. C. Ikonomov, Z. Fu, T. Ijuin, J. Gruenberg, T. Takenawa, and A. Shisheva Core Protein Machinery for Mammalian Phosphatidylinositol 3,5-Bisphosphate Synthesis and Turnover That Regulates the Progression of Endosomal Transport: NOVEL SAC PHOSPHATASE JOINS THE ArPIKfyve-PIKfyve COMPLEX J. Biol. Chem., August 17, 2007; 282(33): 23878 - 23891. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. S. L. Thong, C. B. Dugani, and A. Klip Turning Signals On and Off: GLUT4 Traffic in the Insulin-Signaling Highway Physiology, August 1, 2005; 20(4): 271 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Kalant, R. MacLaren, W. Cui, R. Samanta, P. N. Monk, S. A. Laporte, and K. Cianflone C5L2 Is a Functional Receptor for Acylation-stimulating Protein J. Biol. Chem., June 24, 2005; 280(25): 23936 - 23944. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. C. Berwick, G. C. Dell, G. I. Welsh, K. J. Heesom, I. Hers, L. M. Fletcher, F. T. Cooke, and J. M. Tavare Protein kinase B phosphorylation of PIKfyve regulates the trafficking of GLUT4 vesicles J. Cell Sci., December 1, 2004; 117(25): 5985 - 5993. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Sbrissa, O. C. Ikonomov, J. Strakova, and A. Shisheva Role for a Novel Signaling Intermediate, Phosphatidylinositol 5-Phosphate, in Insulin-Regulated F-Actin Stress Fiber Breakdown and GLUT4 Translocation Endocrinology, November 1, 2004; 145(11): 4853 - 4865. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. C. Ikonomov, D. Sbrissa, K. Mlak, R. Deeb, J. Fligger, A. Soans, R. L. Finley Jr., and A. Shisheva Active PIKfyve Associates with and Promotes the Membrane Attachment of the Late Endosome-to-trans-Golgi Network Transport Factor Rab9 Effector p40 J. Biol. Chem., December 19, 2003; 278(51): 50863 - 50871. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sbrissa, O. C. Ikonomov, R. Deeb, and A. Shisheva Phosphatidylinositol 5-Phosphate Biosynthesis Is Linked to PIKfyve and Is Involved in Osmotic Response Pathway in Mammalian Cells J. Biol. Chem., November 27, 2002; 277(49): 47276 - 47284. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), PIKfyve
), and empty virus at MOI = 5, resulting in 5–7% efficiency of infection, as identified by the reporter GFP expressed under an independent promoter. Two days after infection, the fibroblasts were subjected to a standard differentiation protocol detailed in Materials and Methods. The accumulation of lipid droplets was monitored by fluorescence microscopy using a Hoffman modulation contrast system with 40 x objective or postfixation, by oil red O staining. A, The percentage of green cells with lipid droplets was plotted over time. Data are collected by observing 10 randomly selected fields (150–200 cells) per dish in three separate experiments. B, Depicted are typical examples of fluorescent images of PIKfyve
