Rac1 and Cdc42 but Not RhoA or Rho Kinase Activities Are Required for Neurite Outgrowth Induced by the Netrin-1 Receptor DCC (Deleted in Colorectal Cancer) in N1E-115 Neuroblastoma Cells*

Netrins are chemotropic guidance cues that attract or repel growing axons during development. DCC (deleted in colorectalcancer), a transmembrane protein that is a receptor for netrin-1, is implicated in mediating both responses. However, the mechanism by which this is achieved remains unclear. Here we report that Rho GTPases are required for embryonic spinal commissural axon outgrowth induced by netrin-1. Using N1E-115 neuroblastoma cells, we found that both Rac1 and Cdc42 activities are required for DCC-induced neurite outgrowth. In contrast, down-regulation of RhoA and its effector Rho kinase stimulates the ability of DCC to induce neurite outgrowth. In Swiss 3T3 fibroblasts, DCC was found to trigger actin reorganization through activation of Rac1 but not Cdc42 or RhoA. We detected that stimulation of DCC receptors with netrin-1 resulted in a 4-fold increase in Rac1 activation. These results implicate the small GTPases Rac1, Cdc42, and RhoA as essential components that participate in signaling the response of axons to netrin-1 during neural development.

Netrins are a small family of secreted proteins that guide growing axons during neural development (1,2). The first netrin cloned, UNC-6, was identified using a genetic screen for mutations affecting axon guidance in Caenorhabditis elegans (3). Netrins were first identified in vertebrates on the basis of their ability to promote commissural axon outgrowth from explants of embryonic spinal cord (4,5). Netrin family members have now been identified in multiple vertebrate and invertebrate species and shown to have a highly conserved function as axon guidance cues (6). Netrins are bifunctional molecules attracting and repelling different classes of axons. Growth cone attraction mediated by netrin-1 involves the transmembrane netrin receptor DCC (7,8). In C. elegans, the identification of UNC-5 first implicated it as a receptor required for the repellent response to UNC-6 (9). Three UNC-5 homologs have now been identified in mammals (10,11). Current evidence suggests that netrin-mediated repulsion requires the function of both UNC-5 and DCC family members in some and perhaps all cases, suggesting that UNC-5 and DCC may form a netrin receptor complex (12).
The intracellular mechanisms mediating the response of an axon to netrin-1 are currently unclear. Previous studies indicate that extracellular guidance cues induce the neuronal growth cone to advance, retract, or turn by regulating the actin cytoskeleton within the growth cone (13). The Rho family of small GTPases, in particular, RhoA, Rac1, and Cdc42, are well established regulators of actin reorganization in non-neuronal cells (14), and there is now compelling evidence demonstrating roles for RhoA, Rac1, and Cdc42 as signaling elements within the neuronal growth cone (15,16). Here, we report that members of the Rho family of GTPases are required for commissural axon outgrowth produced by netrin-1 from explants of embryonic rat spinal cord. Both Rac1 and Cdc42 are required for neurite outgrowth promoted by DCC in N1E-115 neuroblastoma cells, and in contrast, inhibition of RhoA and Rho kinase increases the ability of DCC to induce neurite outgrowth. In Swiss 3T3 cells, DCC was found to trigger actin reorganization through activation of Rac1 in a netrin-1-dependent manner. In fibroblasts, DCC did not activate Cdc42 or RhoA.

EXPERIMENTAL PROCEDURES
Explant Assay-Embryonic day 13 rat dorsal spinal cord and floor plate explants were dissected and cultured in three-dimensional collagen gels as described (5). Recombinant chick netrin-1 protein was produced and purified as described (5). Toxin B was purified as previously described (17). Both netrin-1 protein and toxin B were added to the culture medium at the beginning of the culture period. The explants were cultured for 14 h, then fixed with 4% paraformaldehyde, and photographed with an Optronics MagnaFire camera and a Carl Zeiss Axiovert microscope using a 20ϫ objective lens and phase contrast 5optics. The length of axon fascicles growing out of the explants were quantified using Northern Eclipse Software (Empix Imaging). The total length of fascicle growth was then calculated for each explant.
DNA Constructs-Standard DNA protocols were used as described (18). pRK5-DCC-C (3394 -4690 bp) was generated by digestion of pBS-DCC with EcoRI and BglII followed by ligation of a EcoRI-BglII fragment into pRK5 digested with EcoRI and BamHI. To generate pRK5 encoding full-length DCC, a fragment (3061 bp) from the start codon to the EcoRI site of pBS-DCC was amplified by PCR and subcloned into pRK5-DCC-C digested with EcoRI. pDCC-E encoding the N terminus of DCC (1122 amino acids) comprising the extracellular and transmembrane domains tagged with green fluorescent protein (GFP) 1  DNA was purified using a Qiagen kit. For microinjection studies, purified plasmids were filtered through a 0.2-m cellulose acetate membrane (Corning) before microinjection into cells.
Reverse Transcriptase (RT)-PCR-Total RNA was purified using Trizol (Invitrogen) and poly(A) ϩ RNA was isolated using the Oligotex mRNA purification kit (Qiagen). First strand cDNA was synthesized using superscript reverse transcriptase (Invitrogen). PCR was used to amplify cDNA using the following primers: UNC5 h1, GGA ATT CCC  TCC CTC GAT CCC AAT GTG T and TCC CCG CGG GGC AGG GAA  CGA AAG TAG T, 909 bp; UNC5 h2, GCT CTA GAG TCG CGG CAG  CAG GTG GAG GAA and GGA ATT CAG GGG GCG GCT TTT AGG  GTC GTT, 771 bp; and DCC, CCG CTC GAG TGG TCA CCG TGG GCG  TTC TCA and GGC TGG ATC CTC TGT TGG CTT GTG, 938 bp. The primers were annealed at 60°C, and 35 cycles of amplification were carried out. The size of the predicted amplification product is indicated.
Cell Culture and Microinjection-Mouse fibroblast Swiss 3T3 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 5% fetal calf serum and antibiotics and maintained in an atmosphere of 10% CO 2 . Confluent serum-starved Swiss 3T3 cells were prepared as described (19). Briefly, the cells were plated in 5% serum at a density of 6 ϫ 10 4 onto acid-washed coverslips. 7-10 days later, the cells became quiescent and were subjected to serum starvation for 16 h in Dulbecco's modified Eagle's medium containing 2 g/liter NaHCO 3 . The eukaryotic expression vector pRK5 encoding full-length DCC or truncated DCC (pRK5-DCC-C) was microinjected alone or with pRK5 encoding Myc-tagged Cdc42N17, RacN17, or pEFmyc-C3 transferase at 0.1 mg/ml into the nucleus of ϳ100 cells over a period of 20 min in CO 2 -independent medium (Invitrogen) using an Eppendorf microinjection system 5246. pDCC-E or pEGFP were microinjected at 0.1 mg/ml into the nucleus of ϳ100 cells. During microinjection, the cells were maintained at 37°C within a humidified atmosphere. The cells were returned to the incubator for a further 5 h followed by the addition of purified netrin-1 at 500 ng/ml for up to 30 min.
Mammalian Cell Transfection-N1E-115 neuroblastoma cells and COS-7 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and antibiotics at 10% CO 2 . N1E-115 cells were plated onto coverslips previously coated with laminin (20 g/ml; VWR Canlab) for 24 h at 37°C, washed twice with water, and left to air dry. Transfection was carried out with LipofectAMINE transfection reagent (Invitrogen) according to the manufacturer's protocol. Briefly, the cells were incubated in serum-free medium for 1 h. During this time, pRK5, pRK5-DCC-C, pDCC-E, pRK5-DCC (0.4 g) with or without pRK5myc-RacN17 or -Cdc42N17, or pEFmyc-C3 transferase (0.2 g) were mixed with LipofectAMINE reagent and incubated for 15 min at room temperature followed by the addition of the transfection mix to the cells. 6 h later, the transfection mix was replaced with Dulbecco's modified Eagle's medium containing 5% fetal calf serum and incubated for 12 h with or without the blocking antibodies PN3 against netrin-1 (25 g/ml) (52) (provided by Dr. Tim Kennedy) and DCC (10 g/ml) (AF5, Cedarlane Laboratories, Ltd.) or mouse IgG (25 g/ml) before fixation in freshly prepared 4% (w/v) paraformaldehyde for 10 min. When indicated, the cells transfected with pRK5 or pRK5-DCC were incubated with 10 M Y-27632 compound for 2 h before fixation. COS-7 cells were transfected using the DEAE-dextran method as described previously (20). The amounts of plasmid used per 100-mm dish were as follows: pRK5, 5 g; pRK5-DCC, 5 g; and pRK5myc-Rac1 or -Cdc42, 1.5 g. 24 h after transfection, the cells were serum-starved overnight and treated with netrin-1 (500 ng/ml) for different periods of time.
Immunofluorescence Microscopy-At the indicated times, microinjected Swiss 3T3 cells or transfected N1E-115 cells were rinsed with phosphate-buffered saline and fixed for 10 min in freshly prepared 4% (w/v) paraformaldehyde. All steps were carried out at room temperature, and coverslips were rinsed in phosphate-buffered saline between each of the step. The cells were permeabilized in 0.2% Triton X-100 for 5 min, and free aldehyde groups were reduced with 0.5 mg/ml sodium borohydride for 10 min. The cells were double labeled following the procedure previously described (21). Briefly, the cells were incubated with the primary monoclonal antibodies anti-DCC (Pharmingen, G97-449), anti-Myc (a generous gift from Dr. Nicole Beauchemin, McGill University), or anti-GFP (Molecular Probe) diluted in phosphate-buffered saline for 60 min. Then coverslips were transferred to a second antibody mixture composed of fluorescein isothiocyanate-conjugated goat anti-mouse antibody (Sigma) and tetramethylrhodamine isothiocyanate-conjugated phalloidin (Sigma) for 60 min. In N1E-115 cells, a neurite was defined as a process that measured at least the length of the cell body and stained positively for neurofilament M using a polyclonal anti-neurofilament 150 (Chemicon). Coverslips were mounted by inverting them onto 8 l of Mowiol (Calbiochem) mountant containing p-phenylenediamine as an anti-bleach reagent. After 2 h at room temperature, the coverslips were examined on a Zeiss Axiovert 135 microscope using Zeiss oil immersion 63ϫ objective lens. Fluorescence images were recorded using a digital camera (DVC) and analyzed with Northern Eclipse software (Empix Imaging Inc.).
Purification of GST-PAK and GST-WASP-GST-PAK (amino acids 56 -272) and GST-WASP (amino acids 201-321) were used to isolate GTP-bound Rac1 and Cdc42, respectively. Escherichia coli transformed with GST-PAK and GST-WASP constructs were grown at 37°C to an absorbance of 0.5. Expression of the fusion proteins was induced by isopropyl-␤-D-thiogalactopyranoside (1 mM) for 3 h at 37°C. The cells were washed once in STE buffer (100 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA) prior to sonication in buffer A (20 mM Hepes, pH 7.5, 120 mM NaCl, 2 mM EDTA, 10% glycerol, 10 g/ml aprotinin, 10 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride). The lysates were cleared by centrifugation, and Nonidet P-40 was added to a final concentration of 0.5%. The proteins were stored at Ϫ80°C until use. Protein concentration was determined by comparison with different amounts of bovine serum albumin using SDS-PAGE. For each sample, 10 -15 g of GST-PAK or GST-WASP was purified using glutathione-Sepharose beads (Sigma) for 30 min at 4°C. The beads were washed twice with buffer A, and the protein lysates were added as described below.
Rac/Cdc42 GTP Loading Assay-COS-7 cells cotransfected with pRK5 encoding DCC and Myc-tagged Rac1 or Cdc42 were serumstarved overnight and treated with purified netrin-1 for different periods of time. The cells were lysed in 25 mM Hepes, pH 7.5, 1% Nonidet P-40, 10 mM MgCl 2 , 100 mM NaCl, 5% glycerol, 1 mM sodium vanadate, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, and 10 g/ml leupeptin. The protein concentrations were determined using a Bio-Rad protein assay kit. An equal amount of proteins was incubated for 1 h at 4°C with GST-PAK or GST-WASP (10 -15 g) purified on glutathione-Sepharose beads in binding buffer (25 mM Hepes, pH 7.5, 30 mM MgCl 2 , 40 mM NaCl, 0.5% Nonidet P-40, 1 mM dithiothreitol) as described (22). The beads were washed three times with washing buffer (25 mM Hepes, pH 7.5, 30 mM MgCl 2 , 40 mM NaCl, 1 mM dithiothreitol) and twice with washing buffer containing 1% Nonidet P-40 before being boiled in SDS sample buffer. The proteins were separated on 12% SDS-PAGE. GTPbound Rac1 and Cdc42 were revealed by immunoblotting using the anti-Myc antibody and ECL reagent detection kit (Amersham Biosciences). The levels of GTP-bound Rac1 and Cdc42 in each sample were assessed using densitometry.

Rho GTPases
Are Required for Commissural Axon Outgrowth Evoked by Netrin-1-To determine whether Rho GT-Pases are involved in mediating the axon outgrowth promoting activity of netrin-1, we examined the effect of adding the Rho GTPase inhibitor toxin B to explants of E13 rat dorsal spinal cord cultured in a three-dimensional collagen gel in the presence of netrin-1. Explants of dorsal spinal cord cultured in the presence of recombinant netrin-1 (200 ng/ml) produced maximal commissural axon outgrowth from these explants as previously reported (4) (Fig. 1, A and C). The addition of increasing concentrations of toxin B from 0.001 to 1 ng/ml in the presence of maximal concentrations of netrin-1 resulted in increasing inhibition of commissural axon outgrowth from explants (Fig.  1, A and D-G). In the presence of toxin B, cells at the edge of the explants were clearly rounded (Fig. 1E, arrowhead), a characteristic effect of toxin B on other cell types (23). In addition, although a small amount of axon outgrowth occurred at higher concentrations of toxin B, the axons were much less fasciculated than normal (Fig. 1, compare C with D and E). These findings indicate that netrin-1-mediated commissural axon outgrowth requires the activity of one or more Rho GTPases.  (24). Using immunoblotting analyses, N1E-115 cells were found to constitutively express netrin-1. However, these cells did not express DCC ( Fig. 2A). RT-PCR analysis revealed the expression of mRNAs encoding the netrin-1 receptors UNC5 h1 and UNC5 h2 but not DCC (Fig. 2B). In the presence of 5% serum, the cells are round and extend lamellipodia and multiple filopodia (Fig.  3G). When DCC is expressed in N1E-115 cells in the presence of 5% serum, 62% of transfected cells exhibited neurite outgrowth (Fig. 4). The majority of DCC-expressing cells contained one long neurite (ϳ30 m) per cell with thin filopodia along the neurite (Fig. 3, A and B). DCC protein was consistently enriched at the extending tip of the neurite as shown in Fig. 3A (arrow). The presence of antibodies blocking the function of DCC or netrin-1 inhibited the ability of DCC to induce neurite outgrowth in N1E-115 cells (Fig. 4). In addition, truncated DCC proteins lacking the majority of the extracellular domain (DCC-C) or the cytoplasmic domain (DCC-E) of DCC were unable to produce neurite outgrowth (Fig. 4). These results strongly suggest that netrin-1 binding to DCC is necessary to mediate intracellular signaling events leading to neurite outgrowth in N1E-115 cells.
C3 transferase has been shown to inactivate RhoA by ADPribosylation at residue Asn 41 (23). When C3 transferase is expressed in N1E-115 cells, 50% of transfected cells showed neurite outgrowth as previously reported (53) (Fig. 4). When DCC is expressed in the presence of C3 transferase, the number of transfected cells with neurite outgrowth increased to 80% (Fig. 4). Similarly, when DCC-expressing cells are incubated with the Y-27632 compound that inhibits the Rho effector Rho kinase, neurite outgrowth is stimulated in more than 80% of transfected cells (Fig. 4). These results suggest that inhibition of RhoA and its effector Rho kinase known to mediate the effects of RhoA on neurite retraction in N1E-115 cells (51) increases the ability of DCC to stimulate neurite extension in N1E-115 cells.
Expression plasmids encoding dominant negative RacN17 or Cdc42N17 were transfected together with pRK5-DCC into N1E-115 cells. As shown in Fig. 3 (D and F), both dominant negative Rac1 and Cdc42 significantly inhibited neurite outgrowth induced by DCC. In DCC-expressing cells, the dominant negative Rac1 and Cdc42 mutants reduced neurite extension by 55 and 45%, respectively (Fig. 4). Cells expressing both RacN17 and DCC were rounded and flattened and exhibited long filopodia but not lamellipodia. These findings suggest that although Rac1 has been inhibited, Cdc42 remained activated in these cells (Fig. 3D). Cells expressing both Cdc42N17 and DCC exhibited lamellipodia and short microspikes at the plasma membrane, suggesting that Rac1 remained activated in these cells (Fig. 3F). Therefore, both Rac1 and Cdc42 activities are required for neurite outgrowth induced by DCC in N1E-115 cells.
The Netrin-1 Receptor DCC Activates Rac1 but Not Cdc42 or RhoA in Swiss 3T3 Fibroblasts-To further dissect the mechanisms used by netrin-1 to signal through Rho GTPases, we reconstituted the phenomenon in Swiss 3T3 fibroblasts by transiently expressing DCC and using the organization of the actin cytoskeleton as a functional read-out. Following serum starvation, Swiss 3T3 cells lose most of the actin based structures usually found in a fibroblast: lamellipodia, filopodia, and stress fibers. However, the cells do remain attached to the supporting extracellular matrix (Fig. 5A). Microinjection of constitutively active Cdc42L61, RacL61, and RhoL63 proteins into quiescent, serum-starved Swiss 3T3 cells has been shown to rapidly induce the formation of three distinct actin based structures: filopodia, lamellipodia, and stress fibers, respectively (19,(25)(26)(27). In addition, in some cell types, such as fibroblasts and epithelial cells, activation of Cdc42 leads to rapid activation of Rac1, which in turn leads to activation of RhoA (26). Netrin-1 and DCC proteins were undetectable by Western blot analyses of Swiss 3T3 cell lysates ( Fig. 2A). RT- PCR analyses detected the expression of mRNAs encoding UNC5 h2 but not UNC5 h1 or DCC. As shown in Fig. 5B, the addition of recombinant netrin-1 protein does not affect the reorganization of polymerized actin in uninjected cells. Microinjection of the eukaryotic expression vector, pRK5, encoding full-length rat DCC into quiescent, serum-starved Swiss 3T3 cells led to the expression of DCC (Fig. 5C), and no spontaneous reorganization of actin was observed (Fig. 5D). However, 10 min after the addition of 500 ng/ml of purified netrin-1 to the medium, assemblies of polymerized actin were detected at the leading edge of the plasma membrane in DCC expressing cells. These developed into lamellipodia and membrane ruffles as shown in Fig. 5F (arrows). The minimum concentration of netrin-1 required to cause actin reorganization in DCC expressing cells was 100 ng/ml (data not shown). However, optimal effects were obtained at 500 ng/ml of netrin-1. 30 min after the addition of netrin-1, in cells expressing DCC, actin assembled into stress fibers that traverse the cell (Fig. 5H). Recently, it has been reported that netrin-1 binds to the extracellular domain of DCC (28). A truncated DCC protein lacking the majority of the extracellular domain did not induce actin reor-ganization after the addition of netrin-1 for 30 min, suggesting that netrin-1 binding to DCC is essential to activate Rho GTPase signaling pathways (Fig. 6B). Similarly, the expression of a truncated DCC protein lacking the cytoplasmic domain and coupled to green fluorescent protein did not lead to actin reorganization after the addition of netrin-1 (Fig. 6D). We conclude that DCC is essential to activate the cascade of Rho GTPases in Swiss 3T3 cells in a ligand-dependent manner.
To determine whether DCC activates the cascade of Rho GTPases through Cdc42 or Rac1 in Swiss 3T3 fibroblasts, we microinjected quiescent, serum-starved Swiss 3T3 cells with pRK5-DCC together with eukaryotic vectors encoding either Myc-tagged dominant negative RacN17 or Cdc42N17 or C3 transferase. As shown in Fig. 7, the expression of dominant negative Cdc42N17 did not inhibit actin reorganization induced by netrin-1 in DCC-expressing cells (Fig. 7, compare D  with B), whereas dominant negative RacN17 inhibited the formation of both lamellipodia and stress fibers (Fig. 7F). C3 transferase blocked the formation of stress fibers but not the formation of polymerized actin at the leading edge of the plasma membrane (Fig. 7H). Hence, DCC activates Rac1-induced signaling pathways but not Cdc42-dependent signals in Swiss 3T3 fibroblasts. These findings indicate that in these cells, activation of RhoA by DCC is a consequence of cross-talk between Rac1 and RhoA.
The Netrin-1 Receptor DCC Promotes Rac1 GTP Loading-Pull-down assays were carried out in which Rac1 and Cdc42 GTP loading was assessed by specific binding of the active GTPases to the Cdc42/Rac interactive binding domain (29) of p65 PAK or WASP (Wiskott-Aldrich syndrome protein) fused to glutathione S-transferase (GST-PAK or GST-WASP), respectively. DCC and Myc-tagged Rac1 or Cdc42 were coexpressed in COS-7 cells for 24 h, and the cells were serum-starved overnight followed by the addition of netrin-1 to the medium. The lysates were prepared, and the amount of Rac1 or Cdc42 precipitated with GST-PAK or GST-WASP, respectively, was determined by Western blot analysis. Netrin-1 stimulated a 4-fold increase in the level of activated Rac1 (Fig. 8, A and B), whereas no increase in GTP-Cdc42 was observed after stimulation with netrin-1 (Fig. 8, C and D). The cells expressing Rac1 in the absence of DCC showed no increase in Rac1-GTP after 5 min of stimulation with netrin-1, suggesting that DCC is required for Rac1 activation. These data are consistent with the microinjection studies in Swiss 3T3 cells, suggesting that netrin-1 receptor DCC activates Rac1 but not Cdc42 in fibroblasts. DISCUSSION Netrin-1 and its receptor, DCC, are widely expressed in embryonic and adult tissues (4, 30 -32, 52). Their function in many cell types is poorly understood, but in the embryonic central nervous system they act as attractive and repulsive cues that guide the migration of developing axons (8). Here we demonstrate that toxin B inhibits commissural axon outgrowth evoked by netrin-1, thereby implicating Rho GTPases in mediating the effect of netrin-1 on these axons. Both Rac1 and Cdc42 were found to be necessary for DCC-induced neurite outgrowth in N1E-115 neuroblastoma cells. When RhoA and Rho kinase were inhibited, respectively, by C3 transferase or Y-27632 in N1E-115 cells, 80% of DCC-expressing cells exhibit neurite outgrowth, suggesting that down-regulation of RhoA and Rho kinase is required for DCC to induce neurite outgrowth in N1E-115 cells. In fibroblasts, the expression of DCC triggered actin reorganization in a netrin-1-dependent manner through the activation of Rac1 but not RhoA or Cdc42. Netrin-1 stimulation of DCC resulted in a 4-fold increase of Rac1 activation without affecting the level of activated Cdc42. Interestingly, these results suggest that a neuronal-specific guanine nucleotide exchange factor required for DCC to activate Cdc42 may be absent in fibroblasts. Alternatively, a specific coreceptor for netrin-1 that is required for DCC to activate Cdc42 may not be expressed in fibroblasts. Altogether, this study provides compelling evidence for a key role of regulated activities of Rac1, Cdc42, and RhoA in the cytosolic signaling mechanisms induced by DCC when it binds to netrin-1. Consistent with our findings, it has been reported that some of the defects caused by an activated form of the C. elegans DCC homolog, UNC-40, could be partly suppressed by mutations in Ced-10, a member of the Rac family in C. elegans (33).
In addition to the formation of lamellipodia in DCC-expressing fibroblasts, DCC also induced the formation of stress fibers as a result of cross-talk between Rac1 and RhoA. A current model suggests that attractive guidance cues activate Rac1 or Cdc42 and inhibit RhoA to promote directed axonal outgrowth, whereas repulsive cues inhibit Rac1 or Cdc42 and stimulate RhoA to induce retraction (15,16). In support of this model, Wahl et al. (34) showed that Ephrin-A5 activates Rho and inhibits Rac in cultured retinal ganglion cells. Here we propose that when a growth cone is attracted by netrin-1, DCC may activate Rac1 while inhibiting RhoA in neuronal cells. It may be the case that the activation of RhoA by Rac1 in fibroblasts reported here is restricted to non-neuronal cells.
The implication that second messengers, Ca 2ϩ and cAMP, modulate the response to netrin-1 has emerged from in vitro studies of growth cone turning using Xenopus spinal neurons (35)(36)(37)(38). Using the same assay, coactivation of phosphatidylinositol 3-kinase and phospholipase C␥ pathways were shown to be required for the turning response of the growth cone (39). Phosphatidylinositol 3-kinase mediates activation of Rac1 downstream of many tyrosine kinase receptors (40). However, it has not yet been determined whether phosphatidylinositol 3-kinase links DCC to activation of Rac1 upon binding to netrin-1. Protein kinase A phosphorylation of RhoA and the intracellular level of cAMP negatively regulate the activity of RhoA in different cell types (41), and the inhibition of RhoA and of its effector, Rho kinase, is required for cAMP-induced outgrowth of dendrites in B16 cells (42). The effects shown here FIG. 7. Netrin-1 receptor DCC activates Rac1 but not RhoA or Cdc42 in Swiss 3T3 fibroblasts. Serum-starved Swiss 3T3 cells were microinjected with pRK5-DCC alone (A and B) or with pRK5 encoding Myc-tagged Cdc42N17 (C and D), RacN17 (E and F), or pEF-mycC3 transferase (G and H). 5 h after microinjection, netrin-1 (500 ng/ml) was added to the medium for 30 min. F-actin (B, D, F, and H) and DCC (A, C, E, and G) were visualized as in Fig. 3. Myc-tagged proteins were revealed by costaining with an anti-Myc antibody and by indirect immunofluorescence (not shown). Approximately 100 cells were microinjected per coverslip. Scale bar, 10 m. The arrows indicate membrane ruffles. mediated by DCC may be a consequence of a coordinated activation of Rac1 leading to actin polymerization at the advancing edge of the growth cone and inactivation of RhoA through the maintenance of the intracellular levels of cAMP in neurons.
The cytoplasmic domain of DCC did not interact physically with Rac1 (data not shown), suggesting an indirect link between DCC and Rac1. A candidate protein that may link DCC to activation of Rac1 is the UNC-73 ortholog Trio, a guanine nucleotide exchange factor with activity toward both Rac1 and RhoA (43), found to play a major role in axonal development and pathfinding (44 -48). The cytoplasmic tail of DCC contains several putative SH3-binding motifs, PXXP (49), that may interact with the two SH3 domains of Trio or to an SH3-containing adapter molecule.
Before its discovery as an axon guidance receptor in the development of the nervous system, DCC was identified as a tumor suppressor gene in colorectal cancer and appears to activate signaling pathways affecting both cell proliferation and differentiation (30,31,50). Consistent with Rho proteins as potential oncogenes playing important roles in the development of cell transformation and metastasis (40), the identification of Rho GTPase activities in DCC-induced signaling pathways now provides new insight into unraveling the molecular mechanisms underlying the tumor suppressor function of DCC.