If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
To whom correspondence should be addressed: Dept. of Neurology, University of Michigan, 200 Zina Pitcher Place, 4414 Kresge III, Ann Arbor, MI 48109-0588. Tel.: 734-763-7274; Fax: 734-763-7275; E-mail: [email protected]
Department of Neurology, University of Michigan, Ann Arbor, Michigan 48109
* This work was supported by National Institutes of Health Grants NS01938 (to J. W. R.), NS36778, and NS38849, grants from the Juvenile Diabetes Foundation Center of Excellence for the Study of the Complications of Diabetes, the American Diabetes Association (to E. L. F.), and the Program for Understanding Neurological Diseases.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Previously, we reported insulin-like growth factor-I (IGF-I) promotes motility and focal adhesion kinase (FAK) activation in neuronal cells. In the current study, we examined the role of IGF-I in Schwann cell (SC) motility. IGF-I increases SC process extension and motility. In parallel, IGF-I activates IGF-I receptor, insulin receptor substrate-1 (IRS-1), phosphatidylinositol 3 (PI-3)-kinase, and FAK. LY294002, a PI-3 kinase inhibitor, blocks IGF-I-induced motility and FAK phosphorylation. The Rho family of GTPases is important in the regulation of the cytoskeleton. Overexpression of constitutively active Leu-61 Cdc42 and Val-12 Rac1 enhances SC motility which is unaffected by LY294002. In parallel, stable transfection of SC with dominant negative Asn-17 Rac1 blocks IGF-I-mediated SC motility and FAK phosphorylation, implying Rac is an upstream regulator of FAK. Collectively our results suggest that IGF-I regulates SC motility by reorganization of the actin cytoskeleton via the downstream activation of a PI-3 kinase, small GTPase, and FAK pathway.
insulin-like growth factor-I
focal adhesion kinase
insulin receptor substrate-1
dorsal root ganglion
Dulbecco's modified essential media
scatter factor/hepatocyte growth factor
Insulin-like growth factor-I (IGF-I)1 is a polypeptide growth factor essential for normal nervous system development (
). IGF-I treatment of SH-SY5Y human neuroblastoma cells results in redistribution of the actin cytoskeleton with the formation of rapidly moving membrane ruffles. Ruffling is followed by protrusion of lamellipodia that adhere to specific extracellular matrix molecules and activate cell surface integrins (
Focal adhesion kinase (FAK) is a key focal adhesion protein present in focal adhesion sites. Focal adhesion sites are aggregations of protein complexes consisting of activated FAK that in turn recruits other proteins, including paxillin, p130Cas, vinculin, and talin. These protein complexes anchor the actin cytoskeleton and provide structural integrity to cells (
). Included in this family are Rho, Cdc42, and Rac. In fibroblasts, microinjection of wild-type Rac, a constitutively active mutant Rac, or GTPγS-bound (activated) Rac causes growth factor-independent actin polymerization and concomitant membrane ruffling and lamellipodial advance (
Little is known about the interactions between IGF-I, the small GTPases, and the focal adhesion proteins, including FAK, and how these interacting factors function to promote the motility and/or migration of cells in the nervous system during normal development and in response to injury (
). In the current study, we explored the downstream signaling pathways of IGF-I-mediated cellular motility and the roles of FAK and the small Rho family GTPases, Rac and Cdc 42 in Schwann cell (SC) motility. These cells migrate and attach to axons as the nervous system develops and in response to injury in the adult (
In the current study, IGF-I enhances motility of cultured rat SC. Cytochalasin D blocks IGF-I-induced SC motility, suggesting this process is dependent on rearrangement of the actin cytoskeleton. In SC, IGF-I activates mitogen-activated protein kinase and phosphatidylinositol 3 (PI-3)-kinase pathways, and FAK. An inhibitor of PI-3 kinase (LY294002) blocks the effects of IGF-I on cell motility and FAK phosphorylation, implying PI-3 kinase signaling is important for IGF-I-induced motility. To test the role of the small GTPases in IGF-I-mediated changes in motility, we used constitutively active or dominant negative mutants of Cdc42 and Rac. Cdc42 and Rac mediate IGF-I-induced cell motility, and both proteins are activated upstream of FAK. We then cocultured SC with dorsal root ganglion (DRG) neurons as an in vitro model of developing peripheral nerves. In this model, IGF-I enhanced attachment of SC to axons. Our results suggest that small GTPases and FAK may serve as linkers between growth factor activation and SC motility which, in turn, promotes the attachment of SC to axons in the nervous system.
Recombinant human IGF-I was a generous gift of Cephalon (West Chester, PA) and was stored in 100 mm acetic acid in −80 °C until use. All media, sera, and bovine pituitary extract were from Life Technologies, Inc. All other chemicals were purchased from Sigma.
Whole explant DRG from 15-day-old embryonic (E15) Harlan Sprague-Dawley rats (Madison, WI) were prepared as described previously (
). Explants were dissociated and plated at a density of approximately 10,000 cells per collagen-coated glass slide in media free of serum and insulin with 30 μm5-fluoro-2′-deoxyuridine to ensure removal of endogenous SC and fibroblasts (
). SC were cultured on poly-l-lysine-coated plates in culture media (DMEM containing 10% fetal bovine serum, 2 μm forskolin, and 10 μg/ml bovine pituitary extract). Cells were passaged upon confluency and used for 4 passages. For serum-free conditions, SC were washed twice and cultured in defined media (DM) (DMEM/F-12 media supplemented with transferrin (10 mg/liter), putrescine (10 μm), progesterone (20 nm), and sodium selenite (30 nm)).
Anti-IRS-1 antibody was a generous gift from Dr. Morris White (Joslin Diabetes Center, Boston). Anti-phosphotyrosine (G410) antiserum was purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). Anti-focal adhesion kinase (FAK), anti-p130Cas, and anti-phosphotyrosine (PY20) antisera were purchased from Transduction Laboratories (Lexington, KY). Anti-IGF-IR β subunit and anti-ERK1 antisera were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-S-100 and anti-neurofilament antisera were purchased from Chemicon (Temecula, CA).
Morphological Studies and Immunocytochemistry
To observe SC-axonal interactions in SC-DRG cocultures, SC were labeled with 10 μm dioctadecylindocarbocyanine (DiI, Molecular Probes Inc., Eugene, OR) per the manufacturer's protocol. After rinsing, 100,000 SC were added to each DRG culture dish in SHTE media (DMEM supplemented with transferrin (100 mg/liter), sodium selenite (30 nm), hydrocortisone (10 nm), β-estradiol (10 nm)) with the addition of ascorbic acid (10 mg/ml), and 10 ng/ml nerve growth factor ± 10 nm IGF-I for 6 or 12 h. Changes in SC morphology were documented with fluorescence microscopy (Leitz Fluorovert Microscope, W. Nuhsbaum Inc., McHenry, IL). The percent of SC with processes (processes longer than the width of a cell body) and SC attached on axons were measured.
To demonstrate localization of SC and axons in SC-DRG cocultures, unlabeled SC were cocultured with DRG and treated with IGF-I as described. The cells were then fixed with 4% paraformaldehyde and processed for double immunostaining for S-100 and neurofilament to demonstrate SC and axons, respectively. Briefly, cells were incubated with anti-S-100 (1:500, mouse) and anti-neurofilament (1:100, rabbit) for 2 h at room temperature. After rinsing, cells were incubated with rhodamine-conjugated anti-mouse and fluorescein isothiocyanate-conjugated anti-rabbit secondary antisera at 1:100 dilution for 1 h at room temperature, rinsed, and coverslipped, and immunofluorescence was documented with fluorescence microscopy.
For labeling the actin cytoskeleton, cells were fixed with 4% paraformaldehyde and stained with Oregon Green phalloidin (1:1000, Molecular Probes) for 15 min as suggested by the manufacturer.
Leu-61 Cdc42 and Asn-17 Cdc42 constructs in a pcDNA3 expression vector were generous gifts from Dr. Martin Schwartz (Scripps Research Institute, La Jolla, CA). Val-12 Rac1 and Asn-17 Rac1 constructs in a pSFFV expression vector were kindly donated by Dr. Lilli Petruzzelli (University of Michigan, Ann Arbor, MI).
). Briefly, 1 μg of cDNA was incubated with 10 μl of LipofectAMINE for 45 min before adding to cells cultured in DM. After 6 h, DM was replaced by culture media, and cells were then recovered for 3 days before being selected by 425 μg/ml G418. Cells were used for experiments after at least 4 weeks of G418 selection. Experiments were performed using cells from at least three different transfection pools.
SC motility was measured using a gold particle motility assay as described previously (
). Briefly, a uniform carpet of gold particles was prepared on bovine serum albumin-coated glass coverslips in 6-well plates. Colloidal gold-coated coverslips were rinsed and then seeded with 1 × 105SC in 2 ml of either DM or DM + 10 nm IGF-I ± 0.5 μm cytochalasin D or 10 μm LY294002. After 6 h, cells were fixed with 3.5% glutaraldehyde, air-dried, and mounted on glass slides with Permount (Fisher). The gold particle-free clear zone surrounding cells was photographed, and SC motility was represented as the average area of at least 100 pericellular clear zones, quantitated using an NIH Image program (
SC were serum-deprived for 6 h before IGF-I treatment. For experiments using LY294002, cells were pretreated with LY294002 30 min before IGF-I addition. SC were then rinsed with ice-cold phosphate-buffered saline and lysed in lysis buffer (50 mm Tris-HCl, pH 7.4, 1% Nonidet P-40, 150 mm NaCl, 1 mm EDTA, 1 mm phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin and leupeptin, 1 mm sodium orthovanadate, and 1 mmNaF). 500 μg of protein from each sample was immunoprecipitated with 4 μg/ml antisera overnight at 4 °C. Immunoprecipitates were separated by SDS-polyacrylamide gel electrophoresis (12.5%) and electrophoretically transferred to nitrocellulose membranes followed by immunoblotting procedures as described previously using primary antisera for phosphotyrosine (1 μg/ml PY20 and 0.4 μg/ml 4G10) (
). Briefly, the immunoprecipitates were rinsed and then incubated with 0.2 mg/ml phosphoinositides and 10–50 μCi of [γ-32P]ATP in reaction buffer (50 μm ATP, 10 mmMgCl2, 20 mm HEPES, pH 7.5) for 10 min. The lipid products were extracted by methanol/chloroform and loaded on oxylated silica gel plates (Analtech, Newark, DE). The phosphatidylinositides were separated by thin layer chromatography using a running solvent (chloroform/methanol/ H2O/ammonium chloride (30:13.5:5.65:1)).
IGF-I Induction of SC Motility Is Actin Cytoskeleton-dependent
To study the effects of IGF-I on SC motility, gold motility assays were performed. SC were plated on colloidal gold-covered coverslips in serum-free media (Fig.1A), 10 nm IGF-I (Fig. 1B), or 10 nm IGF-I + 0.5 μmcytochalasin D (Fig. 1C). Cytochalasin D is used to block actin cytoskeletal formation. Moving SC (Fig. 1, arrowheads) clear paths (Fig. 1, arrows) that were quantified using NIH image. The area of each path is representative of actual cell movement analyzed statistically to represent cell motility. Statistical analyses demonstrate there is a 2-fold increase of cell motility due to IGF-I treatment, and this increase can be blocked by cytochalasin D (Fig.1D). These results suggest that IGF-I-induced cell motility is an actin cytoskeleton-dependent event.
IGF-IR Signaling in SC
To investigate further the mechanisms of IGF-I-induced SC motility, we examined the IGF-IR signaling cascades in SC. SC were serum-deprived and treated with 10 nm IGF-I for 2–30 min. Tyrosine-phosphorylated proteins were then analyzed by Western blotting and immunoprecipitation. Anti-phosphotyrosine Western blotting demonstrated IGF-I-induced tyrosine phosphorylation of proteins with molecular masses of 42, 44, 66–69, 97, 110–140, and 185 kDa (Fig. 2A). Immunoprecipitation using specific antibodies revealed that these proteins included FAK (125 kDa), IGF-IR β subunit (97 kDa), and IRS-1 (185 kDa); all were tyrosine-phosphorylated after IGF-I treatment (Fig.2B). In addition to tyrosine phosphorylation, IGF-I also activates PI-3 kinase. To measure PI-3 kinase activity in SC, cells were serum-deprived and treated with or without 10 nm IGF-I for 2 min. Immunoprecipitation was performed using an anti-IRS-1 antibody, and the immunoprecipitates were processed for PI-3 kinase activity assay as described under “Experimental Procedures.” The formation of phosphatidylinositol phosphate and phosphatidylinositol diphosphate demonstrates PI-3 kinase activation. IGF-I (I) treatment activates PI-3 kinase of SC in comparison to untreated control SC (C) (Fig. 2C).
To observe changes in the actin cytoskeleton with IGF-I treatment, SC were serum-deprived and treated with IGF-I for 30 min. Cells were fixed and stained with fluorescein phalloidin which has high affinity for the actin cytoskeleton. Untreated SC maintain their actin stress fibers in cell bodies (Fig. 3A, arrows). After IGF-I treatment, there is dissociation of stress fibers in cell bodies (Fig. 3B, arrowhead), and SC extend long processes with concentrated actin filaments (Fig. 3B, arrows). We then determined the signaling pathways that mediate the effects of IGF-I on the SC cytoskeleton by pretreating SC with either a mitogen-activated protein kinase/extracellular signal-regulated kinase kinase inhibitor (PD98059), which blocks activation of both ERK1 and ERK2, or LY294002, an inhibitor of PI-3 kinase. Pretreatment of LY294002 (Fig. 3D) but not PD98059 (Fig. 3C) blocks IGF-I induced changes in the SC cytoskeleton (Fig. 3,C and D, arrows), suggesting PI-3 kinase mediates IGF-I-induced cytoskeletal changes.
PI-3 Kinase but Not MAP Kinase Pathway Mediates IGF-I-induced Cell Motility and FAK Phosphorylation
To study the roles of PI-3 kinase in SC motility, motility studies were performed using SC in serum-free conditions with increasing concentrations (0, 1, 10 nm), IGF-I ± 10 μm LY294002 for 6 h. IGF-I mediates a dose-dependent increase in SC motility which is blocked by LY294002, suggesting IGF-I-induced cell motility is mediated by PI-3 kinase (Fig.4A). In parallel, IGF-I enhancement of FAK phosphorylation is also prevented by LY294002 treatment (Fig. 4B), suggesting FAK is a downstream target of PI-3 kinase in IGF-IR-signaling cascades that regulate SC motility. In contrast to LY294002, PD98059 has no effect on IGF-I-induced cell motility (Fig. 4C) or FAK phosphorylation (data not shown).
IGF-I-induced SC Motility Is Mediated by Cdc42 and Rac Downstream of PI-3 Kinase
The Rho family of small GTPases, including Rho, Rac, and Cdc42, are instrumental in the organization of the actin cytoskeleton (
). We examined the roles of Cdc42 and Rac in SC motility. To localize Cdc42 and Rac in SC, immunostaining of Cdc42 and Rac was performed in SC in DM (control) or DM + 10 nm IGF-I for 30 min. In controls, both Cdc42 (Fig.5C) and Rac (Fig.5A) immunoreactivity are cytoplasmic. After 30 min of IGF-I treatment, Cdc42 relocalizes from the cytoplasm into filopodia (Fig.5D, arrow), whereas Rac immunoreactivity is now detected in membrane ruffles (Fig. 5B, arrow).
To explore further the roles of Cdc42 and Rac in SC motility, SC were stably transfected with constitutively active Leu-61 Cdc42 or Val-12 Rac1. Both constitutively active mutants enhanced SC base-line motility, 1.5-fold for Leu-61 Cdc42 and 1.7-fold for Val-12 Rac1-transfected SC (Fig. 6, Aand B). LY294002 had no effect on the motility of Leu-61 Cdc42 and Val-12 Rac1-transfected SC; however, LY294002 slightly decreases the motility of control cells. IGF-I treatment further increased cell motility of both control and mutant-transfected cells. Our data suggest Cdc42 and Rac participate in IGF-I-induced cell motility downstream of PI-3 kinase. In support of our findings using active mutants, dominant negative Asn-17 Cdc42 or Asn-17 Rac1 transfection blocks both IGF-I-induced cell motility (Fig.7A) and FAK phosphorylation (Fig. 7B).
IGF-I Enhances SC Migration toward Axons in SC/DRG Coculture
Previously, we reported IGF-I enhances SC differentiation and myelination (
). We observed that SC migration toward axons is essential for myelin sheath formation in SC/DRG coculture. In the current study, IGF-I promotes the attachment of SC on axons in serum-free media; the SC extend processes and move toward axons in 6 h (Fig. 8A). SC then align on and attach to axons (Fig. 8B), and gradually the myelin sheath is formed (
). We used DiI-labeled SC to quantify SC-axonal attachment. IGF-I treatment produces a 2-fold increase in the SC as follows: 1) with processes, and 2) in contact with axons, in comparison with control, untreated SC (Fig.8C). To study if SC migration toward axons is due to IGF-I-induced chemotaxis, transwell inserted chambers were used. SC were cultured in the upper inserted chambers, and DRG were cultured in the bottom chambers under serum-free conditions. The porous (8 μm pore diameter) membrane between the inserted chamber and bottom chamber serves as a barrier for SC to penetrate in response to chemotactants. 10 nm IGF-I was administered either in the media of inserted chambers, the bottom chambers, or both. Under all conditions tested, IGF-I failed to induce SC migration through the membranes, suggesting SC migration is not due to IGF-I-induced SC chemotaxis toward axons but to IGF-I enhancement of SC motility which increases the probability for SC-axonal contact (data not shown).
). Our results suggest IGF-I is also a motility factor for SC. In the current studies, IGF-I enhances SC motility which facilitates the contact of SC with axons, and, as we have previously reported leads to nervous system myelination (
To understand the mechanisms underlying IGF-I-mediated SC motility, we initially examined the effects of IGF-I on SC architecture. Phalloidin staining of actin reveals that IGF-I treatment reorganized the SC actin cytoskeleton leading to SC process extension. These findings are in agreement with our recent reports (
) in kidney mesangial cells. In both cell types, IGF-I treatment results in reorganization of the actin cytoskeleton and membrane ruffling.
Cytochalasin D treatment blocks the effects of IGF-I on SC motility, suggesting that the rearrangement of the SC actin cytoskeleton after IGF-I treatment is essential to initiate motility. Similar findings are reported in microglial cells (
). In our studies, IGF-I reorganization of the SC actin cytoskeleton and subsequent enhancement of motility is temporally associated with activation of the IGF-IR and FAK. Our results support the idea that growth factors that reshape the cytoskeleton also activate focal adhesion proteins allowing cells to crawl over their substrates (
Our results with IGF-I in SC are in agreement with recent reports in non-nervous system cells using other growth factors. Scatter factor/hepatocyte growth factor (SF/HGF) increases motility and induces FAK phosphorylation in fibroblasts (
) reported PI-3 kinase activation is required for migration of epithelial cells after treatment with human macrophage-stimulating protein. PI-3 kinase also mediates SF/HGF-induced motility in kidney cells (
) find that Rac mediates nerve growth factor and phorbol 12-myristate 13-acetate-induced lamellipodia formation and cell migration in PC12 cells. By using constitutively active mutants (Leu-61 Cdc42 and Val-12 Rac1), we find both Cdc42 and Rac are downstream of PI-3 kinase. These results agree with data from Hoosehmand-Rad and colleagues (
); in their studies platelet-derived growth factor mediates rearrangement of the actin cytoskeleton and motility in porcine aortic endothelial cells via activation of PI-3 kinase and more downstream activation of Rac. Indeed, Tolias and colleagues (
) speculate Cdc42 is upstream of Rac in actin cytoskeletal reorganization. Asn-17 Cdc42 blocks both bradykinin-induced filapodial extension and membrane ruffling; however, Asn-17 Rac inhibits only membrane ruffling. Our data would suggest both Cdc42 and Rac1 play similar roles in IGF-I-induced motility.
In our studies, stable transfection of SC with dominant negative Asn-17 Rac1 blocks IGF-I-mediated FAK phosphorylation, implying Rac is an upstream regulator of FAK. Several proteins, including FAK, may interact with Rac to mediate changes in the actin cytoskeleton (
) find that inhibition of Rho activity in rat pancreatic acini blocks epidermal growth factor-induced FAK phosphorylation and integrity of the actin cytoskeleton. We failed to coimmunoprecipitate FAK with Rac (data not shown), suggesting in SC there are intermediate molecules between Rac and FAK. In support of this idea, Taylor and colleagues (
) recently discovered a GTPase-activating protein (Graf) contains an SH3 domain that associates with FAK, providing a potential link between Rac and FAK.
In addition to playing important roles in growth factor receptor signaling of cellular motility, PI-3 kinase, small GTPases, and FAK are also essential for integrin-mediated cell motility. Like growth factor receptor activation, integrin ligand binding by extracellular matrix proteins stimulates cytoskeletal rearrangement and cell motility (
). Thus, both IGF-IR and integrin-signaling cascades activate PI-3 kinase, small GTPases, and FAK, suggesting bidirectional communication between IGF-IR and integrin signaling (Fig. 9). In support of this idea, occupancy of αVβ3integrin inhibits IGF-I signaling in vascular smooth muscle cells (
). We are currently examining the interactions between integrin and IGF-I signaling in SC and speculate that cross-talk between the downstream signals of integrin ligand binding and IGF-IR activation are essential for not only cell motility but also cell survival in the nervous system.
We thank Dr. Cunming Duan and Dr. Xiping Xin for technical assistance and Judy Boldt for secretarial assistance.