Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nat Cell Biol. Author manuscript; available in PMC Mar 16, 2010.
Published in final edited form as:
PMCID: PMC2839190

Netrin-1 Mediates Neuronal Survival Through PIKE-L Interaction With the Dependence Receptor UNC5B


Netrins, a family of secreted molecules, play critical roles in axon guidance and cell migration during neuronal development 1,2. In addition to its role as a chemotropic molecule, netrin-1 also acts as a survival factor 37. Both UNC5 (i.e. UNC5A, B, C or D) and DCC are transmembrane receptors for netrin-18,9. In the absence of netrin-1, DCC and UNC5 act as dependence receptors and trigger apoptosis 3,6,10. However, how netrin-1 suppresses the apoptotic activity of the receptors remains elusive. Here, we show that netrin-1 induces interaction of UNC5B with the brain specific GTPase PIKE-L. This interaction triggers activation of PI 3-kinase signaling, prevents UNC5B’s pro-apoptotic activity and enhances neuronal survival. Moreover, this process tightly relies on Fyn as PIKE-L is tyrosine phosphorylated in response to netrin-1 and the netrin-1-mediated interaction of UNC5B with PIKE-L is inhibited in Fyn null mice. Thus, PIKE-L acts as a downstream survival effector for netrin-1 through UNC5B in the nervous system.

PIKE-L is a brain specific GTPase, which binds and stimulates PI 3-kinase in a GTP-dependent manner 11,12. PIKE-L binds Homer, an adaptor protein for metabotropic glutamate receptor (mGluR). Activation of mGluRIs enhances formation of an mGluRI-Homer-PIKE-L complex, leading to activation of PI 3-kinase and prevention of neuronal apoptosis 13. PIKE is also a substrate for caspases. PIKE can be phosphorylated on tyrosine residues by Fyn, leading to its resistance to caspase cleavage 14. To search for PIKE-L-binding proteins, we conducted yeast two-hybrid screening using GTPase domain as bait. Four out of twelve clones are both His and β-gal positive, one of which encodes the C-terminus of UNC5B (Figure 1A). In HEK293 cells, transfected GFP-PIKE-L selectively binds to 569–946 fragment of UNC5B but not other fragments. Compared to the binding by the C-terminal motif (a.a. 569–946), truncation of death domain (a.a. 854–946) decreases UNC5B affinity to PIKE-L. Reciprocal immunoprecipitation reveals that the interaction occurs no matter PIKE-L or UNC5B is precipitated by its antibody (Figure 1B, middle panels). Full-length UNC5B and its C-terminal fragment released after caspase cleavage 5,7 specifically interact with GTPase domain but not with other regions of PIKE-L, consistent with our yeast two-hybrid findings (Figure 1B, right panels). We also observed the robust interaction between endogenous PIKE-L and UNC5B in both the cortex and hippocampus of rat brain (Figure 1C). Immunostaining of hippocampal and cortical primary neurons reveals that PIKE-L and UNC5B colocalize in the cell body and throughout all neuronal processes (Figure 1D, left panel). The staining is specific as GST-PIKE-L (a.a. 268–384) antigen but not control GST abolishes PIKE-L staining in neurons (Figure 1D, right panels).

Figure 1
PIKE-L interacts with UNC5B

To examine whether netrin-1 modulates PIKE-L interaction with UNC5B, we cotransfected UNC5B into HEK293 cells with wild-type or dominant-negative PIKE-L-KS (K413AS414N). This mutant cripples PIKE-L’s GTPase activity. It binds PI 3-kinase but fails to activate it 11,12. Compared to control, netrin-1 elicits PIKE-L interaction with UNC5B; by contrast, PIKE-L-KS fails to bind UNC5B regardless of netrin-1 stimulation (Figure 2A), indicating that GTPase activity is essential for the association. Coimmunoprecipitation demonstrates that PIKE-L does not bind to death domain containing p75NTR or UNC5A, but it faintly associates with UNC5C irrespective of netrin-1 treatment (data not shown), suggesting that PIKE-L specifically interacts with UNC5B in response to netrin-1.

Figure 2
Netrin-1 mediates the interaction between PIKE-L and UNC5B through tyrosine phosphorylation

PIKE can be phosphorylated on tyrosine residues by Fyn 14. Netrin-1 has also been shown to trigger its receptor DCC tyrosine phosphorylation 1518. To test whether phosphorylation on tyrosine modulates the interaction between PIKE-L and UNC5B, we pretreated the cotransfected HEK293 cells with several tyrosine kinase inhibitors, followed by netrin-1. Netrin-1-mediated PIKE-L/UNC5B interaction is selectively disrupted by Genistein, PP2 and compound 5. PP2 is a Fyn selective inhibitor, whereas Genistein and compound 5 are tyrosine kinase inhibitors. By contrast, Daidzein and PP3, inactive controls for Genistein and PP2, respectively, fail to dissociate the complex (Figure 2B, top panel). Accordingly, netrin-1 triggers tyrosine phosphorylation on both PIKE-L and UNC5B, which are selectively blocked by Genistein, PP2 and Compound 5 (Figure 2B, 2nd panel). In primary cortical cultures, netrin-1 elicits endogenous PIKE-L to bind UNC5B, which is blocked by PP2 pretreatment, and Genistein also decreases the interaction (Figure 2C, top panel). PIKE-L tyrosine phosphorylation is substantially abolished by PP2 or Genistein pretreatment. UNC5B tyrosine phosphorylation is evidently blocked by PP2 and partially prevented by Genistein (Figure 2C, 2nd and 5th panels).

To explore whether Fyn contributes to phosphorylate UNC5B or PIKE-L, we cotranfected UNC5B into HEK293 cells with various tyrosine kinases. Coimmunoprecipitation shows that constitutively active FynA but not kinase-dead FynD or other tyrosine kinases including Src and Pyk2 interacts with UNC5B (Supplemental Figure 1A, left panels). Moreover, FynA also elicits robust tyrosine phosphorylation on UNC5B. The 100 kDa band below UNC5B might be an unknown tyrosine phosphorylated protein associated with UNC5B (Supplemental Figure 1A, top middle panel). We conducted the similar experiments with PIKE-L, and found that FynA but not FynD or other tyrosine kinases selectively bound and phosphorylated PIKE-L (Supplemental Figure 1B). Coimmunoprecipitation demonstrates that PIKE-L tightly binds UNC5B in FynA but not FynD cotransfected cells even in the absence of netrin-1, no matter PIKE-L or UNC5B is immunoprecipitated (Supplemental Figure 1C, top left panels). Both UNC5B and PIKE-L are potently phosphorylated in FynA but not FynD transfected cells (Supplemental Figure 1C, middle panels). Noticeably, compared to FynD cells, PIKE-L apoptotic cleavage is partially decreased in FynA cells (Supplemental Figure 1C, right lower panel), an effect similar to netrin-1-treated cells. PIKE-A is mainly phosphorylated by Fyn on Y681 and Y774 residues 14, which correspond to Y1032 and Y1124 in PIKE-L. Immunoprecipitation assay reveals that wild-type PIKE-L strongly interacts with UNC5B, whereas Y1032F, Y1124F and Y1032, 1124F mutants fail to interact with UNC5B. As expected, PIKE-L wild-type is evidently tyrosine phosphorylated, which is abolished in PIKE-L mutants (Figure 3A, top and 2nd panels). Reciprocal immunoprecipitation reveals similar results (Figure 3A, 4th and 5th panels), suggesting that PIKE-L tyrosine phosphorylation is necessary for its association with UNC5B. To explore further whether Fyn regulates the association between PIKE-L and UNC5B, we conducted immunoprecipitation with wild-type and Fyn −/− mice. UNC5B robustly interacts with PIKE-L in +/+ but not −/− mouse brain (Figure 3B), indicating that Fyn is required for the association. PY99 antibody but not control IgG selectively pulls down UNC5B and PIKE-L from +/+ but not −/− brain lysates, suggesting that UNC5B and PIKE-L are strongly tyrosine phosphorylated in wild-type but not Fyn-null mice (Figure 3B, left panels). To explore whether netrin-1 triggers Fyn activation through UNC5B, we tested this notion in HEK293 cells, which were transfected with PIKE-L or UNC5B alone or in a combination. Netrin-1 elicits a UNC5B-dependent Fyn activation. Cotransfection of both PIKE-L and UNC5B enhances Fyn activation upon netrin-1 treatment (Figure 3C, top left panel). Depletion of UNC5B in cortical neurons abolishes netrin-1-provoked Fyn activation (Figure 3C, bottom right panel), underscoring that UNC5B is critical for Fyn activation. Collectively, Fyn plays a critical role in regulating UNC5B/PIKE-L complex formation in mouse brain.

Figure 3
Interaction of PIKE-L and UNC5B is regulated by Fyn tyrosine kinase

Netrin-1 activates PI 3-kinase, which is essential for netrin-1’s role in axon guidance 1921. To assess whether UNC5B/PIKE-L complex plays any role in mediating netrin-1-provoked PI 3-kinase activation, we transfected HEK293 cells with PIKE-L and UNC5B. In the absence of UNC5B, netrin-1 fails to provoke PI 3-kinase activity in PIKE-L transfected cells while netrin-1 elicits robust PI 3-kinase activity in wild-type PIKE-L and UNC5B cotransfected cells. This netrin-1-dependent PI 3-kinase activation is decreased in Wortmannin or LY294002 pretreated cells or when a dominant-negative PIKE-L mutant is transfected (Figure 4A, bottom panel). These data suggest that netrin-1-triggered PI 3-kinase activation is mediated by PIKE-L. To explore whether PIKE-L is necessary for this action, we infected cortical neurons with various adenovirus expressing sh-RNA of PIKE, wild-type PIKE-L or PIKE-L-KS and monitored Akt phosphorylation. Netrin-1-provoked PI 3-kinase activity is substantially reduced when PIKE-L is depleted. Infection of PIKE-L increases netrin-1’s stimulatory effect on PI 3-kinase, which is massively attenuated in PIKE-L–KS infected cells (Figure 4B, 2nd panel).

Figure 4
PIKE-L and Fyn are necessary for netrin-1-triggered PI 3-kinase activation

To evaluate whether Fyn mediates netrin-induced PI 3-kinase activity, we pretreated primary neurons with PP2, and stimulated cells with netrin-1 for 30 min. Netrin-1 induces potent Akt activation, which is completely blocked by PP2 (Figure 4C, top panel), suggesting that Fyn is essential for netrin-1 to elevate PI 3-kinase activity. Compared to wild-type neurons, netrin-1-triggered Akt activation is only slightly decreased in DCC −/− neurons, indicating that DCC is not the major receptor for this effect (Figure 4C, 3rd panel). Moreover, activation of Src family tyrosine kinases is partially diminished in DCC-null neurons (Figure 4C, bottom panel), suggesting that DCC contribute to Src family tyrosine kinases activation by netrin-1. To explore the physiological role of Fyn in regulating PI 3-kinase activation by netrin-1, we monitored Akt activation in wild-type and Fyn knockout neurons. Netrin-1 strongly elicits PI 3-kinase activation in wild-type neurons and PIKE-L further enhances its effect. Strikingly, PIKE-L (Y1032, 1124F) fails to increase further Akt activation by netrin-1 (Figure 4D, top panel), indicating that PIKE-L binding to UNC5B is essential for PI 3-kinase activation by netrin-1. As expected, netrin-1 fails to activate Akt in Fyn −/− neurons, regardless of PIKE-L expression (Figure 4D, 3rd panel). PI 3-kinase activation status tightly couples to Akt phosphorylation (data not shown). Depletion of UNC5B in cortical neurons evidently attenuates Akt activation. Nonetheless, PIKE-L expression level remains constant irrespective of UNC5B alteration (Figure 4E). Taken together, this finding supports that UNC5B but not DCC plays an essential role for mediating netrin-1-provoked PIKE-L dependent PI 3-kinase/Akt signaling activation.

To determine whether PIKE-L modulates netrin-1’s anti-apoptotic action, we infected primary neuronal cultures with various adenoviruses and provoked apoptosis with glutamate, a broadly used excitatory amino acid (EAA) neurotransmitter for neuronal cell death. In control neurons, netrin-1 evidently blocks glutamate-triggered apoptosis. Infection of wild-type PIKE-L strongly suppresses apoptosis, and netrin-1 further enhances the protective effect. By contrast, the anti-apoptotic effect by netrin-1 is reduced in PIKE-L-KS-infected neurons. Moreover, knocking down of PIKE-L almost completely abolishes netrin-1’s survival activity. The shRNA resistant PIKE-L mutant infection restores the survival effect, suggestive of the specificity of the siRNA effect. Fyn phosphorylation deficient mutant PIKE-L(Y1032, 1124F) also attenuates apoptosis, but it is irresponsive to netrin-1 (Figure 5A). Thus, PIKE-L binding to UNC5B is necessary for netrin-1’s anti-apoptotic effect. The anti-apoptotic activity by PIKE-L couples to its stimulatory effects on PI 3-kinase in netrin-1-treated cells. In Fyn −/− neurons, netrin-1’s anti-apoptotic effect is substantially impaired. PIKE-L(Y1032, 1124F) modestly represses apoptosis in a netrin-independent manner. Wild-type PIKE-L decreases apoptosis, but netrin-1 fails to enhance its survival activity (Figure 5B, upper panel), supporting that Fyn is required for PIKE-L to mediate netrin-1’s anti-apoptotic effect. Interestingly, depletion of PIKE-L slightly decreases UNC5B but not DCC expression level. Overexpression of PIKE-L enhances UNC5B and DCC expression, indicating that PIKE-L stabilizes netrin-1 receptors in neurons (Figure 5B, lower panels). In the absence of netrin-1, depletion of PIKE-L enhances UNC5B cleavage and PARP fragmentation in wild-type neurons elicited by glutamate, which are attenuated by PIKE-L overexpression. Netrin-1 further elevates PIKE-L’s anti-apoptotic effect. In the absence of PIKE-L, netrin-1 fails to block PARP and UNC5B apoptotic cleavage (Figure 5C, left panel). In contrast, in Fyn −/− neurons, netrin-1 loses its anti-apoptotic effect as PARP and UNC5B are robustly cleaved no matter netrin-1 is present or not. Nevertheless, overexpression of PIKE-L slightly decreases UNC5B cleavage (Figure 5C, right panels).

Figure 5
PIKE-L and Fyn are essential for the neuronal survival effect by netrin-1

Excitotoxic neuronal cell death in brain is often induced experimentally by the administration of kainic acid (KA), a potent agonist of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate class of glutamate receptors 22. Depletion of PIKE-L in hippocampus by injecting adenovirus expressing its shRNA into mouse brain enhances UNC5B cleavage triggered by KA, whereas overexpression of PIKE-L completely suppresses UNC5B degradation. Accordingly, PARP cleavage, a marker for apoptosis, tightly correlates with PIKE-L expression levels (Figure 5D). Since UNC5B knockout mice are embryonic lethal 23, we can not use the UNC5B deficient mice to explore its role in neuronal apoptosis. Alternatively, we deplete UNC5B in mouse brain utilizing adenovirus expressing its shRNA. Compared to control, elimination of UNC5B in hippocampus evidently blocks KA-provoked apoptosis in mouse brain (Figure 5E), underscoring that UNC5B implicates in neuronal cell death in vivo. Collectively, our data show that netrin-1 triggers PIKE-L/UNC5B complex formation, which is regulated by Fyn tyrosine kinase, and represses neuronal apoptosis.

Our finding that PI 3-kinase is coupled to UNC5B through PIKE-L in response to netrin-1, suppressing neuronal apoptosis, provides a molecular mechanism that explains netrin-1’s stimulatory effect on PI 3-kinase. Conceivably, such active effect by netrin-1 on PI 3-kinase is associated with the activation by PIKE-L of PI 3-kinase. PIKE activates PI 3-kinase in a GTP-dependent manner 11. PLC-γ1 acts as a physiological guanine nucleotide exchange factor (GEF) for PIKE12. Recently, it has been shown that netrin-1 induces activation of PLC-γ1, which is mediated by DCC but not UNC5B or neogenin 24. Presumably, PLC-γ1 is activated by netrin-1 and functions as a GEF for PIKE-L. The in vitro binding study reveals that the cytoplasmic domain of DCC associates with PIKE-L and coimmunoprecipitation also indicates that DCC interacts with PIKE-L in brain lysates, for which Fyn might not be essential (Supplemental Figure 2A–C). These data provide a molecular mechanism explaining PIKE-L activation in response to extracellular stimuli such as netrin-1. Netrin-1 provokes the association between DCC and UNC5B 25. Strikingly, we found that PIKE-L or Fyn triggers DCC/UNC5B complex formation even in the absence of netrin-1. Netrin-1 further elevates the binding between DCC and UNC5B in the presence of PIKE-L or Fyn (Supplemental Figure 2D). Conceivably, Fyn and PIKE-L play critical roles in mediating netrin-1-triggered DCC/UNC5B heterodimer formation in neurons. UNC5B mediates apoptosis through DAP-kinase (DAPK) 7. However, neither PIKE-L nor Fyn affects the association between UNC5B and DAPK (data not shown). In addition, PIKE-L does not regulate UNC5B-dependent DAPK activation either (Supplemental Figure 2E).

At E12 in mutant mice deficient for netrin-1 expression, apoptosis is strongly enhanced in the whole brainstem and the ventricular zone 5. However, Hinck et al., reported Ntn1 is not a necessary survival factor for UNC5A-expressing neurons in vivo, as the hypomorphic netrin-1 mutant mice failed to show increased apoptosis in neural tube ventral neurons 26. To investigate netrin-1’s role in neuronal survival in vivo, we conducted TUNEL staining on E18 brain sections. Evident TUNEL-positive cells in Ntn1 −/− mice are found in ventrolacteral periaqueductal area, where both UNC5B and DCC are expressed (Supplemental Figure 3). Thus, our findings suggest that netrin-1 contributes to suppressing neuronal cell death in region close to the Aqueduct in mouse brain.

Previous studies reveal that PKC/PICK1 complex mediates UNC5A membrane surface removal upon PKC activation, which inhibits netrin-1 dependent collapse of hippocampal growth cones 27. Since PIKE-L does not bind UNC5A, we conducted immunofluorescent staining on hippocampal neurons to explore whether PIKE-L plays any role in UNC5B surface expression. UNC5B surface expression is not altered by PMA treatment; nevertheless, depletion of PIKE-L enhances its surface distribution, and PMA further escalates it. By contrast, overexpression of PIKE-L decreases UNC5B surface expression, PMA and netrin-1 combined treatment further attenuates its surface residency (Supplemental Figure 4A), supporting that PIKE-L might promote UNC5B internalization. Netrin-1 provokes UNC5B internalization, but PI 3-kinase inhibitors display negligible effect on UNC5B surface expression, indicating that PI 3-kinase signaling is not implicated in this event (Supplemental Figure 4B). Interestingly, knockout of Fyn elicits UNC5B surface insertion, suggesting that netrin-1-triggered Fyn but not PI 3-kinase signaling mediates UNC5B trafficking. On the other hand, depletion of PIKE-L abolishes netrin-1-triggered actin filaments in cortical neurons (Supplemental Figure 5). Together, these data support that PIKE-L might not only be involved in the survival activity of netrin-1 but also in its putative role in neuronal cell adhesion and migration. Collectively, in this report, we show that Fyn is sufficient and necessary for triggering PIKE-L and UNC5B tyrosine phosphorylation and association upon netrin-1 treatment. Our finding that PIKE-L selectively associates with UNC5B but not other family member of UNC5 or p75NTR provides strong evidence indicating that PIKE-L/UNC5B complex is responsible to activate PI 3-kinase cascade and suppress apoptosis upon netrin-1 stimulation.


Cells and Reagents

HEK293T cells were maintained in DMEM, supplemented with 10% fetal bovine serum (FBS), 2 mg/ml glutamine and 100 units penicillin-streptomycin at 37°C with 5% CO2 atmosphere in a humidified incubator. Stable HEK293 cells expressing myc tagged chicken netrin-1 were provided by Dr. Yi Rao (Northwestern University). The purified anti-PIKE-N antibody is against the N-terminus from residues 1 to 384 in PIKE-L. Anti-HA-HRP, anti-DAP-Kinase antibody was from Sigma. Anti-PY99, anti-p110, anti-DCC, anti-GFP, anti-Akt antibodies were from Santa Cruz Biotech. Anti-p-Akt473 antibody was from Cell Signaling. Recombinant chicken netrin-1 protein and anti-UNC5B antibody were from R&D system. Anti-UNC5B (clone 1A9, antigen 27 aa-127 a.a.) was from Novus Bio. Glutathione Sepharose 4B was supplied by Pharmacia Biotech. All the tyrosine kinase inhibitors and Anti-myc antibody were from Calbiochem. The primer sequence for preparing siRNA resistant PIKE-L: GCA CTT AT(C)T TG(C)T ATC GAA TG. The C nucleotides in the parenthesis were changed into T. The siRNA sequence for UNC5B (sense) is: 5’ AGA CTG GAT TCC AGC TCA A 3’. Protein A/G-conjugated agarose beads were from Sigma.

Yeast Two-hybrid Screen

The yeast two-hybrid screen was performed using the MATCH-MAKER Two-hybrid System 2 (Clontech) according to the manufacturer’s protocol. The GTPase domain of PIKE was subcloned downstream of the Gal4 DNA-binding domain in pAS2-1 and was used as bait to screen a human Fetal Brain cDNA Library in pACT2. Clones that grew on plates lacking leucine, tryptophan and histidine with 50 mM 3-aminotriazole were selected and assayed for β-galactosidase activity 28.

Protein-protein interaction assays

Ten-cm dishes of HEK293T cells were transfected with 10 µg DNA by the calcium phosphate precipitation method. In 48 h, the transfected cells were treated as indicated, collected and washed once in PBS, lysed in 1 ml lysis buffer (50 mM Tris, pH 7.4, 40 mM NaCl, 1 mM EDTA, 0.5% Triton X-100, 1.5 mM Na3VO4, 50 mM NaF, 10 mM sodium pyrophosphate, 10 mM sodium β-glycerophosphate, 1 mM phenylmethylsulfonyl fluoride (PMSF), 5 mg/ml aprotinin, 1 mg/ml leupeptin, 1 mg/ml pepstatin A), and was centrifuged for 10 min at 14,000 × g at 4°C. Experimental procedures for co-immunoprecipitation were carried out as described 29. Immune complexes were resolved by SDS-PAGE and were subjected to immunoblotting. GST pull down assay was carried out as described previously 14.

PI 3-kinase assay

The cell lysate was prepared as described above. PI3-kinase was immunoprecipitated with rabbit anti-p110 from the cell extract, and washed with the following buffers: 3 times with buffer A (PBS, 1% NP-40, 1 mM dithiothreitol (DTT)); 2 times with buffer B (PBS, 0.5 M LiCl, 1 mM DTT); and 2 times with buffer C (10 mM Tris-HCl, pH 7.4, 0.1 M NaCl, 1 mM DTT). Subsequent steps were performed as described 12.

Primary rat cortical neuron culture and apoptotic assay

Primary cultured mouse cortical neurons were prepared as follows. P1 mouse pups were decapitated and cortex was extirpated, cross chopped and suspended by pipetting for separation in 5% fetal calf serum (FCS), 5% horse serum (HS) Dulbecco’s modified Eagle’s medium (DMEM) gently. The cell suspension was then centrifuged at 250 × g for 5 min. This operation was repeated again. Cells were seeded into polyethyleneimine-coated 10 cm dishes and 12-well plates including coated-coverslips and incubated at 37 °C in 5% CO2/95% air. After 3 h, culture medium was changed to Neurobasal containing B-27 supplement (Invitrogen) and incubated for 4 days. For maintenance, a half medium is change to fresh Neurobasal/B27 in every 4 days. After 1 week, the dished cultured neurons are employed in various experiments. After infection with various adenovirus expressing PIKE-L, PIKE-L-KS, PIKE-L (Y1032, 1124F) adenovirus and control virus, wild-type cortical neurons were treated with 200 ng/ml netrin for 30 min before 80 µM glutamate was introduced. In 16–18 h, propidium iodide and Hoechst 33342 were added. The stained neurons were fixed and studied under fluorescent microscope. The neuronal apoptosis was also verified by an independent approach: caspase-3 activated fluorescent dye MR(DEVD)2. Due to the sensitivity, 80 µM glutamate was added into Fyn −/− cortical neurons to trigger apoptosis for 12–14 h.

Adenovirus injection into mouse hippocampus

We determined the viral titer by measuring the number of infected HEK293 cells expressing GFP, and adjusted all viral stocks to 1 × 108 plaque-forming units/ml before use. For injection of adenoviral vectors into the hypothalamus, we anesthetized male C57BL/6J mice aged 8–12 weeks (n = 6 per group) with 2.5% avertin. We performed standard surgical procedures to inject adenoviral vectors expressing either PIKE-L or UNC5B-specific siRNA with GFP or GFP alone using a stereotaxic table (David Kopf Instruments). The experiment was conducted as previously described 30. In 4 days, we intraperitoneally administrated 25 mg/kg kainic acid. In 5 days, the hippocampal regions were dissected and homogenated and analyzed by immunoblotting.

Statistical analysis

The results were expressed as means ± S.D. calculated from the specified numbers of determination. A Student’s t-test was used to compare individual data with control value.

Supplementary Material

Supplemental Figures:

Figure 1. Fyn phosphorylates both PIKE-L and UNC5B and regulates the interaction. (A) Active Fyn tyrosine kinase binds UNC5B and phosphorylates UNC5B. UNC5B was cotransfected with various tyrosine kinases into HEK293T cells. The transfected tyrosine kinases were immunoprecipitated and analyzed with anti-HA-HRP and PY99 antibodies, respectively. Only active fyn selectively bound UNC5B and phosphorylated it (top left and middle panels). Verification of transfected constructs (left lower panels). The associated active FynA was also tyrosine phosphorylated (middle 2nd panel). Confirmation of transfected UNC5B, fyn and Src A (right panel). (B) Active fyn tyrosine kinase binds and phosphorylates PIKE-L. PIKE-L was cotransfected with various tyrosine kinases into HEK293T cells. Only active fyn specifically bound PIKE-L and phosphorylated it (left and middle panels). Reciprocal immunoprecipitation verified the similar results (right panels). (C) Active Fyn triggers the interaction between PIKE-L and UNC5B. HEK293T cells were co-transfected with myc-PIKE-L, UNC5B-HA in the presence of fynA or fynD. The expression of PIKE-L, UNC5B and fyn were determined by western blotting. PIKE-L strongly bound to UNC5B in the presence of fynA but not fynD (left top panels). FynA but not FynD provoked tyrosine phosphorylation in PIKE-L and UNC5B (middle top panels). Compared to fynD sample, PIKE-L apoptotic cleavage was decreased when cotransfected with fynA (right lower panel).

Figure 2. PIKE-L interacts with DCC and mediates its association with UNC5B (A) In vitro binding between PIKE-L and DCC. Purified GST control, GST-DCC-CTD, GST-DCC-CTD-ΔP3 recombinant proteins were respectively incubated with cell lysates of HEK293, transfected with myc-PIKE-L. PIKE-L selectively bound to both DCC-CTD and ΔP3 but not GST control. (B) PIKE-L binds DCC in HEK293T cells. Flag-DCC was co-transfected with PIKE-L and its fragments into HEK293T cells. PIKE-L full-length showed the strongest binding with DCC (top panel). Verification of transfected constructs (middle and bottom panels). (C) Interaction of PIKE-L and DCC is not regulated by fyn. Brain homogenates from wild-type and fyn−/− mice were immunoprecipitated by anti-DCC and control IgG, respectively, followed by immunoblotting with anti-PIKE antibody. PIKE-L bound to DCC regardless of Fyn expression. (D) PIKE-L and Fyn mediate UNC5B and DCC association. UNC5B and DCC were cotransfected with either PIKE-L or Fyn into HEK293 cells, followed by netrin-1 treatment. UNC5B was immunoprecipitated with anti-HA antibody, its associated proteins were analyzed with anti-myc antibody. In the absence of netrin-1, DCC did not bind to UNC5B; however, in the presence of PIKE-L or Fyn, DCC tightly associated with UNC5B even in the absence of netrin-1. Netrin-1 treatment further enhanced the interaction between DCC and UNC5B (top panel). Confirmation of transfected constructs (2nd to bottom panels). (E) PIKE-L does not regulate DAP-kinase autophosphorylation. HEK293T cells were transiently transfected with a control vector or Myc-tagged PIKE-L vector together with HA- tagged-UNC5B and Flag-tagged DAP-kinase expressing constructs. Recombinant netrin-1 was added 48 h after transfection. Pull-down with a phospho-serine308-DAPK antibody followed by western blot using anti-Flag was performed. Overexpression of PIKE-L did not mediate DAPK autophosphorylation, whereas netrin-1 treatment slightly increased DAPK phosphorylation irrespective of PIKE-L expression (top panels). Verification of transfected DAPK, UNC5B and PIKE-L constructs (2nd, 3rd and bottom panels).

Figure 3. Netrin-1 deficiency triggers apoptosis in cerebral aqueduct around ventricular zone (A) TUNEL staining on wild-type and netrin-1 knockout embryonic brain sections (E18). Evident TUNEL staining was found around the cerebral aqueduct region with most concentrated in the ventrolacteral periaqueductal gray area. No apoptosis was detected in other regions. Eighteen embryos were analyzed in this experiment. (B) Immunohistochemistry staining of UNC5B and DCC. DCC and UNC5B were expressed in the region close to the Aqueduct (white arrows) in both wild-type and Netrin −/− mouse brain. Bar = 50 µm

Figure 4. Depletion of PIKE-L and Fyn enhances UNC5B surface expression in hippocampal neurons (A) Depletion of PIKE-L upregulates UNC5B surface expression. 14 DIV hippocampal neurons were infected with control adenovirus or adenovirus expressing wild-type PIKE-L or its shRNA. The infected neurons were treated with netrin-1 or PMA. Cell were live-labeled to immunostain surface UNC5B or fixed and permeabilized to stain the internalized UNC5B. The ratios between surface and internalized were calculated. Removal of PIKE-L increased UNC5B surface expression, which was enhanced by PMA treatment. Nevertheless, overexpression of PIKE-L diminished UNC5B surface expression. (*P<0.05; student t-test). (B) Knockout Fyn enhances UNC5B surface expression. Hippocampal neurons were cultured from wild-type and Fyn −/− mice. The wild-type neurons were pretreated with 100 nM wortmannin or 10 µM LY294002 for 30 min, respectively. The drug-treated neurons were stained as described. The ratios between surface and internalized were calculated. Depletion of Fyn increased UNC5B surface expression. Netrin-1 treatment decreased surface expression. However, PI 3-kinase inhibitors have no effect on UNC5B surface expression. (*P<0.05; student t-test). (C) Immnofluorescent staining of UNC5B. Surface or intracellular UNC5B was stained in hippocampal neurons from Fyn knockout hippocampal neurons. Neurons were live-labeled to immunostain surface UNC5B (right), then fixed and permeabilized to immunostain intracellular UNC5B (left).

Figure 5. PIKE-L contributes to netrin-1-induced axonal F-actin polymerization (A) Immunofluorescent staining of F-actin and PIKE-L on cortical neurons. Rat E17 cortical neurons were infected with control or sh-RNA-PIKE-L virus for 6 hours, then changed medium and cultured for additional 48 hours. Prior to fixation and staining, netrin-1/control medium was treated for 1 hour. The fixed neurons were stained with anti-PIKE-L (red) + FITC-conjugated phalloidin (green). Netrin-1-induced filamentous actin (F-actin polymerization), particularly at neurite growth cones, was abolished by shRNA-PIKE-L. The circled regions were amplified at the bottom of the figure. (B) Quantification of actin polymerization on control and sh-RNA PIKE-L-infected neurons.


This work is supported by grant from National Institute of Health (RO1, NS045627) to K. Ye and by grants from ANR and ligue contre le Cancer to P. Mehlen.


1. Keino-Masu K, et al. Deleted in Colorectal Cancer (DCC) encodes a netrin receptor. Cell. 1996;87:175–185. [PubMed]
2. Serafini T, et al. Netrin-1 is required for commissural axon guidance in the developing vertebrate nervous system. Cell. 1996;87:1001–1014. [PubMed]
3. Mehlen P, et al. The DCC gene product induces apoptosis by a mechanism requiring receptor proteolysis. Nature. 1998;395:801–804. [PubMed]
4. Thiebault K, et al. The netrin-1 receptors UNC5H are putative tumor suppressors controlling cell death commitment. Proc Natl Acad Sci U S A. 2003;100:4173–4178. [PMC free article] [PubMed]
5. Llambi F, Causeret F, Bloch-Gallego E, Mehlen P. Netrin-1 acts as a survival factor via its receptors UNC5H and DCC. Embo J. 2001;20:2715–2722. [PMC free article] [PubMed]
6. Tanikawa C, Matsuda K, Fukuda S, Nakamura Y, Arakawa H. p53RDL1 regulates p53-dependent apoptosis. Nat Cell Biol. 2003;5:216–223. [PubMed]
7. Llambi F, et al. The dependence receptor UNC5H2 mediates apoptosis through DAP-kinase. Embo J. 2005;24:1192–1201. [PMC free article] [PubMed]
8. Ackerman SL, et al. The mouse rostral cerebellar malformation gene encodes an UNC-5-like protein. Nature. 1997;386:838–842. [PubMed]
9. Leonardo ED, et al. Vertebrate homologues of C. elegans UNC-5 are candidate netrin receptors. Nature. 1997;386:833–838. [PubMed]
10. Williams ME, Strickland P, Watanabe K, Hinck L. UNC5H1 induces apoptosis via its juxtamembrane region through an interaction with NRAGE. J Biol Chem. 2003;278:17483–17490. [PubMed]
11. Ye K, et al. Pike. A nuclear gtpase that enhances PI3kinase activity and is regulated by protein 4.1N. Cell. 2000;103:919–930. [PubMed]
12. Ye K, et al. Phospholipase C gamma 1 is a physiological guanine nucleotide exchange factor for the nuclear GTPase PIKE. Nature. 2002;415:541–544. [PubMed]
13. Rong R, et al. PI3 kinase enhancer-Homer complex couples mGluRI to PI3 kinase, preventing neuronal apoptosis. Nat Neurosci. 2003;6:1153–1161. [PubMed]
14. Tang X, Feng Y, Ye K. Src-family tyrosine kinase fyn phosphorylates phosphatidylinositol 3-kinase enhancer-activating Akt, preventing its apoptotic cleavage and promoting cell survival. Cell Death Differ. 2007;14:368–377. [PubMed]
15. Li W, et al. Activation of FAK and Src are receptor-proximal events required for netrin signaling. Nat Neurosci. 2004;7:1213–1221. [PMC free article] [PubMed]
16. Liu G, et al. Netrin requires focal adhesion kinase and Src family kinases for axon outgrowth and attraction. Nat Neurosci. 2004;7:1222–1232. [PMC free article] [PubMed]
17. Ren XR, et al. Focal adhesion kinase in netrin-1 signaling. Nat Neurosci. 2004;7:1204–1212. [PubMed]
18. Meriane M, et al. Phosphorylation of DCC by Fyn mediates Netrin-1 signaling in growth cone guidance. J Cell Biol. 2004;167:687–698. [PMC free article] [PubMed]
19. Chang C, et al. MIG-10/lamellipodin and AGE-1/PI3K promote axon guidance and outgrowth in response to slit and netrin. Curr Biol. 2006;16:854–862. [PubMed]
20. Campbell DS, Holt CE. Chemotropic responses of retinal growth cones mediated by rapid local protein synthesis and degradation. Neuron. 2001;32:1013–1026. [PubMed]
21. Ming G, et al. Phospholipase C-gamma and phosphoinositide 3-kinase mediate cytoplasmic signaling in nerve growth cone guidance. Neuron. 1999;23:139–148. [PubMed]
22. Schauwecker PE, Steward O. Genetic determinants of susceptibility to excitotoxic cell death: implications for gene targeting approaches. Proc Natl Acad Sci U S A. 1997;94:4103–4108. [PMC free article] [PubMed]
23. Lu X, et al. The netrin receptor UNC5B mediates guidance events controlling morphogenesis of the vascular system. Nature. 2004;432:179–186. [PubMed]
24. Xie Y, et al. DCC-dependent phospholipase C signaling in netrin-1-induced neurite elongation. J Biol Chem. 2006;281:2605–2611. [PubMed]
25. Hong K, et al. A ligand-gated association between cytoplasmic domains of UNC5 and DCC family receptors converts netrin-induced growth cone attraction to repulsion. Cell. 1999;97:927–941. [PubMed]
26. Williams ME, et al. UNC5A promotes neuronal apoptosis during spinal cord development independent of netrin-1. Nat Neurosci. 2006;9:996–998. [PubMed]
27. Williams ME, Wu SC, McKenna WL, Hinck L. Surface expression of the netrin receptor UNC5H1 is regulated through a protein kinase C-interacting protein/protein kinase-dependent mechanism. J Neurosci. 2003;23:11279–11288. [PubMed]
28. Tang X, et al. Sperm membrane protein (hSMP-1) and RanBPM complex in the microtubule-organizing centre. J Mol Med. 2004;82:383–388. [PubMed]
29. Tang X, Ye K. Pike tyrosine phosphorylation regulates its apoptotic cleavage during programmed cell death. Adv Enzyme Regul. 2006;46:289–300. [PubMed]
30. Sheng G, et al. Hypothalamic huntingtin-associated protein 1 as a mediator of feeding behavior. Nat Med. 2006;12:526–533. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • Conserved Domains
    Conserved Domains
    Link to related CDD entry
  • Gene
    Gene links
  • Gene (nucleotide)
    Gene (nucleotide)
    Records in Gene identified from shared sequence links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene links
  • MedGen
    Related information in MedGen
  • Nucleotide
    Published Nucleotide sequences
  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • Protein
    Published protein sequences
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...