Advertisement

Phosphatidylinositol 3-Kinase Activity Is Required at a Postendocytic Step in Platelet-derived Growth Factor Receptor Trafficking ∗

  • Marguerite Joly
    Affiliations
    Program in Molecular Medicine and the Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01605
    Search for articles by this author
  • Andrius Kazlauskas
    Affiliations
    National Jewish Hospital for Immunology and Respiratory Medicine, Denver, Colorado 80206
    Search for articles by this author
  • Silvia Corvera
    Correspondence
    To whom correspondence should be addressed: Dept. of Cell Biology, University of Massachusetts Medical School, 373 Plantation St., Worcester, MA 01605. Tel.: 508-856-6898; Fax: 508-856-4289;
    Affiliations
    Program in Molecular Medicine and the Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01605
    Search for articles by this author
  • Author Footnotes
    ∗ This work was supported by National Institutes of Health Grants DK-40330 (to S. C.) and GM48339 (to A. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:June 02, 1995DOI:https://doi.org/10.1074/jbc.270.22.13225
      We have previously reported that platelet-derived growth factor (PDGF) receptor mutants that lack high affinity binding sites for phosphatidylinositol 3-kinase (PI 3-kinase) fail to concentrate in juxtanuclear vesicular structures after activation with PDGF. We have now identified the point in the endocytic pathway at which PI 3-kinase binding sites are required. Receptor internalization from the plasma membrane, measured as the acquisition of acid resistance of prebound 125I-PDGF, was only slightly decreased in cells expressing a PDGF receptor mutant (F5) lacking PI 3-kinase, GTPase-activating protein (GAP), phospholipase Cγ, and Syp binding sites but not expressing mutants where any of these individual sites were restored nor expressing a mutant lacking exclusively PI 3-kinase binding sites. In contrast, the extent of down-regulation of PDGF binding sites from the cell surface after prolonged incubation with PDGF as well as the degradation of [35S]methionine-labeled receptor were markedly reduced in cells expressing the F5 mutant, mutants restored in GAP, phospholipase Cγ, or Syp binding sites or expressing the mutant exclusively lacking PI 3-kinase binding sites but not in cells expressing the mutant where PI 3-kinase binding sites were restored. Inhibition of PI 3-kinase activity with wortmannin caused a dramatic decrease in the rates of down-regulation and degradation of wild-type receptors. These results suggest that PI 3-kinase binding sites are not required for internalization of PDGF receptor but are required to divert the PDGF receptor to a degradative pathway. Furthermore, the requirement for PI 3-kinase binding sites on the receptor appears to be due to a requirement for PI 3-kinase catalytic activity.

      INTRODUCTION

      Stimulation of receptor tyrosine kinases rapidly elicits a large number of intracellular events including actin filament reorganization, membrane ruffling, transmembrane ion fluxes, and changes in the activity of metabolic enzymes. One of the earliest responses to stimulation is the internalization of the activated receptor. Internalized receptors can recycle back to the cell surface (
      • Marshall S.
      • Green A.
      • Olefsky J.M.
      ), or can be sorted to the lysosome for degradation. This latter process results in the down-regulation of cell surface receptors. Receptor down-regulation appears to be an important element in the control of cellular proliferation, since mutations that impair the ability of some receptors to internalize and degrade cause cell transformation (
      • Chen W.S.
      • Lazar C.S.
      • Lund K.A.
      • Welsh J.B.
      • Chang C.P.
      • Walton G.M.
      • Der C.J.
      • Wiley H.S.
      • Gill G.N.
      • Rosenfeld M.G.
      ).
      In addition to its physiological relevance, the down-regulation of receptor tyrosine kinases offers a useful model system to study the molecular events involved in the trafficking of proteins from the plasma membrane to the lysosome, because receptor movement can be selectively and synchronously elicited by addition of specific growth factors (
      • Hiles I.D.
      • Otsu M.
      • Volinia S.
      • Fry M.J.
      • Gout I.
      • Dhand R.
      • Panatoyou G.
      • Ruiz-Larrea F.
      • Thompson A.
      • Totty N.F.
      • Hsuan J.J.
      • Courtneidge S.A.
      • Parker P.J.
      • Waterfield M.D.
      ). Internalization of receptor tyrosine kinases often involves both high affinity, low capacity and low affinity, high capacity pathways. The high affinity, low capacity internalization depends on tyrosine phosphorylation of the receptor cytoplasmic domain (
      • Backer J.M.
      • Khan C.R.
      • Cahill D.A.
      • Ullrich A.
      • White M.F.
      ,
      • Lohse M.J.
      ,
      • Felder S.
      • LaVin J.
      • Ullrich A.
      • Schlessinger J.
      ,
      • Backer J.M.
      • Shoelson S.E.
      • Haring E.
      • White M.F.
      ,
      • Sorkin A.
      • Westermark B.
      • Heldin C-H.
      • Claesson-Welsh L.
      ). Following internalization, some receptor tyrosine kinases are directed to the lysosomal pathway. The molecular requirements involved in this sorting process are not understood.
      We have recently explored the hypothesis that the SH2 domain-containing proteins that are recruited to receptor cytoplasmic domains after activation may play a role in the trafficking of receptor tyrosine kinases (
      • Kapeller R.
      • Chakrabarti R.
      • Cantley L.
      • Fay F.
      • Corvera S.
      ,
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ). Tyrosine autophosphorylation of the PDGF1(
      The abbreviations used are: PDGF
      platelet-derived growth factor
      GAP
      GTPase-activating protein
      PI 3-kinase
      phosphatidylinositol 3-kinase
      DMEM
      Dulbecco's modified Eagle's medium
      EGF
      epidermal growth factor.
      ) receptor creates high affinity binding sites for several proteins including phosphatidylinositol 3-kinase (PI 3-kinase), phospholipase Cγ, a GTPase-activating protein for Ras (GAP), a 65-kDa phosphotyrosine phosphatase (Syp), non-receptor tyrosine kinases, and the adaptor proteins Grb2 and Nck (
      • Koch C.A.
      • Anderson D.
      • Moran M.F.
      • Ellis C.
      • Pawson T.
      ,
      • Cantley L.C.
      • Auger K.R.
      • Carpenter C.
      • Duckworth B.
      • Graziani A.
      • Kapeller R.
      • Soltoff S.
      ,
      • Escobedo J.
      • Navankasattusas S.
      • Kavanaugh W.M.
      • Milfay D.
      • Fried V.
      • Williams L.
      ,
      • Fantl W.
      • Escobedo J.
      • Martin G.
      • Turck C.
      • del Rosario M.
      • McCormick F.
      • Williams L.
      ,
      • Kazlauskas A.
      • Kashishian A.
      • Cooper J.
      • Valius M.
      ,
      • Kypta R.
      • Goldberg Y.
      • Ulug E.
      • Courtneidge S.
      ,
      • Nishimura R.
      • Li W.
      • Mondino A.
      • Zhou M.
      • Cooper J.
      • Schlessinger J.
      ). A PDGF receptor mutant that can autophosphorylate but lacks the major high affinity binding sites for PI 3-kinase, GAP, phospholipase Cγ, and Syp displays striking alterations in cellular trafficking patterns (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ,
      • Valius M.
      • Kazlauskas A.
      ); whereas wild-type receptors concentrate in a juxtanuclear vesicular compartment, the mutant remains dispersed at the cell periphery (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ). Restoration of the PI 3-kinase binding sites completely restores the ability of receptors to concentrate intracellularly.
      These results raise important questions regarding the point in the endocytic pathway at which PI 3-kinase binding sites are required and the biochemical basis for this requirement. The results shown here strongly suggest that PI 3-kinase binding sites are required at a postendocytic step, where the sorting of activated PDGF receptors to the lysosome occurs. Furthermore, wortmannin, which has recently been shown to be a potent (IC50 < 10 nM) inhibitor of the p110 catalytic subunit of mammalian PI 3-kinase (
      • Yano H.
      • Nakanishi S.
      • Kimura K.
      • Hanai N.
      • Saitoh Y.
      • Fukui Y.
      • Nonomura Y.
      • Matsuda Y.
      ,
      • Okada T.
      • Kawano Y.
      • Sakakibara T.
      • Hazeki O.
      • Ui M.
      ), inhibits the down-regulation and degradation of activated receptor, suggesting that the requirement for PI 3-kinase binding sites reflects a requirement for PI 3-kinase catalytic activity.

      MATERIALS AND METHODS

      Cells and Constructs

      The F52(
      The mutants used in this study are designated in the following way. The receptor in which tyrosines at positions 740, 751, 771, 1009, and 1021 were replaced by phenylalanine is called F5. The “add-back” mutants represent F5 receptors in which phenylalanines at positions 740, 751, 771, 1009, or 1021 were replaced by tyrosine individually or in combination and are designated by the single letter code for tyrosine, followed by the position number of the replaced residue. The add-back mutant in which phenylalanines at positions 740 and 751 were replaced by tyrosine is called Y40/51. The wild-type receptor in which tyrosines at positions 740 and 751 were replaced by phenylalanine is called F40/51.
      ) mutant is a receptor in which tyrosines 740, 751, 771, 1009, and 1021 were substituted with phenylalanine and is deficient in the binding of PI 3-kinase, phospholipase Cγ, GAP and Syp (
      • Valius M.
      • Kazlauskas A.
      ). Receptor constructs selectively restored in the high affinity binding of PI 3-kinase (Y40/51), phospholipase Cγ (Y1021), GAP (Y771), or Syp (Y1009) were generated by mutating the corresponding phenylalanines in F5 back to tyrosine. In addition, a receptor exclusively deficient in PI 3-kinase binding (F40/51) was constructed by mutating Tyr740 and Tyr751 in the wild-type receptor to phenylalanine. Each construct was introduced into human HepG2 cells, which do not express endogenous β-PDGF receptors, with the pLXSN retroviral expression vector (
      • Valius M.
      • Kazlauskas A.
      ). All constructs were expressed at approximately 5 × 105 receptors/cell (
      • Valius M.
      • Kazlauskas A.
      ) and were activated by ligand (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ,
      • Valius M.
      • Kazlauskas A.
      ). Cells were grown to 80% confluence in Dulbecco's modified Eagle's medium (DMEM) supplemented with fetal calf serum (10%) (UBI) containing G418 (0.5 mg/ml).

      125I-PDGF Internalization

      Recombinant PDGF-BB was labeled with 125I to 21,000 cpm/ng by the method of Bolton and Hunter according to the manufacturer's instructions (Amersham Corp.). Cells expressing wild-type or mutant PDGF receptors or the empty pLXSN vector were grown in 12-well multiwell dishes to 75% confluence and serum-starved for 18 h. PDGF internalization experiments were conducted as described by Mori et al.(
      • Mori S.
      • Claesson-Welsh L.
      • Heldin C-H.
      ). Cells were washed twice with ice-cold DMEM and incubated with 0.5 ml of DMEM containing 1 mg/ml bovine serum albumin and 50,000 cpm of 125I-PDGF. After 90 min of incubation at 5°C, monolayers were washed three times with ice-cold binding medium. Binding medium at 37°C was then added for 0-20 min, and cells were rapidly cooled by placing the multiwell dishes on ice and washing with ice-cold binding medium. Cells were then incubated for 5 min with 500 μl of phosphate-buffered saline containing 1 mg/ml bovine serum albumin, pH 7.5, or with the same buffer at pH 3.5. This acid wash procedure has been shown to release approximately 90% of cell surface bound radioactivity (
      • Mori S.
      • Claesson-Welsh L.
      • Heldin C-H.
      ). Cells were then washed three times with binding medium, and monolayers were dissolved in a lysis buffer composed of Triton X-100 (1%), glycerol (10%), and Tris, pH 7.5 (20 mM). Radioactivity associated with the acid-washed monolayers (internalized) or the non-acid washed monolayers (total cell-associated) was measured by γ counting.

      Receptor Down-regulation

      Cells expressing wild-type or mutant PDGF receptors or the empty pLXSN vector were cultured and serum-starved as described above. Monolayers were incubated with DMEM or DMEM containing 0-20 ng/ml PDGF-BB for 30-90 min at 37°C. Cells were then cooled and acid-washed as described above. After the acid wash, cells were washed three times with ice-cold DMEM and incubated for 90 min at 5°C with 0.5 ml of DMEM containing 1 mg/ml bovine serum albumin and 100,000 cpm of 125I-PDGF. Monolayers were washed three times and dissolved in lysis buffer. Nonspecific binding was defined as the radioactivity bound to monolayers of HepG2 cells transfected with pLXSN empty vector.

      Receptor Degradation

      Cells were grown to 85% confluence in 60-mm culture dishes and incubated for 18 h in serum-free, methionine-free DMEM containing [35S]methionine (100 μCi/ml) (Amersham Corp.) and bovine serum albumin (1%) and then for 45 min in DMEM containing bovine serum albumin (1%) and methionine (0.3 mg/ml) (DMEM-bovine serum albumin). Cells were then incubated without or with PDGF-BB (50 ng/ml) (UBI) for 2 h. Cells were washed twice in ice-cold phosphate-buffered saline and lysed in 800 μl of a buffer containing 20 mM Tris, pH 8.0, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, 2 mM phenylmethylsulfonyl fluoride, 1 mM benzamidine, 1 mM 1,10 phenanthroline, 10 μg/ml leupeptin, and 1 mM sodium vanadate (Sigma). Receptors were immunoprecipitated with a polyclonal antibody to α- and β-PDGF receptors (UBI), resolved on 7.5% polyacrylamide gels, transferred to nitrocellulose, and exposed to autoradiographic film for 48 h. Anti-phosphotyrosine antibody 4G10 (UBI) was used for immunoblotting.

      Effects of Wortmannin

      Wortmannin (Sigma) was dissolved in Me2SO to a final concentration of 10 mM, dispensed into 5-μl aliquots, and stored at −80°C. Wortmannin aliquots were thawed and diluted 1:1000 in ice-cold phosphate-buffered saline, and aliquots from this diluted stock were added directly to the cells to achieve the final concentrations indicated in each experiment. Because wortmannin is photosensitive and unstable in aqueous solutions, it was routinely thawed, diluted, and added to cells within 10 min. Thawed aliquots were discarded.

      RESULTS

      Initial Rates of Internalization of PDGF Receptors

      Several trafficking and sorting steps are involved in targeting activated PDGF receptors into the lysosome. Receptors must first clear the plasma membrane by a ligand-dependent process that involves clustering in clathrin-coated pits, invagination, and scission of clathrin-coated vesicles (
      • Lamaze C.
      • Baba T.
      • Redelmeir T.E.
      • Schmid S.L.
      ). Subsequently, receptors are sorted to a nonrecycling, degradative pathway. The observed peripheral distribution of receptor mutants lacking PI 3-kinase binding sites (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ) could result from a defect either in clearing the plasma membrane or in leaving the early endosomal/recycling pathway. We performed experiments to evaluate the ability of wild-type and mutant receptors to undergo the initial steps of internalization.
      Cells were incubated with tracer concentrations (2 ng/ml) of 125I-labeled PDGF for 60 min at 4°C. Unbound ligand was washed, and cells were placed at 37°C for 0-10 min. At specific time points, cells were returned to 4°C, and the ligand remaining accessible to the extracellular space was removed by acid washing (
      • Mori S.
      • Claesson-Welsh L.
      • Heldin C-H.
      ). Uptake was only observed in cells expressing PDGF receptors and not in cells transfected with empty pLXSN vector, demonstrating the specificity of the assay (Fig. 1). In further experiments internalization was measured after 10 min of incubation.
      Figure thumbnail gr1
      Fig. 1Internalization of 125I-PDGF. Cells were seeded in 24-well multiwell dishes and serum-starved for 18 h. Monolayers were cooled by washing in ice-cold DMEM and incubated for 60 min at 5°C with 2 ng/ml 125I-PDGF. Cells were then warmed to 37°C for the times indicated, cooled with ice-cold DMEM, and placed on ice. Cell surface 125I-PDGF was removed by acid washing, and intracellular radioactivity was determined in detergent lysates. Closedcircles, cells expressing wild-type receptors; opencircles, cells transfected with empty vector.
      Approximately 60% of the PDGF initially bound to the cell surface (determined in cells that were not acid washed) became resistant to acid washing after 10 min of incubation (Fig. 2). Thus, the initial uptake of PDGF receptor in these cells was rapid and comparable with that observed by other investigators using other cell lines (
      • Mori S.
      • Claesson-Welsh L.
      • Heldin C-H.
      ). The initial uptake of the F5 mutant was slightly but significantly decreased compared with that of the wild-type receptor. Add-back mutants, as well as a mutant lacking only PI 3-kinase binding sites (F40/51), internalized PDGF like the wild-type receptor. These results suggest that the initial steps of PDGF receptor internalization are independent of PI 3-kinase, GAP, Syp, or phospholipase Cγ. The results also indicate that the large differences in the distribution of PDGF mutant receptors in the cell, observed by immunofluorescence (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ), are not due to differences in the ability of receptors to undergo the initial steps of internalization (Fig. 2).
      Figure thumbnail gr2
      Fig. 2Internalization of PDGF receptor mutants. Cells expressing the PDGF receptor constructs indicated were seeded in 24-well multiwell dishes and serum-starved for 18 h. Monolayers were incubated for 60 min at 5°C with 2 ng/ml 125I-PDGF and then warmed to 37°C for 10 min. Cells were then cooled and incubated for 5 min with phosphate-buffered saline, pH 7.4, to determine total cell-associated ligand or with phosphate-buffered saline, pH 3.5, to determine internalized ligand. Nonspecific binding was determined in cells expressing the empty vector and was subtracted. The acid-resistant (internalized) radioactivity was expressed as a percentage of the total cell-associated radioactivity. Bars represent the mean and lines the S.E. of three experiments performed in triplicate. Statistical comparison with the wild-type receptor was done using paired two-tail t test analysis. The only significant difference was between the wild-type and the F5 receptor.

      Down-regulation of PDGF Binding Sites from the Cell Surface

      Previous studies have shown that prolonged exposure of cells to PDGF leads to a rapid disappearance of PDGF receptors from the plasma membrane and to their subsequent degradation (
      • Lohse M.J.
      ). This down-regulation of PDGF receptors occurs rapidly, indicating that once internalized, receptors must move from the early endosomal/recycling pathway to the lysosomal pathway. The peripheral distribution of receptor mutants lacking PI 3-kinase binding sites could be due to an impairment in the ability of receptors to progress beyond the early endosomal/recycling pathway. To explore this possibility, we analyzed the extent to which PDGF receptor mutants down-regulate from the plasma membrane. Cells expressing the indicated constructs were incubated with unlabeled PDGF for 60 min at 37°C and subsequently placed at 4°C. Unlabeled PDGF was removed by a brief acid wash, and the remaining PDGF binding sites on the cell surface were then measured with 125I-PDGF. Approximately 70% of the total cell surface binding sites for PDGF disappeared after exposure of cells expressing wild-type receptors to 20 ng/ml of PDGF for 60 min (Fig. 3). In contrast, only 20% of the binding sites disappeared after a similar incubation of cells expressing the F5 mutant or a receptor lacking only sites 740 and 751 (F40/51). Only the add-back mutant restored in PI 3-kinase binding sites (Y40/51) down-regulated to an extent similar to that of the wild-type receptor. Thus, the ability of receptors to efficiently down-regulate from the plasma membrane and to concentrate in juxtanuclear structures depended on the presence of functional PI 3-kinase binding sites. Because the lack of PI 3-kinase binding sites does not impair the initial internalization of receptors from the plasma membrane (Fig. 2), the observed lack of efficient down-regulation is probably due to recycling of internalized receptors to the cell surface. Thus, lack of PI 3-kinase binding sites on the PDGF receptor results in its failure to progress from early and/or recycling endosomes to later steps in the endocytic pathway.
      Figure thumbnail gr3
      Fig. 3Down-regulation of cell surface PDGF receptors. Cells expressing the PDGF receptor constructs indicated were seeded in 24-well multiwell dishes and serum-starved for 18 h. Cells were incubated without or with 20 ng/ml PDGF-BB for 90 min at 37°C, acid-washed with buffer at pH 3.5, and then washed twice in DMEM to remove any remaining cell surface bound ligand and neutralize the pH. Monolayers were then incubated for 60 min at 5°C with 2 ng/ml 125I-PDGF. After washing, cell-associated radioactivity was determined in detergent cell lysates. Nonspecific binding was determined in cells expressing the empty vector and was subtracted. The radioactivity associated with cells incubated with 20 ng/ml PDGF-BB was expressed as a percentage of the radioactivity associated with cells incubated without PDGF. Bars represent the mean and lines the S.E. of five experiments performed in duplicate. Statistical comparison with the wild-type receptor was done using paired two-tail t test analysis. Only the Y40/51 mutant was not significantly different from the wild-type receptor (NS).

      Tyrosine Phosphorylation of PDGF Receptors

      The results presented above suggest that efficient down-regulation of PDGF receptors from the cell surface requires intact PI 3-kinase binding sites. A prediction made by this hypothesis is that receptors that fail to down-regulate could undergo several rounds of ligand binding and activation, and this event would be reflected by the tyrosine phosphorylation state of the receptor. To test this prediction, we analyzed the tyrosine phosphorylation state of PDGF receptor mutants by immunoblotting of receptor immunoprecipitates with anti-phosphotyrosine antibodies (Fig. 4).
      Figure thumbnail gr4
      Fig. 4Tyrosine phosphorylation of PDGF receptor mutants. HepG2 cells expressing the PDGF receptor mutants indicated were serum-starved and then exposed to PDGF-BB (50 ng/ml) for the times indicated. Receptors were immunoprecipitated, resolved on SDS-PAGE, transferred to nitrocellulose, and probed with anti-phosphotyrosine antibody. Leftpanel, an example of an immunoblot from one experiment. Middle and rightpanels, the intensity of the phosphotyrosine signal on the PDGF receptor bands after 5 min (middlepanel) or 120 min (rightpanel) of exposure to PDGF was quantified by densitometric scanning. The values obtained by densitometric scanning of four independent experiments were averaged and plotted. The means and S.E. of these values were as follows: 76 ± 13, 15 ± 6, 37 ± 16, 29 ± 9, 47 ± 6, and 39 ± 13 for WT, F5, Y40/51, Y771, Y1009 and Y1021 at 5 min and 6 ± 4, 49 ± 10, 3 ± 2, 37 ± 7, 38 ± 6, 40 ± 10 at 120 min. Statistical comparison with the wild-type receptor was done using paired two-tail t test analysis. At 5 min, only the F5 mutant was significantly different (p < .016) from the wild-type receptor. At 120 min, Y40/51 was not significantly different from WT, but all other mutants were (p < .005-.05).
      Tyrosine phosphorylation of wild-type and all mutant receptors could be detected within 5 min of exposure of cells to 50 ng/ml PDGF. The extent of tyrosine phosphorylation on the mutant receptors was significantly lower than for the wild type after 5 min of stimulation (Fig. 4, left and middlepanels). This decrease may be due to the absence in the mutants of three to five major autophosphorylation sites. Phosphorylation sites remaining on the mutant receptors probably include Tyr857, a major phosphorylation site involved in activation of the receptor kinase, as well as residues in the juxtamembrane region of the receptor (
      • Kazlauskas A.
      • Durden D.L.
      • Cooper J.A.
      ,
      • Mori S.
      • Ronnstrand L.
      • Yokote K.
      • Engstrom A.
      • Courtneidge S.A.
      • Claesson-Welsh L.
      • Heldin C-H.
      ). More importantly, striking differences were observed in the kinetics of tyrosine phosphorylation of wild-type and mutant receptors. The phosphorylation of receptors containing PI 3-kinase binding sites reached maximal levels after 5-10 min of incubation and then declined. Virtually no phosphotyrosine was detected in these receptors after 120 min of stimulation. In contrast, tyrosine phosphorylation of the F5 mutant and of the add-back mutants containing phospholipase Cγ, GAP, or Syp binding sites, increased steadily during the first 60 min of incubation. A substantial amount of phosphotyrosine (50% of maximal value) could still be detected after 2 h of incubation with ligand (Fig. 4, left and rightpanels). The persistence of tyrosine phosphorylation on receptors lacking PI 3-kinase binding sites is consistent with the possibility that these receptors undergo additional rounds of ligand binding and activation as a consequence of recycling to the cell surface.

      Rates of Degradation of PDGF Receptors

      The results shown above indicate that a lack of PI 3-kinase binding sites on the PDGF receptor results in its failure to progress from early and/or recycling endosomes to later steps in the endocytic pathway leading to the lysosome. To directly determine the rates of ligand-induced receptor degradation, cells were labeled with [35S]methionine and then incubated for 2 h in the absence or presence of 50 ng/ml PDGF. PDGF receptors were then immunoprecipitated with a polyclonal antibody raised against the receptor cytoplasmic domain. Receptors were then resolved by SDS-PAGE and visualized by autoradiography.
      A pronounced decrease in the levels of labeled wild-type and Y40/51 receptors was observed in these experiments (Fig. 5). In contrast, neither F5 mutant receptors nor the other add-back mutant receptors were appreciably degraded after a similar time of exposure to ligand. These results support the hypothesis that intact PI 3-kinase binding sites are required for the trafficking of activated PDGF receptors to the lysosomal pathway. Receptors that rapidly concentrate in juxtanuclear vesicular structures (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ) after exposure to PDGF are efficiently degraded, consistent with these structures corresponding to late endosomes, multivesicular bodies, and lysosomes.
      Figure thumbnail gr5
      Fig. 5Degradation of PDGF receptor mutants. HepG2 cells expressing the indicated PDGF receptor mutants were labeled with [35S]methionine and incubated for 2 h in the absence (-) or presence (+) of PDGF-BB (50 ng/ml). Receptors were immunoprecipitated, resolved on SDS-PAGE, transferred to nitrocellulose, and exposed to autoradiographic film. An example of the autoradiograph from one experiment is shown, where the major band corresponds to the expected position of the PDGF receptor. The increased intensity of the control band in the Y1009 lanes detected in this experiment was not reproducible. The intensity of the PDGF receptor bands was quantified using a Pharmacia Biotech Inc. laser densitometer. The values obtained by densitometric scanning of three to six independent experiments were averaged and plotted (lowerpanel). The percentage of receptor degraded after 2 h was (mean ± S.E.) 54 ± 4, 9 ± 6, 49 ± 7, 0, 17 ± 10, and 17 ± 10 for WT, F5, Y40/51, Y771, Y1009, and Y1021, respectively. Statistical comparison was done using paired two-tail t test analysis. Only the WT and Y40/51 mutant were significantly degraded (p < .005 and p < .02, respectively).

      Effects of Wortmannin on PDGF Receptor Down-regulation

      The results shown above indicate a requirement for PI 3-kinase binding sites for receptor trafficking into the lysosomal pathway. However, these results do not prove that PI 3-kinase is required for this trafficking process, because the same phosphotyrosine residues that bind PI 3-kinase could conceivably also interact with other cellular components. For example, Tyr751 has been shown to be a binding site for the adaptor protein Nck (
      • Nishimura R.
      • Li W.
      • Mondino A.
      • Zhou M.
      • Cooper J.
      • Schlessinger J.
      ). To explore the specific role of PI 3-kinase, we used a fungal toxin, wortmannin, which has recently been shown to be a potent inhibitor of the catalytic p110 subunit of mammalian PI 3-kinase (
      • Yano H.
      • Nakanishi S.
      • Kimura K.
      • Hanai N.
      • Saitoh Y.
      • Fukui Y.
      • Nonomura Y.
      • Matsuda Y.
      ,
      • Okada T.
      • Kawano Y.
      • Sakakibara T.
      • Hazeki O.
      • Ui M.
      ).
      We first analyzed the effects of wortmannin on receptor down-regulation from the cell surface (Fig. 6). Cells expressing wild-type receptors, the F5 mutant, and the Y40/51 add-back mutant were incubated without or with wortmannin (50 nM) for 10 min and then exposed to PDGF (50 ng/ml) for 30-90 min at 37°C. After a brief acid wash, the binding of 125I-PDGF to the cell surface was measured. As shown above, incubation of cells for 60-90 min with PDGF resulted in the efficient down-regulation of wild-type and Y40/51 receptors but not of the F5 mutant receptor. In contrast, when cells were treated with wortmannin, PDGF-induced down-regulation of wild-type and Y40/51 receptors was impaired and was now indistinguishable from that of the F5 receptor (Fig. 6). Thus, the ability of receptors that bind PI 3-kinase to efficiently down-regulate from the plasma membrane is abolished by wortmannin treatment.
      Figure thumbnail gr6
      Fig. 6Effect of wortmannin on the down-regulation of PDGF receptors. Cells expressing wild-type (solidbar), F5 (stripedbar), or Y40/51 (shadedbar) mutant PDGF receptors were seeded in 24-well multiwell dishes and serum-starved for 18 h. Cells were incubated without or with 20 ng/ml PDGF-BB at 37°C for the times indicated. Where indicated, wortmannin (50 nM) was added to cells 10 min before addition of PDGF and remained in the medium for the rest of the experiment. Cells were cooled, acid-washed, and incubated for 60 min at 5°C with 2 ng/ml 125I-PDGF. After washing, cell-associated radioactivity was determined in detergent cell lysates. Nonspecific binding was determined in cells expressing the empty vector and was subtracted. The radioactivity associated with cells incubated with 20 ng/ml PDGF-BB was expressed as a percentage of the radioactivity associated with cells incubated without PDGF and plotted. This experiment was repeated three times with similar results.

      Effects of Wortmannin on PDGF Receptor Degradation

      Cells expressing wild-type PDGF receptors were incubated with [35S]methionine, chased for 45 min with unlabeled methionine, and then incubated for 2 h in the absence or presence of PDGF. Wortmannin, where indicated, was added to the cells for the last 10 min of the chase period and remained in the medium for the duration of the experiment. PDGF receptors were immunoprecipitated, resolved by SDS-PAGE, and visualized by autoradiography (Fig. 7). Treatment of cells with PDGF resulted in an almost complete degradation of labeled receptor. Wortmannin fully inhibited receptor degradation in response to PDGF but did not significantly alter the levels of labeled receptor immunoprecipitated from cells incubated in the absence of PDGF (Fig. 7, leftpanel). The maximal inhibitory effect was observed between 10 and 25 nM wortmannin (Fig. 7, rightpanel), which is similar to the dose required to fully inhibit PI 3-kinase invitro(
      • Valius M.
      • Kazlauskas A.
      ,
      • Yano H.
      • Nakanishi S.
      • Kimura K.
      • Hanai N.
      • Saitoh Y.
      • Fukui Y.
      • Nonomura Y.
      • Matsuda Y.
      ). At these concentrations, the tyrosine kinase activity of the PDGF receptor was unaffected (not shown). Thus, the down-regulation and degradation of PDGF receptors in human HepG2 cells is fully blocked in response to concentrations of wortmannin that inhibit PI 3-kinase. These results support the hypothesis that the requirement for PI 3-kinase binding sites for PDGF receptor trafficking into the lysosomal pathway is due to a requirement for PI 3-kinase activity.
      Figure thumbnail gr7
      Fig. 7Effect of wortmannin on the degradation of wild-type PDGF receptors. HepG2 cells expressing the wild-type PDGF receptor were labeled with [35S]methionine and incubated for 2 h in the absence (-) or presence (+) of PDGF-BB (50 ng/ml). Wortmannin was added at the concentrations indicated 10 min before exposure to PDGF. A representative autoradiograph of the receptor immunoprecipitates is shown in the leftpanel. The expected position of the PDGF receptor is indicated with an arrow. The intensity of the PDGF receptor band was quantified with a Pharmacia laser densitometer, and the results were plotted (rightpanel).

      DISCUSSION

      We have previously reported that the trafficking of PDGF receptors requires PI 3-kinase binding sites (
      • Joly M.
      • Kazlauskas A.
      • Fay F.
      • Corvera S.
      ). A critical question raised by this finding is at what point in the pathway of PDGF receptor internalization are these sites required. Here we show that the requirement for PI 3-kinase binding sites is not during the initial steps of endocytosis. These results are consistent with the findings of Mori et al.(
      • Mori S.
      • Ronnstrand L.
      • Claesson-Welsh L.
      • Heldin C-H.
      ) who have shown that only mutations in tyrosine 579 of the PDGF receptor cytoplasmic domain significantly affect this process. The initial internalization of a receptor mutant lacking tyrosines 740, 751, 771, 1009, and 1021 was decreased by 30%, but restoration of any of the missing tyrosine residues was sufficient to restore the wild-type internalization rates. These results indicate that these residues play a small individual role in the internalization of the PDGF receptor. However, the results also suggest that multiple tyrosine residues in the receptor cytoplasmic domain can function as weak internalization signals. The net rate of internalization is probably the product of all these signals. PI 3-kinase binding sites appear to be necessary at a step subsequent to internalization. Without PI 3-kinase binding sites receptors escape degradation and appear to recycle to the cell surface.
      A second question addressed in this study relates to the nature of the requirement for PI 3-kinase binding sites. These sites are known to bind the p85 subunit of PI 3-kinase but also the adaptor protein Nck (
      • Nishimura R.
      • Li W.
      • Mondino A.
      • Zhou M.
      • Cooper J.
      • Schlessinger J.
      ) and possibly other unknown factors. The p85 subunit of PI 3-kinase interacts with dynamin, a GTPase that is critically required for endocytosis. Thus, the requirement for PI 3-kinase binding sites for PDGF receptor trafficking could reflect a requirement for p85, for p110 catalytic activity, for Nck, or for a yet unknown factor. The finding that wortmannin, a potent inhibitor of PI 3-kinase activity, blocks efficient down-regulation and degradation of wild-type receptors suggests strongly that the requirement for PI 3-kinase binding sites reflects a requirement for PI 3-kinase activity.
      A role for PI 3-kinase activity in sorting of membrane proteins has been clearly established in yeast, where the only known PI 3-kinase, vps34, is required for the sorting of newly synthesized hydrolases to the yeast vacuole (
      • Herman P.K.
      • Stack J.H.
      • Emr S.E.
      ). Interestingly, vps34 has also been recently discovered in a screen for endocytosis-dependent growth, which monitors endocytosis from the cell surface to the vacuole (
      • Munn A.L.
      • Riezman H.
      ). Experiments on these mutants have shown that vps34/end12 is not necessary for internalization of α-factor but is critically required for its subsequent delivery into the vacuole and its degradation (
      • Munn A.L.
      • Riezman H.
      ). These results are consistent with a direct role for PI 3-kinase in sorting in the endocytic pathway both in yeast and mammalian cells.
      A requirement for PI 3-kinase in steps subsequent to initial internalization suggests that postendocytic receptor sorting requires active receptor tyrosine kinase activity in order to maintain an association with the SH2 domains of p85. It has been shown that the internalized EGF and PDGF receptors have an active tyrosine kinase activity (
      • Kapeller R.
      • Chakrabarti R.
      • Cantley L.
      • Fay F.
      • Corvera S.
      ,
      • Honegger A.M.
      • Dull T.J.
      • Felder S.
      • VanOberghen E.
      • Bellot F.
      • Szapary D.
      • Schmidt A.
      • Ullrich A.
      • Schlessinger J.
      ) and that the activity of the EGF receptor kinase is necessary for sorting the EGF receptor to the lysosome (
      • Futter C.E.
      • Felder S.
      • Sclessinger J.
      • Ullrich A.
      • Hopkins C.R.
      ). This view, however, has been questioned recently in experiments that conclude that postendocytic trafficking of EGF receptors is independent of their intrinsic tyrosine kinase activity. In this latter paper however, degradation of EGF ligand, not of EGF receptor, was analyzed (
      • Herbst J.J.
      • Opresko L.K.
      • Walsh B.
      • Lauffenburger D.A
      • Wiley H.S.
      ).
      An important question raised by these results is whether PI 3-kinase binding is required for the down-regulation of other receptor tyrosine kinases in mammalian cells. In this regard, it is interesting that the insulin receptor, which escapes degradation after internalization (
      • Marshall S.
      • Green A.
      • Olefsky J.M.
      ), fails to bind PI 3-kinase directly (
      • Myers Jr., M.G.
      • Sun X.J.
      • White M.F.
      ). The postendocytic destiny of the EGF receptor varies. In some cells internalized receptors down-regulate efficiently (
      • Futter C.E.
      • Felder S.
      • Sclessinger J.
      • Ullrich A.
      • Hopkins C.R.
      ,
      • Herbst J.J.
      • Opresko L.K.
      • Walsh B.
      • Lauffenburger D.A
      • Wiley H.S.
      ), whereas in others they recycle to the cell surface (
      • Dunn W.A.
      • Connoly T.P.
      • Hubbard A.L.
      ,
      • Masui H.
      • Castro L.
      • Mendelsohn J.
      ). Interestingly, the EGF receptor binds PI 3-kinase poorly, with a 12-fold lower affinity than the PDGF receptor (
      • Hu P.
      • Margolis B.
      • Skolnik E.Y.
      • Lammers R.
      • Ullrich A.
      • Schlessinger J.
      ). However, in some cells EGF receptors dimerize with ErbB3, which binds PI 3-kinase with an affinity comparable with the PDGF receptor (
      • Fedi P.
      • Pierce J.H.
      • DiFiore P.P.
      • Kraus M.H.
      ). It is interesting to speculate on the possibility that the observed differences in the ability of EGF receptors to efficiently down-regulate may correlate with their ability to dimerize with ErbB3 and thus interact strongly with PI 3-kinase. The availability of wortmannin will be useful in determining whether the requirement for PI 3-kinase for receptor down-regulation is universal.
      PI 3-kinase in mammalian cells has been implicated in several functions of receptor tyrosine kinases, including mitogenic signaling, actin filament organization, chemotaxis, and down-regulation (
      • Kapeller R.
      • Cantley L.C.
      ). The molecular mechanisms whereby PI 3-kinase mediates these diverse cellular functions are unknown. It is possible that each of the lipid products of mammalian PI 3-kinase, which include PI-3-P, PI-3,4-P2, and PI-3,4,5-P3, may have unique targets and thus coordinately but independently regulate multiple cellular responses. In contrast to mammalian PI 3-kinase, yeast vps34 only phosphorylates PI to produce PI-3-P, suggesting that this lipid may be the principal product involved in protein sorting events. In summary, genetic and pharmacological evidence presented here suggests a critical requirement for PI 3-kinase activity in postendocytic trafficking and down-regulation of β-PDGF receptors. Important questions remain relating to the biochemical basis for this requirement.

      Acknowledgments

      Immunofluorescence studies were done using instrumentation of the Biomedical Imaging Group at the University of Massachusetts. We thank David Hartley and Howard Shpetner for helpful discussions and critical reading of the manuscript.

      REFERENCES

        • Marshall S.
        • Green A.
        • Olefsky J.M.
        J. Biol. Chem. 1981; 256: 11464-11470
        • Chen W.S.
        • Lazar C.S.
        • Lund K.A.
        • Welsh J.B.
        • Chang C.P.
        • Walton G.M.
        • Der C.J.
        • Wiley H.S.
        • Gill G.N.
        • Rosenfeld M.G.
        Cell. 1989; 59: 33-43
        • Backer J.M.
        • Khan C.R.
        • Cahill D.A.
        • Ullrich A.
        • White M.F.
        J. Biol. Chem. 1990; 265: 16450-16454
        • Lohse M.J.
        Biochim. Biophys. Acta. 1993; 1179: 171-188
        • Felder S.
        • LaVin J.
        • Ullrich A.
        • Schlessinger J.
        J. Cell Biol. 1992; 117: 203-212
        • Backer J.M.
        • Shoelson S.E.
        • Haring E.
        • White M.F.
        J. Cell Biol. 1991; 115: 1525-1545
        • Sorkin A.
        • Westermark B.
        • Heldin C-H.
        • Claesson-Welsh L.
        J. Cell Biol. 1991; 112: 469-478
        • Kapeller R.
        • Chakrabarti R.
        • Cantley L.
        • Fay F.
        • Corvera S.
        Mol. Cell. Biol. 1993; 13: 6052-6063
        • Joly M.
        • Kazlauskas A.
        • Fay F.
        • Corvera S.
        Science. 1994; 263: 581-724
        • Koch C.A.
        • Anderson D.
        • Moran M.F.
        • Ellis C.
        • Pawson T.
        Science. 1991; 252: 669-674
        • Cantley L.C.
        • Auger K.R.
        • Carpenter C.
        • Duckworth B.
        • Graziani A.
        • Kapeller R.
        • Soltoff S.
        Cell. 1991; 64: 281-302
        • Escobedo J.
        • Navankasattusas S.
        • Kavanaugh W.M.
        • Milfay D.
        • Fried V.
        • Williams L.
        Cell. 1991; 33: 75-82
        • Fantl W.
        • Escobedo J.
        • Martin G.
        • Turck C.
        • del Rosario M.
        • McCormick F.
        • Williams L.
        Cell. 1992; 69: 413-423
        • Kazlauskas A.
        • Kashishian A.
        • Cooper J.
        • Valius M.
        Mol. Cell. Biol. 1992; 12: 2534-2544
        • Kypta R.
        • Goldberg Y.
        • Ulug E.
        • Courtneidge S.
        Cell. 1990; 62: 481-492
        • Nishimura R.
        • Li W.
        • Mondino A.
        • Zhou M.
        • Cooper J.
        • Schlessinger J.
        Mol. Cell. Biol. 1993; 13: 6889-6896
        • Valius M.
        • Kazlauskas A.
        Cell. 1993; 73: 321-334
        • Yano H.
        • Nakanishi S.
        • Kimura K.
        • Hanai N.
        • Saitoh Y.
        • Fukui Y.
        • Nonomura Y.
        • Matsuda Y.
        J. Biol. Chem. 1993; 268: 25846-25856
        • Okada T.
        • Kawano Y.
        • Sakakibara T.
        • Hazeki O.
        • Ui M.
        J. Biol. Chem. 1994; 269: 3568-3573
        • Mori S.
        • Claesson-Welsh L.
        • Heldin C-H.
        J. Biol. Chem. 1991; 266: 21158-21164
        • Lamaze C.
        • Baba T.
        • Redelmeir T.E.
        • Schmid S.L.
        Mol. Biol. Cell. 1993; 4: 715-727
        • Kazlauskas A.
        • Durden D.L.
        • Cooper J.A.
        Cell Regul. 1991; 2: 413-425
        • Mori S.
        • Ronnstrand L.
        • Yokote K.
        • Engstrom A.
        • Courtneidge S.A.
        • Claesson-Welsh L.
        • Heldin C-H.
        EMBO J. 1993; 12: 2257-2264
        • Hiles I.D.
        • Otsu M.
        • Volinia S.
        • Fry M.J.
        • Gout I.
        • Dhand R.
        • Panatoyou G.
        • Ruiz-Larrea F.
        • Thompson A.
        • Totty N.F.
        • Hsuan J.J.
        • Courtneidge S.A.
        • Parker P.J.
        • Waterfield M.D.
        Cell. 1992; 70: 419-429
        • Mori S.
        • Ronnstrand L.
        • Claesson-Welsh L.
        • Heldin C-H.
        J. Biol. Chem. 1994; 269: 4917-4921
        • Herman P.K.
        • Stack J.H.
        • Emr S.E.
        Trends Cell Biol. 1992; 2: 363-368
        • Munn A.L.
        • Riezman H.
        J. Cell Biol. 1994; 127: 373-386
        • Honegger A.M.
        • Dull T.J.
        • Felder S.
        • VanOberghen E.
        • Bellot F.
        • Szapary D.
        • Schmidt A.
        • Ullrich A.
        • Schlessinger J.
        Cell. 1987; 51: 199-209
        • Futter C.E.
        • Felder S.
        • Sclessinger J.
        • Ullrich A.
        • Hopkins C.R.
        J. Cell Biol. 1993; 120: 77-83
        • Herbst J.J.
        • Opresko L.K.
        • Walsh B.
        • Lauffenburger D.A
        • Wiley H.S.
        J. Biol. Chem. 1994; 269: 12865-12873
        • Myers Jr., M.G.
        • Sun X.J.
        • White M.F.
        Trends Biochem. Sci. 1994; 19: 289-294
        • Dunn W.A.
        • Connoly T.P.
        • Hubbard A.L.
        J. Cell Biol. 1986; 102: 24-36
        • Masui H.
        • Castro L.
        • Mendelsohn J.
        J. Cell Biol. 1993; 120: 85-93
        • Hu P.
        • Margolis B.
        • Skolnik E.Y.
        • Lammers R.
        • Ullrich A.
        • Schlessinger J.
        Mol. Cell. Biol. 1992; 12: 981-990
        • Fedi P.
        • Pierce J.H.
        • DiFiore P.P.
        • Kraus M.H.
        Mol. Cell. Biol. 1994; 14: 492-500
        • Kapeller R.
        • Cantley L.C.
        BioEssays. 1994; 16: 565-575