Differential responses to nerve growth factor and epidermal growth factor in neurite outgrowth of PC12 cells are determined by Rac1 activation systems.

Neurite outgrowth of PC12 cells is induced by nerve growth factor (NGF) but not by epidermal growth factor (EGF). This differential response has been explained by the duration of mitogen-activated protein kinase (MAPK) activation; NGF induces sustained MAPK activation but EGF leads short-lived activation. However, precise mechanisms have not yet been understood. Here we demonstrate the difference between NGF and EGF in regulation of Rac1, a small GTPase involved in neurite outgrowth, in PC12 cells. NGF phosphoinositide 3-kinase dependently induces transient activation of Rac1 and accumulation of active Rac1 at protrusion sites on the cell surface, inducing filamentous actin-rich protrusions and subsequent neurite formation in a Rac1-dependent manner. On the other hand, EGF phosphoinositide 3-kinase independently induces more transient Rac1 activation but neither accumulates active Rac1 nor forms Rac1- and filamentous actin-rich protrusions. Difference in the Rac1 localization between NGF and EGF was also observed with the localization of exogenously expressed green fluorescent protein-tagged Rac1. The Rac1-mediated protrusion by NGF is independent of MAPK cascade, but the subsequent neurite extension requires the cascade. Thus, the differential activation of Rac1 and localization of active Rac1 contribute to the difference in the ability of NGF and EGF to induce neurite outgrowth, and we propose that the MAPK cascade-independent prompt activation of Rac1 and recruitment of active Rac1 at the protrusion sites trigger the initiation of neurite formation.

Rat pheochromocytoma PC12 cells have been used as a model system for neuronal differentiation and neurite outgrowth. After stimulation with nerve growth factor (NGF), 1 they stop growing and begin to extend neurites. In contrast, epidermal growth factor (EGF) does not induce neurite outgrowth but stimulates proliferation of PC12 cells. The receptors for NGF and EGF belong to a family of tyrosine kinase receptors, and they transduce signals via similar signal transduction pathways, including a Ras-dependent mitogen-activated protein kinase (MAPK) cascade (1)(2)(3). It has been proposed that the sustained activation of Ras and MAPK by NGF is involved in neuronal differentiation of PC12 cells (2,4,5). However, a previous study reported that the receptor-mediated sustained activation of MAPK alone is insufficient to induce neurite outgrowth (6). Thus, an additional signaling pathway is suggested to be required for the NGF-induced neurite outgrowth. Morphological analysis of NGF-and EGF-stimulated PC12 cells revealed that stimulation with NGF, but not with EGF, induces the rapid formation of filamentous actin (F-actin)-rich protrusions, followed by the extension of neuritic processes with growth conelike structures at their tips (7,8). Furthermore, the rapid redistribution of F-actin induced by NGF was reported to be suppressed by an inhibitor of phosphoinositide 3-kinase (PI3K) (9), and overexpression of a constitutively active mutant of PI3K was shown to induce neuritic process formation (10,11), suggesting the involvement of PI3K in the NGF-induced neurite outgrowth. However, precise mechanisms involved in the cytoskeletal reorganization required for the NGF-induced neurite outgrowth have not yet been understood.
The Rho family of small GTPases, including Rho, Rac, and Cdc42, serves as molecular switches by cycling between an inactive GDP-bound state and an active GTP-bound state, and has been implicated in the reorganization of actin cytoskeleton in various cell types (12,13). In PC12 cells, activation of Rho induces the growth cone collapse and the retraction of neurites (14,15). In contrast, studies using a dominant negative Rac1 show that Rac1 is involved in the NGF-induced neurite outgrowth (16,17). However, it remains obscure how NGF regulates Rac1, inducing neurite outgrowth. Here we demonstrate that NGF PI3K dependently induces Rac1 activation and formation of cell surface protrusions where active Rac1 and Factin are accumulated, whereas EGF PI3K independently activates Rac1 but fails to form Rac1-and F-actin-rich protrusions. We propose that the differential activation and localization of Rac1 contribute to the difference in the ability of growth factors to induce neurite outgrowth.
Construction of Expression Plasmids-Mammalian expression vector pEF-BOS was kindly provided by Dr. S. Nagata (Osaka University). Human Rac1 was obtained as described previously (18). cDNA for Ha-Ras was obtained from Health Science Research Resources Bank (Osaka, Japan). Ras N17 and Rac1 N17 were generated by polymerase chain reaction-mediated mutagenesis (19) and fused in-frame with a sequence encoding an initiating methionine followed by the Myc epitope tag sequence at the NH 2 terminus contained in pEF-BOS. Green fluorescent protein (GFP)-Rac1 was obtained by insertion of the coding sequence of wild-type Rac1 into pEGFP-C1 (CLONTECH). The coding sequence for the Cdc42/Rac interacting binding (CRIB) domain (amino acids 70 -150) of rat ␣PAK (20) was obtained by reverse transcriptasepolymerase chain reaction from PC12 cells, using primers 5Ј-AAGG-GATTCAAGGAGCGGCCAGAGATTTCT-3Ј containing a BamHI site and 5Ј-GAAGAATTCTAATCTTAAGCTGACTTATCT-3Ј containing a stop codon followed by an EcoRI site. The polymerase chain reaction product was subcloned into the BamHI/EcoRI sites of pGEX-4T-2 (Amersham Pharmacia Biotech) and sequenced. The CRIB domain of ␣PAK was then expressed in Escherichia coli as a fusion protein with glutathione S-transferase (GST), purified on glutathione-Sepharose beads, and isolated from the beads with 16 mM reduced glutathione. The purified proteins were dialyzed with 25 mM Tris-HCl, pH 7.5, 1 mM MgCl 2 , 0.2 mM dithiothreitol, and 5% glycerol, and stored at Ϫ80°C.
Cell Culture and Transfection-PC12 cells were cultured in Dulbecco's modified Eagle's medium containing 5% fetal bovine serum, 10% horse serum, 4 mM glutamine, 100 units/ml penicillin, and 0.2 mg/ml streptomycin under humidified conditions in 95% air and 5% CO 2 at 37°C. For immunofluorescence analysis, cells were seeded onto poly-Dlysine (Sigma)-coated glass coverslips (circular, 13 mm) in 24-well plates at a density of 2.5 ϫ 10 4 cells/well. Transient transfections were carried out using LipofectAMINE 2000 (Life Technologies Inc.) according to the manufacturer's instructions. Transfected cells were fixed 48 h after transfection.
Measurement of Rac1 and Cdc42 Activities-Measurement of Rac1 and Cdc42 activities was performed according to the modified method of Benard et al. (21). PC12 cells were seeded in 100-mm culture dishes at a density of 1 ϫ 10 7 cells/dish, cultured for 24 h, and serum-starved in serum-free Dulbecco's modified Eagle's medium for 12 h. Cells were then stimulated with 50 ng/ml NGF or 200 ng/ml EGF for the indicated times and lysed for 5 min with the ice-cold cell lysis buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 2 mM MgCl 2 , 1% Nonidet P-40, 10% glycerol, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin, and 1 g/ml leupeptin) containing 8 g of GST-CRIB. Cell lysates were then centrifuged for 5 min at 10,000 ϫ g at 4°C, and the supernatant was incubated with glutathione-Sepharose beads for 30 min at 4°C. After the beads were washed with the cell lysis buffer, the bound proteins were eluted in Laemmli sample buffer and separated by 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The separated proteins were electrophoretically transferred onto a polyvinylidene difluoride membrane (Millipore Corp.). The membrane was blocked with 3% low fat milk in Tris-buffered saline, and then incubated with a mouse monoclonal anti-Rac1 (1:1000 dilution) or rabbit polyclonal anti-Cdc42 antibody (1:100 dilution). The Rac1 and Cdc42 antibodies were detected using horseradish peroxidase-conjugated goat anti-mouse IgG and goat anti-rabbit IgG antibodies (DAKO), respectively, and the ECL detection kit (Amersham Pharmacia Biotech). Densitometry analysis was performed using NIH Image software, and the amounts of GST-CRIB-bound Rac1 and Cdc42 were normalized to the total amounts of Rac1 and Cdc42 in cell lysates, respectively.
Immunofluorescence Microscopy-All steps were carried out at room temperature, and cells were rinsed with phosphate-buffered saline (PBS) between each step. At the indicated times, cells on coverslips were fixed with 3.7% formaldehyde, PBS for 15 min. After residual formaldehyde had been quenched with 50 mM NH 4 Cl, PBS for 10 min, cells were permeabilized in 0.2% Triton X-100, PBS for 10 min, and incubated with 10% fetal bovine serum in PBS for 30 min to block nonspecific antibody binding. Endogenous Rac1 was stained with an anti-Rac1 monoclonal antibody in PBS at a 1:1000 dilution for 1 h followed by the incubation with a rhodamine-conjugated donkey antimouse IgG (Chemicon International Inc.) in PBS at a 1:500 dilution for 1 h. For detection of cells expressing Myc-tagged Ras N17 or Rac1 N17 , cells were incubated with an anti-Myc monoclonal antibody 9E10 (0.5 g/ml) in PBS for 1 h followed by the incubation with a rhodamineconjugated donkey anti-mouse IgG in PBS for 1 h. In the case of double stainings with Myc-tagged Ras N17 and endogenous Rac1, expressed Myctagged Ras mutants were visualized using a rabbit polyclonal anti-Myc antibody (MBL) in PBS at a 1:500 dilution for 1 h followed by a fluorescein isothiocyanate-conjugated donkey anti-rabbit IgG (Chemicon International Inc.) in PBS at a 1:250 dilution for 1 h. F-actin was stained with Alexa 488-or rhodamine-conjugated phalloidin (Molecular Probes) in PBS (0.5 units/ml) for 1 h. Cells were mounted in 90% glycerol containing 0.1% p-phenylenediamine dihydrochloride in PBS. Confocal microscopy was performed using an MRC-1024 laser scanning confocal imaging system (Bio-Rad Laboratories) equipped with a Nicon Eclipse E800 microscope and a Nicon Plan Apo 60 ϫ 1.4 oil immersion objective.
To detect active Rac1 in PC12 cells, we made use of the CRIB domain of rat ␣PAK with Myc epitope tag at the NH 2 terminus. Using cell culture overlay technique described by Merilä inen et al. (22) and Kulkarni et al. (23), fixed PC12 cells were incubated with Myc-CRIB domain (100 g/ml) for 1 h and the active Rac1 was detected using anti-Myc polyclonal antibody.

Activation of Rac1 by NGF and EGF in PC12
Cells-Consistent with previous reports (16,17), expression of dominant negative Rac1 completely inhibited the neurite outgrowth induced by NGF in PC12 cells ( Fig. 1), indicating that the activity of Rac1 is critical for the NGF-induced neurite outgrowth. To obtain direct evidence of the activation of endogenous Rac1 by NGF, we measured the amount of cellular GTP-bound Rac1 using the GST-fused CRIB domain of rat ␣PAK (GST-CRIB), which specifically binds to Rac in its active GTP-bound state (20,21). NGF induced a rapid increase in the amount of cellular GTP-bound Rac1, the elevation reaching the maximum at 3 min (Fig. 2, a and e). The level decreased gradually but remained above the basal for over 60 min after the stimulation. In contrast, EGF induced a more transient activation of Rac1 within 1 min, and then the level quickly returned to the basal level within 5 min after the stimulation (Fig. 2, b and e). Cdc42 has been also reported to be implicated in neurite outgrowth similar to Rac1 (16). We then measured the amount of cellular GTP-bound Cdc42 using GST-CRIB. However, the level of GTP-bound Cdc42 was not significantly affected by either NGF or EGF (Fig. 2, c, d, and e).
Redistribution of Rac1 Induced by NGF-We next followed time-dependent changes in the subcellular distribution of Rac1 and F-actin after the stimulation with NGF and EGF. Immunofluorescence staining of Rac1 using an anti-Rac1 antibody revealed that in unstimulated cells Rac1 was present throughout the cytoplasm and at the cell surface ( Fig. 3a and Table I). At 1 min after stimulation with NGF, cells spread with ruffles around the cell periphery, and Rac1 was co-localized with Factin in membrane ruffles. At 3 min after the addition of NGF, corresponding to the time of the highest Rac1 activity, cells produced two to four cell surface protrusions, and Rac1 and F-actin were accumulated and co-localized at these protrusions. Within 15 min, these F-actin-rich protrusions had begun to extend and form the short processes where Rac1 remained to be accumulated, and the co-localization of F-actin and Rac1 at the tips of processes remained unchanged up to 30 min after the NGF stimulation. On the other hand, stimulation with EGF induced ruffles around the cell periphery at 1 min after the stimulation, and Rac1 and F-actin were co-localized to the ruffled area (Fig. 3b). In sharp contrast to the NGF stimulation, the F-actin-and Rac1-rich ruffles declined within 3 min, and protrusions and process formation did not occur. To confirm the NGF-induced dynamic change in the localization of Rac1 observed above, PC12 cells were transfected with NH 2terminal GFP-tagged Rac1, and the fluorescence of GFP was monitored after the stimulation with NGF. GFP-Rac1 was also rapidly accumulated at the F-actin-rich protrusions in response to NGF with similar time course of the movement of endogenous Rac1 (Fig. 4).
To determine whether the Rac1 accumulated at protrusions was the GTP-bound active form, we immunostained for active Rac1 using a Myc-CRIB domain which specifically binds to GTP-bound active Rac1. As shown in Fig. 5, NGF stimulation induced the active Rac1 accumulation at the protrusions where the active Rac1 and F-actin were co-localized. On the other hand, EGF induced the recruitment of active Rac1 to the cell periphery but the active Rac1 accumulation at the protrusions was not observed. The CRIB domain can bind to GTP-bound active form of Cdc42 as well as that of Rac1, but NGF did not increase GTP-bound active Cdc42 (Fig. 2e). Therefore, Rac1 accumulated at the protrusions by NGF is the GTP-bound   [3][4][5][6]8, and 9 Cells exhibiting redistribution of Rac1, GFP-Rac1, or F-actin were defined as the cells that possessed at least one protrusion positively stained with anti-Rac1 antibody, anti-Myc antibody, or phalloidin, and were scored as percentages of the total cells (Figs. 3, 5, and 8) or the total number of transfected cells (Figs. 4, 6, and 9). At least 100 cells were assessed in one experiment, and data are the mean Ϯ S.E. of triplicate experiments. Time Rac1, Fig. 3 GFP-Rac1, Fig. 4

Rac1 Activation System Involved in Neurite Formation
active form of Rac1. We next examined the requirement of Rac1 in the formation of F-actin-rich protrusions by using a dominant negative mutant of Rac1, Rac1 N17 . The transient expression of Rac1 N17 completely inhibited the NGF-induced rapid redistribution of F-actin and the formation of protrusions (Fig. 6). This result indicates that the rapid cytoskeletal response to NGF, the formation of F-actin-rich protrusions at the cell surface, required the activation of Rac1.
Involvement of PI3K in the NGF-induced Activation and Redistribution of Rac1-To determine whether the activation of MAPK cascade or PI3K was involved in the activation of Rac1, we examined the effect of LY294002 (30 M) or wortmannin (1 M), specific inhibitors of PI3K, or PD98059 (25 M), a specific inhibitor of MAPK kinase (also known as MEK), on the NGF-and EGF-induced Rac1 activation. Pretreatment with LY294002 or wortmannin markedly inhibited the NGF-induced activation of Rac1 at 1 and 3 min (Fig. 7, a and c), whereas the pretreatment did not significantly affect the EGFinduced Rac1 activation or the inhibition was very weak (Fig. 7,  b and d). To confirm the action of the PI3K inhibitors, we examined the effects of PI3K inhibitors on the NGF-induced Akt phosphorylation, a known PI3K effect, by using anti-phospho-specific Akt antibody. LY294002 and wortmannin completely inhibited the NGF-induced Akt phosphorylation (Fig.  7f). These results indicate that the NGF-induced Rac1 activation is fully dependent on PI3K, whereas the EGF-induced activation is independent of PI3K. On the other hand, PD98059 had no effect on either the NGF-or the EGF-induced activation of Rac1 (Fig. 7, a, b, and c), although it could suppress the NGFand the EGF-induced activation of MAPK (Fig. 7g). These results indicate that the activation of MAPK cascade is not required for the activation of Rac1. We further examined whether NGF activated GFP-Rac1 in a PI3K-dependent manner. NGF activated exogenously expressed GFP-Rac1 and this activation was inhibited by LY294002 (Fig. 7e).
We next examined the involvement of PI3K and MAPK in the rapid redistribution of F-actin and Rac1 after the stimulation with NGF. A previous study reported that the rapid Factin redistribution induced by NGF was suppressed by wort-

Rac1 Activation System Involved in Neurite Formation
mannin (9). Consistent with this result, we found that the NGF-induced rapid F-actin redistribution was suppressed by LY294002 (Fig. 8a) and wortmannin (data not shown). In addition to the effect on F-actin redistribution, the NGF-induced rapid redistribution of Rac1 was also inhibited by LY294002 (Fig. 8a) and wortmannin (data not shown). The effect of PI3K inhibitors on the rapid redistribution of Rac1 was also examined in GFP-Rac1-transfected cells, and the NGF-induced redistribution of GFP-Rac1 was inhibited by pretreatment with LY294002 (Fig. 8b) and wortmannin (data not shown). These results indicate that the NGF-induced activation and rapid redistribution of Rac1 requires the PI3K activity.
In contrast, when cells were pretreated with PD98059, NGF could induce the accumulation of F-actin and Rac1 at the cell surface protrusions (Fig. 8a). However, in the presence of PD98059, cells did not produce even short processes by 15 min (Fig. 8a) or by 30 min (data not shown) after NGF stimulation, and they did not extend neurites (data not shown). The accumulation of F-actin and Rac1 at the protrusions in the PD98059pretreated cells had disappeared within 60 min (data not shown). These results indicate that the activation of MAPK is not required for the NGF-induced rapid redistribution of F-actin and Rac1 but is crucial for the subsequent neurite outgrowth.
Redistribution of Rac1 by NGF Requires Ras-Small GTPase Ras is a critical component in the signaling pathway of NGFinduced PC12 cell differentiation (24 -26). To examine whether Ras is involved in the rapid redistribution of F-actin and Rac1 induced by NGF, PC12 cells were transiently transfected with a dominant negative mutant of Ras, Ras N17 . Expression of Ras N17 completely suppressed the rapid redistribution of Factin (Fig. 9a) and Rac1 (Fig. 9b) induced by NGF. To confirm the effect of Ras N17 expression on the localization of Rac1, we co-transfected cells with Ras N17 and GFP-Rac1. Ras N17 completely suppressed the NGF-induced redistribution of GFP-Rac1 as well (Fig. 9c). We further examined the effect of the dominant negative Ras on the NGF-induced GFP-Rac1 activation. Ras N17 completely suppressed the NGF-induced GFP-Rac1 activation (Fig. 7e). These results indicate that the activity of Ras is required for the rapid redistribution of F-actin and Rac1 induced by NGF in PC12 cells.

DISCUSSION
PC12 cells differentially respond to NGF and EGF in neuronal differentiation, but the precise mechanism for this difference remains to be defined. In this study, we focused our attention on the regulation of Rac1 activity which is thought to be required for neuronal morphology (16,17), and we demonstrate here that NGF but not EGF PI3K dependently activates Rac1 and recruits active Rac1 to the protrusion sites, initiating the neurite formation. We suggest that a spatial determination mechanism of the signaling pathway involved in the activation of Rac1 plays a critical role in the initiation of the neurite outgrowth in PC12 cells.
In PC12 cells, NGF induces neurite outgrowth and eventual cessation of cell division, whereas EGF leads to a proliferative signal without neurite formation, although their receptors transduce signals via similar signal transduction pathways, including MAPK cascade. These differential responses are thought to be determined by the duration of MAPK activation; NGF induces sustained MAPK activation for several hours, but EGF leads short-lived activation (2). However, the receptormediated persistent activation of MAPK alone is shown to be insufficient to induce neurite outgrowth (6). Furthermore, a recent study showed that although the sustained activation of MAPK is mediated by small GTPase Rap1, expression of a dominant negative mutant of Rap1 blocks the sustained activation of MAPK but fails to inhibit the NGF-induced neurite outgrowth (27). These results indicate that the NGF-induced neurite outgrowth is not merely determined by the sustained MAPK activation and another mechanism must exist for the neurite outgrowth in PC12 cells. In this study, we have shown that NGF PI3K dependently activated and accumulated Rac1 at the F-actin-rich protrusions on the cell surface, whereas   9. Effect of Ras N17 expression on the NGF-induced rapid redistribution of Rac1 and F-actin. a, PC12 cells were transiently transfected with an expression vector encoding Myc-tagged Ras N17 . 36 h after transfection, they were serum-starved for 12 h, and then stimulated with NGF for 3 min. The cells were fixed and stained with an anti-Myc polyclonal antibody (left panels) and Alexa-488-conjugated phalloidin (right panels). b, after the Ras N17 -transfected cells had been stimulated with NGF for 3 min, they were fixed and stained with an anti-Myc polyclonal antibody (left panels) and an anti-Rac1 monoclonal antibody (right panels). c, PC12 cells were co-transfected with expression vectors encoding Myc-Ras N17 and GFP-Rac1. 36 h after transfection, the cells were serum-starved for 12 h, and then stimulated with NGF for 3 min. The cells were fixed and stained with an anti-Myc polyclonal antibody (left panels). Localization of GFP-Rac1 was examined by the fluorescence of GFP (right panels). The results shown are representative of three independent experiments. The bar represents 10 m. EGF PI3K independently activated Rac1 but failed to form Rac1-and F-actin-rich protrusions. Furthermore, we revealed that Rac1 accumulated at the protrusions by NGF was active Rac1. Considering the inhibition by dominant negative Rac1 of the NGF-induced formation of F-actin-rich protrusions and process formation, it is inferred that PI3K-dependent activation of Rac1 and accumulation of active Rac1 at the specific sites on the cell surface induces actin reorganization at the sites, resulting in the formation of the cell surface protrusions and subsequent extension of neurites. From these results taken together, we speculate that temporally and spatially regulated Rac1 activation initiates the neurite formation.
Previous studies have suggested the involvement of PI3K in the NGF-induced neurite outgrowth in PC12 cells (9,28). Here, we showed that inhibition of PI3K activity suppressed the activation and redistribution of Rac1 induced by NGF. Therefore, one role of PI3K in neurite outgrowth in PC12 cells is to regulate the activation and the localization of Rac1 in response to NGF. In this study, we found that dominant negative Ras could inhibit the NGF-induced redistribution of Rac1. Accumulated evidence shows that Ras induces the activation of Rac through PI3K (29 -31). Therefore, we suggest that there is a hierarchy of activation from Ras to Rac1 through PI3K in the NGF signaling pathway, and that this signaling pathway is important for the initiation of the neurite outgrowth by NGF. Activation of PI3K leads to the production of phosphatidylinositol-3,4-P 2 and phosphatidylinositol-3,4,5-P 3 . These products can bind to the pleckstrin homology domain of some GEFs for Rac1, such as Sos and Vav, and stimulate the activity of GEFs (32,33). Sos is well known to be located in the downstream signaling pathways of NGF (34), and Sos, known as a GEF for Ras, activates Rac through its NH 2 -terminal Rac GEF domain containing the tandem Dbl homology and pleckstrin homology domaina, and Ras-mediated PI3K activation and subsequent binding of PI3K products to the pleckstrin homology domain appear to be necessary for the activation of Rac by Sos (29,32). The PI3K-dependent Sos activation may be involved in the regulation of the Rac1 activity by NGF in PC12 cells. In addition to the modulation of the GEF activity, the pleckstrin homology domain is known to play an important role in the localization of signaling molecules, including GEFs for Rac1 (35). We showed here that the NGF-induced redistribution of Rac1 depends on the PI3K activity. Therefore, it is conceivable that PI3K recruits a Rac1 GEF via association with the pleckstrin homology domain and induces the accumulation and activation of Rac1 at the protrusion sites on the cell surface. In contrast to NGF, EGF activated Rac1 in a PI3K-independent manner, indicating the existence of at least two distinct mechanisms in the activation of Rac1, PI3K-dependent and -independent pathways, in PC12 cells. NGF and EGF may utilize distinct GEFs for activation of Rac1 in PC12 cells. A previous report indicated the PI3K-independent, but protein kinase Cdependent activation of Rac2 in neutrophils (36). Phosphorylation of a GEF for Rac1 by a protein kinase may be involved in the activation of Rac1 by EGF.
The involvement of MAPK cascade in neuronal differentiation has been extensively investigated, and the MAPK cascade was shown to be required for the NGF-induced neurite outgrowth in PC12 cells (4,5,37). We demonstrated here that inhibition of MAPK cascade by a specific inhibitor of MAPK kinase had no effect on the formation of F-actin-rich protrusions as well as the activation and the redistribution of Rac1 induced by NGF. However, this inhibitor suppressed the subsequent induction of short processes and eventual neurite outgrowth. These results suggest that the MAPK cascade is required for the extension of neurites but not for the initiation of neurites induced by NGF, and that the initiation step and the subsequent extension step of the neurite outgrowth are regulated by different mechanisms.
In conclusion, we demonstrate here that NGF and EGF differentially activate Rac1 in PI3K-dependent and -independent manners, respectively, and this PI3K-dependent activation of Rac1 and accumulation of active Rac1 to the protrusion sites on the cell surface initiate the neurite formation in PC12 cells. This initial marked activation of Rac1 and its recruitment to the protrusion sites are the initial steps for the formation of neurites and the steps do not require MAPK cascade. This work takes a close-up of an important role of Rac1 in the initiation of neurites, and will help to elucidate the molecular mechanism of neurite formation and extension.