| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cell and Tumor Biology |
1 Department of Biochemistry and Molecular Biology, Thoracic Diseases Research Unit, and Mayo Clinic Cancer Center, Mayo Clinic College of Medicine, Rochester, Minnesota and 2 Division of Basic Medical Sciences, Memorial University of Newfoundland, St. Johns, Newfoundland
Requests for reprints: Edward B. Leof, Stabile 8-58, Mayo Clinic, 200 First Street Southwest, Rochester, MN 55905. Phone: 507-284-5717; Fax: 507-284-4521; E-mail: leof.edward{at}mayo.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With such a variety of responses, it is not surprising that TGF-ß receptor (TGF-ßR) activation and signaling are uniquely regulated (7, 8). There are three primary receptors for TGF-ß on most cell types: types I, II, and III (ß-glycan) receptor. TGF-ßR activation is initiated through the binding of ligand to the type II receptor, which is a constitutively active serine/threonine kinase (9). This event facilitates the recruitment and subsequent transphosphorylation of the type I receptor in a glycine- and serine-rich region known as the GS domain. The activated type I receptor then serves as a docking site for receptor-associated Smad (R-Smad) proteins that are brought to the receptor complex associated with the FYVE domain protein Smad anchor for receptor activation (SARA; ref. 9). Following phosphorylation at a specific SSXS site in the COOH terminus by the type I receptor, the R-Smad proteins dissociate from the type I receptor and translocate to the nucleus complexed with the common mediator Smad4, where they serve as comodulators of transcription (7, 8).
Although the Smad pathway is critical for many aspects of TGF-ß signaling, Smad-independent responses have also been documented (10, 11). Recently, we characterized the Smad-independent activation of the yeast STE20 homologue p21-activated kinase-2 (PAK2) by TGF-ß, which occurs in mesenchymal and not epithelial cultures (1). Furthermore, the activation of PAK2 was required for the morphologic alterations and proliferative responses induced in these cells. Other Smad-independent signaling targets include TGF-ß-activated kinase-1 (12), Ras (13), various Rho proteins (14, 15), c-Jun NH2-terminal kinase (10, 11), extracellular signal-regulated kinase (16), p38 (17), and the non–receptor tyrosine kinase c-Abl (18). The manner that these pathways integrate with each other and/or Smad signaling is currently not well understood.
In addition to the above targets, previous work has suggested that phosphatidylinositol 3-kinase (PI3K) might be involved in positively or negatively regulating TGF-ß signaling at least in some cell types (3, 19–21). PI3Ks belong to a family of proteins that phosphorylate the 3'-hydroxyl group on the inositol ring of phosphoinositides and consist of a regulatory p85 subunit and catalytic p110 subunit (22). In addition to generating various lipid effectors, PI3K also phosphorylates several protein substrates whose activity has been linked to cell growth, differentiation, and survival (22, 23). There are multiple isoforms of PI3Ks, which originally have been subdivided into three classes (23). Class I, II, and III enzymes in the PI3K family generate specific lipid mediators that bind to FYVE and pleckstrin homology domains in a variety of proteins, affecting their localization, conformation, and activity (22). Class I PI3Ks catalyze the phosphorylation of the 3'-hydroxyl subunit of phosphoinositides. Although phosphatidylinositol, phosphatidylinositol 4-phosphate, and phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2] can all act as substrates for class I PI3Ks in vitro, PtdIns(4,5)P2 is the main substrate in vivo (23). Thus, class I PI3Ks generally catalyze the conversion of PtdIns(4,5)P2 to phosphatidylinositol 3,4,5-trisphosphate. Class II PI3Ks consist of a small group of larger PI3K molecules whose cellular function is currently unclear (23). Class III PI3Ks are homologues of the yeast vesicular protein sorting protein Vps34p. These PI3Ks can only use phosphatidylinositol as a substrate in vitro and are most likely to be responsible for the generation of a large proportion of the phosphatidylinositol 3-phosphate [PtdIns(3)P] in cells (22, 24). Because the cellular levels of PtdIns(3)P remain fairly constant, this suggests that the physiologic processes in which class III PI3Ks are involved are not acutely triggered by cellular stimulation. As such, the major cellular function of class III PI3Ks is believed to be in intracellular trafficking, with the product of the enzyme [PtdIns(3)P] being recognized by proteins containing FYVE domain, such as EEA1, Vac1p, YOTB, Hrs, and SARA (22, 24).
PI3K signaling has been implicated in the control of a wide range of cellular activities, including (but not limited to) proliferation, survival, adhesion, differentiation, and cytoskeletal organization (22, 23, 25). However, although many growth factors clearly use the PI3K pathway(s) as an integral component to their biological function, the role of PI3K in TGF-ß signaling is presently unclear/unknown. PI3K has been reported to be involved (20), not involved (26), or somewhat involved (27–29) in the cytoskeletal changes associated with TGF-ß-stimulated epithelial-to-mesenchymal transitions (EMT). Although the reasons for these discrepancies are presently unclear, a component of the controversy may simply reflect the fact that the majority of studies were done in epithelial cells undergoing an EMT-like change. If PI3K is primarily a response induced by TGF-ß in fibroblast cultures, only a fraction of the epithelial population would be "mesenchymal" at the times tested. As such, the present study was undertaken to determine (a) whether PI3K was indeed activated in a fibroblast-specific manner to TGF-ß and (b) to further clarify the requirement for PI3K signaling in the cellular response to TGF-ß stimulation. The data document that (a) Smad-independent activation of class I PI3K regulates TGF-ß-mediated morphologic and proliferative responses in a subset of fibroblast not epithelial cultures, (b) PI3K signaling occurs in a plasma membrane proximal compartment independent of receptor internalization, and (c) PI3K-dependent activation of PAK2 is necessary for TGF-ß-stimulated morphologic transformation and cell proliferation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Western blotting. Fibroblast lines with a targeted deletion of Smad3 were obtained from Dr. Anita Roberts (NIH, Bethesda, MD). Cultures were treated overnight in serum-free DMEM and stimulated for the indicated times in the presence or absence of 5 ng/mL TGF-ß, 20 µg/mL LY294002, or 30 µmol/L SH-5. Cells were lysed [50 mmol/L Tris (pH 7.4), 1% NP40, 0.25% sodium deoxycholate, 150 mmol/L NaCl, 1 mmol/L EGTA, 1 mmol/L Na3VO4, 1x Complete protease inhibitor; Roche Applied Science, Indianapolis, IN], and equivalent protein was probed with the indicated antibody. The phosphorylated Smad2 and total Smad2 antibodies were from Upstate Biotechnology, Inc. (Lake Placid, NY), and BD Transduction Laboratories (San Jose, CA), respectively, whereas the total Smad3 antibody was from Zymed Laboratories (San Francisco, CA). A rabbit anti-phosphorylated Smad3 antibody to the peptide COOH-GSPSIRCSpSVpS was generated in our laboratory (1). Other antibodies were from Cell Signaling Technology (Beverly, MA) (anti-PAK2, anti-Akt, and anti-pS473Akt), Upstate Biotechnology (anti-pT308Akt and anti-p85), Santa Cruz Biotechnology (Santa Cruz, CA) (anti-HA, sc-805; anti-p110, sc-8010), Stratagene (La Jolla, CA) (anti-PI3KC3 N-term), BD Transduction Laboratories (San Jose, CA) (anti-PI3K p170), Roche Applied Science [anti–green fluorescent protein (GFP)], Sigma (anti-Flag, A-1205), Amersham Pharmacia (Piscataway, NJ) (anti-His), and Clontech (Mountain View, CA) (anti-myc).
Morphologic transformation and proliferation. AKR-2B cells were plated at 2.5 x 105 per six-well dish and incubated at 37°C for 24 hours. Confluent cultures were placed in serum-free MCDB 402 (JRH Biosciences, Lenexa, KS) for 48 hours and stimulated by the addition of fresh serum-free DMEM alone or containing 5 ng TGF-ß and in the absence or presence of 20 µg/mL LY294002. After 48-hour incubation, images were acquired by phase-contrast microscopy (magnification, x20). Following photography, cultures were trypsinized and the cell number was determined by hemacytometer.
Transfections. For luciferase assays, cells were plated in six-well dishes at 2.5 x 105 per six-well dish before transfection (4 hours) with 2 µg of either 3TP-Lux, Smad-binding element (SBE)-Lux, or 
-smooth muscle actin (SMA)-Lux and 0.5 µg cytomegalovirus (CMV)-ß-galactosidase with Fugene 6 (Roche Applied Science). Cultures were left untreated or stimulated with 5 ng/mL TGF-ß in the absence or presence of 20 µg/mL LY294002 and then assayed for luciferase activity as described previously (30).
To investigate the effect of antisense p85, morpholino antisense oligonucleotides representing nucleotides –22 to +3 of mouse p85 (5' CATGTGTGCGACAGTTTCCTTGGCT 3') or a scrambled control (5' GTACAGACTATTGCCTCTCGATCTT 3') were synthesized with 3' fluorescein by Gene Tools (Philomath, OR). Cos7 cells were plated in six-well dishes at a density of 1.0 x 105 per well in 10% DMEM and incubated at 37°C for 24 hours. Confluent cells were transfected with the p85 antisense or control oligonucleotide along with plasmids expressing the type I and II TGF-ßR at a final concentration of 6 µmol/L with LipofectAMINE 2000 (Invitrogen, Carlsbad, CA). Following 48 hours of incubation in 2% FBS-DMEM and 24-hour incubation in 0.1% FBS-DMEM, cultures were stimulated for 60 minutes at 37°C in 2 mL serum-free DMEM alone or containing 10 ng/mL TGF-ß before Western analysis (described above) or PAK2 kinase assay (described below).
Protein kinase assays. Lysates for kinase assays were prepared as described (1) and equivalent protein (500-700 µg) was incubated overnight at 4°C with the specified antibody. Immune complexes were collected with protein A-Sepharose (Sigma) and washed twice in kinase lysis buffer and twice in kinase buffer [25 mmol/L Tris (pH 7.4), 10 mmol/L MgCl2, 1 mmol/L DTT] before incubation in 50 µL kinase buffer containing 5 µmol/L ATP, 5 µCi [
-32P]ATP per reaction, and 5 µg of either myelin basic protein (Sigma) for PAK2 assays or immunopurified glycogen synthase kinase-3 (GSK-3) to assess Akt activity. Immunopurified GSK-3 substrate was obtained using "catch-and-release" kit supplied by Upstate Biotechnology. The kinase reactions were allowed to proceed for 10 minutes at 37°C, stopped with 50 µL of 2x Laemmli buffer, and processed by SDS-PAGE and autoradiography.
In vitro lipid kinase assays. To determine the ability of PI3K to phosphorylate phosphoinositides, cells were grown to 80% confluency before overnight incubation in serum-free DMEM. Cultures were treated with 5 ng/mL TGF-ß for the indicated times before washing thrice with ice-cold buffer A [137 mmol/L NaCl, 20 mmol/L Tris-HCl (pH 7.4), 1 mmol/L CaCl2, 1 mmol/L MgCl2, 0.1 mmol/L sodium orthovanadate]. Cells were then lysed (buffer A containing 1% NP40 and 1 mmol/L phenylmethylsulfonyl fluoride) with gentle agitation for 20 minutes at 4°C and centrifuged for 10 minutes at 13,000 x g. The supernatant was immunoprecipitated with antibody to the p85 subunit of PI3K for 1 hour at 4°C, and immune complexes were collected with protein A-Sepharose and washed thrice with buffer A containing 1% NP40, thrice with 100 mmol/L Tris-HCl (pH 7.4), 5 mmol/L LiCl, and 0.1 mmol/L sodium orthovanadate, and twice with TNE [10 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 5 mmol/L EDTA, 0.1 mmol/L sodium orthovanadate] before incubation in 75 µL lipid kinase buffer (TNE containing 15 mmol/L MgCl2, 20 µg freshly prepared L--phosphatidylinositol, 60 µmol/L ATP, 400 µCi/mL [-32P]ATP). Phosphatidylinositol (Sigma) was prepared by drying the lipid under a stream of argon before the addition of 10 mmol/L Tris-HCl (pH 7.4) and 1 mmol/L EGTA and suspension by sonication in an ice bath for 10 minutes. The kinase reaction was allowed to proceed for 10 minutes at 37°C and stopped by addition of 20 µL CHCl3/methanol (1:1; high-performance liquid chromatography grade) and brief vortexing. Organic and aqueous phases were separated by centrifugation and 50 µL of the organic phase were spotted on a 1-mm Silica gel 60 F254 TLC plate (EM Science, Gibbstown, NJ). The plate was developed in CHCl3/methanol/H2O/NH4OH [60:47:11:3.2] and 32P-labeled phosphatidylinositol phosphate (PIP) was visualized by autoradiography and referenced to known standards.
Adenovirus constructs. Dominant-negative PAK2-expressing and empty adenovirus were generated by transfection of adenovirus shuttle vector pAdCMV into 293Cre cells plated 24 hours earlier at 9 x 105 per six-well dish. Recombinant clones were determined by induction of cytopathic effects in the monolayers, isolated, and plaque purified in 293Cre cells. Control GFP-expressing adenovirus was purchased from Riken Genbank (Japan). Dominant-negative dynamin-expressing adenovirus was generously supplied by Dr. Jeff Pessin (Stony Brook University, Stony Brook, NY).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Activation of PI3K following TGF-ß treatment has not been consistently observed (20, 26, 27). As many of these studies were done on epithelial cells undergoing an EMT, we next determined whether this might reflect distinct signaling responses of the two cell types. To address this issue, three representative fibroblast (NIH3T3, Swiss3T3, and AKR-2B) and epithelial [HeLa, Mv1Lu, and Madin-Darby canine kidney (MDCK) cells] cell lines were stimulated with TGF-ß and examined for effects on Akt phosphorylation. As shown in Fig. 1C, TGF-ß stimulated Akt phosphorylation in each of the fibroblast cultures yet had no effect on the epithelial lines. This cell tropism is identical to what we reported previously for PAK2 activation (1) and provides evidence (albeit indirect) that PAK2 and PI3K might be integrated within the same TGF-ß signaling pathway (see below).
Phosphatidylinositol 3-kinase activation and Smad phosphorylation occur independently and in separate cellular locales. Although Smad-independent signaling has been reported (10, 11), TGF-ß signaling is primarily mediated through the Smad family of transcriptional coregulators. Because the Smad docking protein SARA contains a FYVE domain, the presence of which is essential for correct localization and activation of Smad2 in response to TGF-ß (9, 31, 32), PI3K activation (Fig. 1) may be required for production of phosphoinositides to recruit SARA and Smad proteins to the activated receptor complex. To determine if PI3K activity was necessary for Smad activation, we first inhibited PI3K with LY294002 and assessed whether the kinetics of Smad2 or Smad3 phosphorylation in response to TGF-ß would be modulated. As shown in Fig. 2A, detectable levels of phosphorylated Smad2 and Smad3 occur within 15 minutes of TGF-ß stimulation; inhibition of PI3K by LY294002 (Fig. 2A) or wortmannin (data not shown) has no discernable effect on either the kinetics or the extent of induction. The concentration of LY294002 used was determined to be sufficient to ablate PI3K activation by TGF-ß as indicated by the failure to phosphorylate phosphatidylinositol in vitro or Akt in vivo (Fig. 1A). Moreover, increasing the LY294002 concentration to levels approaching toxicity (200 µg/mL and theoretically inhibiting class I, II, and III PI3Ks) still failed to diminish Smad phosphorylation after TGF-ß stimulation (data not shown).
|
-SMA luciferase activity in response to TGF-ß (Fig. 2C). As reported previously, each of these reporter constructs required the presence of Smad3 (Fig. 2C; ref. 26). Although Fig. 2 shows that PI3K activity is not required for Smad2 and/or Smad3 phosphorylation or action, it does not address whether Smad2 and/or Smad3 phosphorylation is required for PI3K activation. Accordingly, Smad3-null (Smad3–/–) mouse embryo fibroblasts (MEF) were cotransfected with a Myc-tagged dominant-negative Smad2 (Myc-Smad2S467A) and a His-tagged wild-type Akt (Fig. 3). Following stimulation with TGF-ß, cells were lysed and phosphorylation of the transfected Akt protein was determined. Although Smad2 and Smad3 activation did not occur in the knockout clone (Fig. 3, left), TGF-ß-mediated Akt phosphorylation proceeded normally (Fig. 3, right). The results of Figs. 2 and 3 are consistent with the hypothesis that (in fibroblasts) Smad proteins and PI3K are independently activated by the TGF-ßR complex.
|
|
|
Although Fig. 5B clearly shows that inhibiting Akt activity with SH-5 or dominant-negative AktK179A or Pdk1 does not effect PAK2 activation, it does not address the converse (i.e., whether PAK2 kinase activity is required for Akt phosphorylation). Accordingly, AKR-2B cells were infected with adenovirus expressing dominant-negative PAK2 and the level of Akt phosphorylation was determined following addition of TGF-ß. Expression of dominant-negative PAK2 prevented PAK2 activation but had no effect on TGF-ß-stimulated Akt phosphorylation (Fig. 5C). Lastly, although LY294002 and wortmannin are used at concentrations that are thought to selectively inhibit type I PI3Ks (Figs. 1A, 2, and 5A and B; data not shown), it is still possible that the drugs might affect other kinases. To address that issue, we transfected morpholino antisense oligonucleotides to the regulatory p85 subunit of type I PI3K and assessed their effect on TGF-ß signaling. As shown in Fig. 5D, the antisense oligonucleotides, but not the scrambled control, specifically diminish p85 protein; no effect was observed on the type I PI3K catalytic p110 subunit, class II p170 protein, or class III PI3K protein PI3KC3. Most importantly, in agreement with the LY294002 and wortmannin data, loss of p85 protein resulted in an absence of TGF-ß-stimulated Akt phosphorylation and PAK2 activation with no attenuation in Smad2 phosphorylation.
The previous data indicate that TGF-ß-stimulated PI3K is independent of Smad protein activation and distinctly modulates both Akt and PAK2 signaling in AKR-2B cells. Of interest, however, was whether this new target could be linked to biological phenotypes regulated by TGF-ß in fibroblast cultures. As shown in Fig. 6A, AKR-2B cells treated with TGF-ß transform from large "cobblestone-like" geometric shapes into thin elongated cells that spread and extend over one another, reminiscent of transformed cultures. To investigate the role(s) of PI3K in this morphologic transformation, cells were treated with the PI3K inhibitor LY294002 before and during TGF-ß stimulation. Whereas expression of the delivery vehicle (DMSO) was without effect, treatment with LY294002 or adenovirus expressing dominant-negative PAK2 prevented cells from morphologically transforming in the presence of TGF-ß (Fig. 6A; ref. 1). As these findings indicated a direct role for PI3K and/or its downstream targets in TGF-ß action, we wished to discern if the proliferative responses induced by TGF-ß were similarly effected. Addition of TGF-ß to confluent cultures in serum-free medium resulted in an approximate cell doubling over 48 hours, whereas treatment with LY294002 or dominant-negative PAK2 abolished TGF-ß-stimulated monolayer proliferation (Fig. 6B). Analogous results were observed on fibroblasts in soft agar grown in the presence or absence of TGF-ß and/or LY294002 (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present study shows that TGF-ß stimulation results in the activation of class I PI3K in a subset of fibroblast cell lines. Maximal PI3K activity, as measured by the in vitro production of PIP, occurred between 30 and 60 minutes of TGF-ß treatment, with Akt phosphorylation evident by 60 minutes (Fig. 1A). Although PI3K activation was shown to be dependent on TGF-ßR kinase activity (data not shown), neither the p85 nor the p110 subunit of PI3K was directly phosphorylated by either the type I or II receptor (data not shown). However, a recent article by Yi et al. (42) suggested that TGF-ßRs indirectly associate with p85 and previous reports have indicated a physical interaction between PI3K subunits and the TGF-ßR complex under some conditions (43) but not in others (44). As such, the manner by which PI3K is activated after TGF-ßR binding is unknown.
The role(s) of PI3K in TGF-ß-mediated EMT is controversial (20, 26, 27). However, as TGF-ß is known to induce various phenotypes dependent on the cell context, it is extremely difficult to directly compare responses between distinct cell types and transdifferentiating cultures. For that reason, the current study was designed to examine the role(s) of PI3K activation in a subset of mesenchymal cell models, which similarly respond to TGF-ß by activating PI3K (Fig. 1C). Although we found no effect on PI3K signaling when TGF-ß was added to epithelial cultures (Fig. 1C), it would be premature to conclude that the observed cell tropism indicates a minor role for this pathway in epithelial cell growth and/or tumor progression. There is a large body of literature supporting a paracrine role for an activated stroma in the survival and proliferation of carcinomas (45, 46) and TGF-ß signaling in fibroblasts has been shown recently to modulate the oncogenic potential of specific adjacent epithelia (47, 48). The present results support this concept and indicate a signaling activity that might regulate this response. Moreover, as many carcinomas attain a more "fibroblastic" phenotype in response to TGF-ß during tumor progression (49), characterizing the targets and mediators capable of distinguishing these cell type differences is vital. PI3K may represent on such pathway.
Inhibition of PI3K with LY294002 or wortmannin had no appreciable effect on (a) the kinetics and intensity of Smad2 or Smad3 phosphorylation, (b) Smad2 or Smad3 nuclear translocation, or (c) Smad-dependent transcriptional responses (Fig. 2). Similarly, deletion/inactivation of Smad2 and/or Smad3 did not prevent Akt phosphorylation in response to TGF-ß (Fig. 3; data not shown). Although these findings support the contention that PI3K activation and Smad phosphorylation are independently regulated, a recent article by Runyan et al. (50) points out a potential caveat to generating conclusions based on phosphospecific antibodies directed to a single epitope. Although those investigators observed an inhibition of Smad3 transcriptional activity by LY294002 in mesangial cells, this was reported to be independent of the COOH-terminal serine phosphorylation on Smad3 induced by TGF-ß. However, as we observed no effect on either Smad2/3 phosphorylation or transcriptional activation by prior treatment with LY294002, this is consistent with distinct signals emanating from the TGF-ßR complex regulating PI3K and Smad responses.
A role for the FYVE domain in correctly localizing SARA has been documented (9, 31, 32). However, the concentration of LY294002 used in the present studies would primarily be expected to disrupt class I PI3Ks as it was likely too low to inhibit class III PI3Ks that generate PtdIns(3)P in vivo, the specific phosphoinositide recognized by FYVE domain–containing proteins like SARA. To investigate this issue further, we tested the effect of a range of PI3K inhibitor concentrations (0.1-200 µg/mL) on Smad2 or Smad3 phosphorylation in response to TGF-ß. Although no appreciable diminution in Smad phosphorylation was observed (data not shown), it should be noted that because class III PI3K production is constitutive the short pretreatment before ligand addition may have been insufficient to adequately reduce cellular levels of PtdIns(3)P to observe any response. Thus, although the present results do not directly address the role of phosphoinositides and SARA localization, the data clearly indicate that ligand-mediated PI3K activation by TGF-ß is not required for Smad signaling in the mesenchymal cell models tested (Figs. 1-3).
The role of receptor internalization in TGF-ß signaling is an area of active investigation. Whereas we and others have shown a requirement for clathrin-dependent endocytosis in Smad activation (30, 32, 33), other studies have suggested that TGF-ß or activin activates Smad2 in the absence of TGF-ßR internalization (51, 52). In addition, further complexity is added to this issue as a recent study reported a requirement for TGF-ßR endocytosis in Smad2 nuclear translocation and transcriptional activity but not Smad2 phosphorylation (53). Although it is currently unclear why such a variety of responses has been observed, two issues that need to be considered include (a) the use of various cell types (i.e., Mv1Lu, AKR-2B, Cos1, Cos7, mesangial, and HeLa) as TGF-ß's action is dependent on the cell context and (b) whether endocytic assays were done and quantitated in parallel with all signaling studies. This latter point is extremely important as the fraction of activated/endocytosed TGF-ßRs required to initiate Smad signaling is unknown. For instance, if internalization needs to be inhibited >80% to observe an endocytic requirement for Smad activation, unless one specifically documents that this has occurred it is difficult to directly compare the findings. In the current study, we have extended this endocytosis/signaling question to Smad-independent pathways. As shown in Fig. 4, although dominant-negative dynamin inhibited clathrin (i.e., transferrin) and caveolar (i.e., lactosylceramide) internalization by 
90% as well as Smad2 phosphorylation, no appreciable effect on Akt phosphorylation (Fig. 4A) or PAK2 activation (data not shown) was observed. This differential endocytic requirement suggests that PI3K and Smad activation occur at different cellular locales and provides the cell a unique opportunity to modulate the cellular response to TGF-ß under a variety of external stimuli.
The relation of PI3K in PAK family activation is complex as both PI3K-dependent and PI3K-independent roles have been reported in various cell systems (39, 54). When fibroblast cultures were treated with 20 µg/mL LY294002 or a dominant-negative PAK2 adenovirus, TGF-ß-stimulated PAK2 kinase activity was prevented, whereas the Akt inhibitor SH-5 or dominant-negative Akt or Pdk1 had no effect (Fig. 5B). Although these results are consistent with the hypothesis that PAK2 and Akt define distinct pathways downstream of PI3K, it is also possible that one or both targets reflect a nonspecific action of LY294002 on another kinase(s). To address this concern, we used morpholino antisense oligonucleotides to the regulatory p85 subunit of type I PI3K and showed that (a) p85 protein was decreased, whereas p110, class II p170, and class III PI3KC3 protein levels were unchanged; (b) TGF-ß was unable to stimulate Akt phosphorylation or PAK2 kinase activity; and (c) Smad2 phosphorylation in response to TGF-ß was unchanged or slightly enhanced, potentially reflecting some negative cross-talk between Smad-dependent and Smad-independent signaling (Fig. 5D). Together, the results support the conclusion that TGF-ß stimulates type I PI3K signaling in a subset of fibroblast (not epithelial) cell lines and is required for PAK2 activation.
In agreement with our previous data suggesting independent regulation of Smad and PI3K signals (Figs. 2-4), LY294002 had no effect on Smad2 or Smad3 phosphorylation (Figs. 2 and 5). Although PAK2 and PI3K activities are required for the morphologic and proliferative effects of TGF-ß on AKR-2B cells, they are not sufficient. For instance, constitutively active PI3K is unable to stimulate growth or induce a morphologic change unless TGF-ß is also present (data not shown). These findings agree with data from our laboratory and others (26, 27), indicating an equally important role for Smad signaling in these events.
It is currently unknown how PI3K is activated by the TGF-ßR complex and/or how this results in the activation of PAK2. Although the virally encoded Nef protein has been shown to act as a scaffold capable of facilitating PAK2 activation (36), analogous proteins have not been implicated in TGF-ß action. However, a model depicting our current understanding of the relation between PI3K and Smad pathways in fibroblasts is shown in Fig. 6C. Of note, whereas inhibition of PI3K abrogates TGF-ß-mediated PAK2 activation (Fig. 5A, B, and D), yet dominant-negative PAK2 is unable to prevent Akt phosphorylation (Fig. 5C), these findings are consistent with a model where PI3K is spatially upstream of PAK2 and independently regulates Akt and PAK2 activation. A similar action of PI3K independent of Akt has been shown for PAK1-mediated actin phosphorylation and cytoskeleton reorganization (55). It is clear that TGF-ß exerts its complex and dramatic effects on fibroblast cells through the action of multiple signaling pathways. Understanding the synergistic and/or antagonistic manner by which these events are regulated is critical if the various cell type–dependent phenotypes induced by TGF-ß are to be understood.
|
Acknowledgments |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Anita Roberts for the Smad3–/– MEF cell line, Dr. Jeff Pessin for the dominant-negative dynamin1 (Ad.DynK44A) adenovirus, Becky Bruinsma for excellent technical assistance, and Drs. Margarita Bracamonte, Stephen Murphy, Jennifer Brown, and Nandor Garameszegi for helpful comments.
Received 5/ 2/05. Revised 8/22/05. Accepted 9/13/05.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3ß1 integrin and PI3K. J Cell Biochem 2004;92:770–80.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  X. Liu, R. J. Kelm Jr., and A. R. Strauch Transforming Growth Factor {beta}1-mediated Activation of the Smooth Muscle {alpha}-Actin Gene in Human Pulmonary Myofibroblasts Is Inhibited by Tumor Necrosis Factor-{alpha} via Mitogen-activated Protein Kinase Kinase 1-dependent Induction of the Egr-1 Transcriptional Repressor Mol. Biol. Cell, April 15, 2009; 20(8): 2174 - 2185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Zuo and Y.-G. Chen Specific Activation of Mitogen-activated Protein Kinase by Transforming Growth Factor-{beta} Receptors in Lipid Rafts Is Required for Epithelial Cell Plasticity Mol. Biol. Cell, February 1, 2009; 20(3): 1020 - 1029. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Rahimi, M. Andrianifahanana, M. C. Wilkes, M. Edens, T. J. Kottom, J. Blenis, and E. B. Leof Distinct Roles for Mammalian Target of Rapamycin Complexes in the Fibroblast Response to Transforming Growth Factor-{beta} Cancer Res., January 1, 2009; 69(1): 84 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Gamell, N. Osses, R. Bartrons, T. Ruckle, M. Camps, J. L. Rosa, and F. Ventura BMP2 induction of actin cytoskeleton reorganization and cell migration requires PI3-kinase and Cdc42 activity J. Cell Sci., December 1, 2008; 121(23): 3960 - 3970. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Xia, H. Wang, S. Munk, J. Kwan, H. J. Goldberg, I. G. Fantus, and C. I. Whiteside High glucose activates PKC-{zeta} and NADPH oxidase through autocrine TGF-{beta}1 signaling in mesangial cells Am J Physiol Renal Physiol, December 1, 2008; 295(6): F1705 - F1714. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Liu and A. I. Gotlieb Transforming Growth Factor-{beta} Regulates in Vitro Heart Valve Repair by Activated Valve Interstitial Cells Am. J. Pathol., November 1, 2008; 173(5): 1275 - 1285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. D. Moore, T. Isayeva, G. P. Siegal, and S. Ponnazhagan Silencing of Transforming Growth Factor-{beta}1 In situ by RNA Interference for Breast Cancer: Implications for Proliferation and Migration In vitro and Metastasis In vivo Clin. Cancer Res., August 1, 2008; 14(15): 4961 - 4970. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Leipner, K. Grun, A. Muller, E. Buchdunger, L. Borsi, H. Kosmehl, A. Berndt, T. Janik, A. Uecker, M. Kiehntopf, et al. Imatinib mesylate attenuates fibrosis in coxsackievirus b3-induced chronic myocarditis Cardiovasc Res, July 1, 2008; 79(1): 118 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Dufour, X. Holy, and P. J. Marie Transforming growth factor-{beta} prevents osteoblast apoptosis induced by skeletal unloading via PI3K/Akt, Bcl-2, and phospho-Bad signaling Am J Physiol Endocrinol Metab, April 1, 2008; 294(4): E794 - E801. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cabello-Verrugio and E. Brandan A Novel Modulatory Mechanism of Transforming Growth Factor-beta Signaling through Decorin and LRP-1 J. Biol. Chem., June 29, 2007; 282(26): 18842 - 18850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kamiya, K. Sakakibara, E. J. Ryer, R. P. Hom, E. B. Leof, K. C. Kent, and B. Liu Phosphorylation of the Cyclic AMP Response Element Binding Protein Mediates Transforming Growth Factor {beta}-Induced Downregulation of Cyclin A in Vascular Smooth Muscle Cells Mol. Cell. Biol., May 1, 2007; 27(9): 3489 - 3498. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Suzuki, M. C. Wilkes, N. Garamszegi, M. Edens, and E. B. Leof Transforming Growth Factor {beta} Signaling via Ras in Mesenchymal Cells Requires p21-Activated Kinase 2 for Extracellular Signal-Regulated Kinase-Dependent Transcriptional Responses Cancer Res., April 15, 2007; 67(8): 3673 - 3682. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-P. Kuang, X.-H. Zhang, C. B. Rich, J. A. Foster, M. Subramanian, and R. H. Goldstein Activation of elastin transcription by transforming growth factor-beta in human lung fibroblasts Am J Physiol Lung Cell Mol Physiol, April 1, 2007; 292(4): L944 - L952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Yang, Y. Zhang, Y. Li, Z. Wu, and D. Zhu Myostatin Induces Cyclin D1 Degradation to Cause Cell Cycle Arrest through a Phosphatidylinositol 3-Kinase/AKT/GSK-3beta Pathway and Is Antagonized by Insulin-like Growth Factor 1 J. Biol. Chem., February 9, 2007; 282(6): 3799 - 3808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Seoane Escaping from the TGF{beta} anti-proliferative control Carcinogenesis, November 1, 2006; 27(11): 2148 - 2156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E. Wang, I. Shin, F. Y. Wu, D. B. Friedman, and C. L. Arteaga HER2/Neu (ErbB2) Signaling to Rac1-Pak1 Is Temporally and Spatially Modulated by Transforming Growth Factor {beta} Cancer Res., October 1, 2006; 66(19): 9591 - 9600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Wilkes and E. B. Leof Transforming Growth Factor beta Activation of c-Abl Is Independent of Receptor Internalization and Regulated by Phosphatidylinositol 3-Kinase and PAK2 in Mesenchymal Cultures J. Biol. Chem., September 22, 2006; 281(38): 27846 - 27854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Biswas, T. L. Criswell, S. E. Wang, and C. L. Arteaga Inhibition of Transforming Growth Factor-{beta} Signaling in Human Cancer: Targeting a Tumor Suppressor Network as a Therapeutic Strategy. Clin. Cancer Res., July 15, 2006; 12(14): 4142 - 4146. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
), p-AktS473 (
), and p-AktT308 (
) were normalized to total Akt at the indicated time points. Representative of three separate experiments. B, activation of PI3K does not require new transcription or translation. AKR-2B fibroblasts were grown to confluence, placed in serum-free medium for 24 hours, and treated with 20 µg/mL cycloheximide or 5 µg/mL actinomycin D for 60 minutes. Cultures were then stimulated with 5 ng/mL TGF-ß for 120 minutes before examining PI3K activity by assaying for p-AktS473 (top) or total Akt (bottom). C, PI3K activation occurs in fibroblast but not epithelial cells in response to TGF-ß. Three fibroblast (NIH3T3, Swiss3T3, and AKR-2B) and three epithelial (HeLa, Mv1Lu, and MDCK cells) cell lines were grown to confluence, placed in DMEM containing 0.5% FBS for 24 hours, and assayed for phosphorylated Akt (p-AktS473) and total Akt following addition (+) of 5 ng/mL TGF-ß for 120 minutes.
Pdk1) plasmids. Cultures were either untreated (-) or stimulated (+) with 5 ng/mL TGF-ß (+/– LY or SH-5) for 120 minutes. Following immunoprecipitation, the in vitro kinase activity of PAK2 (top, top) and Akt (middle, top) was determined using myelin basic protein as substrate for PAK2 and immunopurified GSK-3 as substrate for Akt. Separate aliquots were used for Western analysis against total PAK2 (top, bottom), Akt (middle, bottom), Smad2 (bottom, bottom), and phosphorylated Smad2 (bottom, top). C, inhibition of PAK2 does not prevent Akt activation. AKR-2B fibroblasts were grown to confluence and placed in serum-free medium in the absence (-) or presence (+) of adenovirus (MOI 300) expressing dominant-negative PAK2 as an enhanced GFP (EGFP) fusion (Ad-PAK2K278R). Following 24-hour incubation, cultures remained unstimulated (-) or TGF-ß was added to a final concentration of 5 ng/mL (+) for 90 minutes. PAK2 activity (bottom, top) was assessed by in vitro kinase assay and total PAK2 (bottom, bottom; PAK2 represents endogenous protein and EGFP-PAK2K278R the viral fusion protein), phosphorylated Akt (top, top) and total Akt (top, bottom) were determined using Western analysis. D, p85 antisense oligonucleotides inhibit PAK2 kinase activity stimulated by TGF-ß. Cos7 cells were plated at 1.0 x 105 per six-well dish and grown for 24 hours before transient transfection with either p85 antisense morpholino oligonucleotides (p85 MO) or scrambled control (Control MO). Total p85, p110, p170, and PI3KC3 protein was determined to validate effectiveness and specificity of the inhibited target. PAK2, Akt, and Smad2 activity and protein levels were determined as described in Materials and Methods following 60-minute stimulation in the absence (-) or presence (+) of 10 ng/mL TGF-ß.