| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Biological Science, and 2 Department of Chemistry, Tokyo Metropolitan University, Minami-osawa, Hachioji, Tokyo 192-0397, Japan
3 Department of Anatomy, Keio University School of Medicine, Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan
4 Department of Molecular Neurobiology, Institute of DNA Medicine, Jikei University School of Medicine, Nishi-Shinbashi, Minato-ku, Tokyo 105-8461, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionExperimental proceduresReferences |
|---|
/CPβ, which are subunits of barbed end F-actin-capping proteins (CP), as proteins bound to unphosphorylated Dab1 by mass spectrometric analysis. It was shown that the PTPAPR sequence of Dab1, conforming to the PxxxPR atypical SH3-binding motif, was the binding site for SH3 domains of CIN85. The results that phosphorylation at Ser491 close to the PTPAPR sequence inhibited association with CIN85 may provide a mechanism regulating the interaction between the PxxxPR motif proteins and SH3 domains of CIN85 family proteins. Together with previous results that CIN85 regulates actin assembly, present results raise the possibility that Cdk5 modulates actin dynamics through regulation of CIN85–Dab1 interaction by the Dab1 phosphorylation. |
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Cdk5, a member of cyclin-dependent kinases (Cdks), requires association with the neuron-specific activator p35 or p39 for kinase activity (Dhavan & Tsai 2001; Hisanaga & Saito 2003). Mice deficient in Cdk5 or p35 display inverted neuronal cell layer formation similar but not identical to mice bearing mutations for Reelin pathway proteins (Ohshima et al. 1996; Chae et al. 1997). Although genetic approaches have indicated that Cdk5–p35 and Reelin pathway operate in parallel (Ohshima et al. 2001; Beffert et al. 2004), several studies indicate a direct interaction of Cdk5 with Dab1. Cdk5 can phosphorylate Dab1 at Ser491 in vivo (Keshvara et al. 2002) and Cdk5-phosphorylation of Dab1 inhibits the Tyr phosphorylation of Dab1 upon Reelin stimulation (Ohshima et al. 2007). However, the role of Dab1 phosphorylation by Cdk5 remains incompletely explained.
CIN85, also known as Ruk/SETA, and its homologue CD2AP, known as CMS, constitute a family of multidomain adaptor proteins composed of three SH3 domains at the N-terminus, a central proline-rich region, and a coiled-coil domain at the C-terminus (Dikic 2002). CIN85 and CD2AP associate with various proteins, including those involved in signaling, endocytosis and actin dynamics, through their SH3- and proline-rich regions. The critical binding motif recognized by SH3 domains of CIN85 is the atypical proline–arginine sequences composed of PxxxPR or Px(P/A)xxR (Kowanetz et al. 2003a; Kurakin et al. 2003). For example, the SH3 domains of CIN85 and CD2AP bind to a motif within Cbl ubiquitin ligase (implicated in receptor-mediated endocytosis and receptor down-regulation), CD2 and BLNK (involved in T and B cell receptor signaling), and AIP1/Alix (involved in glial cell apoptosis and cell adhesion) (Dikic 2002). The proline-rich region of CIN85 interacts with other types of SH3-containing proteins such as p130Cas and cortactin, which function in cytoskeletal organization (Lynch et al. 2003; Tibaldi & Reinherz 2003). CIN85 and CD2AP bind to CP, the barbed end-capping protein of actin filaments, between proline-rich and coiled-coil regions at their C-termini (Hutchings et al. 2003). However, there are few studies of these adaptor proteins in neurons.
To understand the relationship between Cdk5 and Dab1, we investigated Cdk5-dependent phosphorylation of this adaptor protein, found Ser400 as an in vivo phosphorylation site, and confirmed Ser491 as another major phosphorylation site. We searched for proteins that interact with Dab1 in a phosphorylation-dependent manner. CIN85, CD2AP, CP and CPβ were identified as proteins that specifically bind to non-phosphorylated Dab1. The binding site in Dab1 for CIN85 was the PTPAPR motif encompassed by amino acids 483–488. Dab1–CIN85 interaction was abolished by phosphorylation at Ser491. These findings raise the fascinating possibility that Cdk5 modulates signaling to the actin cytoskeleton through CIN85 and CP by phosphorylation of Dab1 at Ser491.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Cdk5-p35 phosphorylates multiple Ser or Thr sites of Dab1 (Dab1-p80) in vitro (Ohshima et al. 2007), but only Ser491 has been identified as an in vivo phosphorylation site for Cdk5 (Keshvara et al. 2002). To identify the additional phosphorylation sites for Cdk5, we used three Dab1 alternative splicing isoforms, p36, p45 and p80 (Fig. 1A) to detect region- and isoform-specific phosphorylation. These three isoforms of Dab1 have a common N-terminal PTB domain, but differ in their C-terminal tail domains. Dab1-p80, -p45 and -p36 have 13, 1 and 0 Ser/Thr–Pro sequences, respectively, in their C-terminal domains. Each of these isoforms was subjected to phosphorylation by Cdk5 in vitro. Dab1-p80 was strongly phosphorylated by both Cdk5–p35 (Fig. 1B) and Cdk5–p39 (Supplementary Fig. S1A). In contrast, Dab1-p45 and Dab1-p36 were inefficient substrates for either kinase. No signal was detected when dominant-negative Cdk5 (dnCdk5) was used for phosphorylation. These data indicated that Cdk5 phosphorylation sites were predominantly contained within the C-terminal tail of Dab1-p80, and that both Cdk5 complexes can phosphorylate Dab1 in vitro. Because Dab1-p80 (hereafter referred to as Dab1) is the major isoform expressed in brains (Howell et al. 1997), we concentrated on Dab1 phosphorylation by Cdk5–p35.
|
To observe Cdk5-dependent phosphorylation of Dab1 in living cells, Myc-Dab1 was transiently transfected with Cdk5–p35 in HEK293 cells. Dab1 immunoprecipitated from lysates of these cells, which had been metabolically labeled with 32P-phosphate, was radio-labeled (Fig. 1C). Dab1 was phosphorylated even when transfected alone without kinase (Fig. 1C, lane 2), probably due to phosphorylation by endogenous protein kinase(s) in HEK293, as demonstrated previously (Ohshima et al. 2007). When Dab1 was co-expressed with Cdk5–p35, phosphorylation increased dramatically and an upward shift in electrophoretic mobility was also detected (Fig. 1C, lane 3). Co-expression with dnCdk5–p35 suppressed the increased phosphorylation (Fig. 1C, lane 4). These data indicate Dab1 serves as a substrate for Cdk5 in cultured cells.
To determine the in vivo phosphorylation sites, phosphopeptide map analysis was performed with wild-type and a site-directed mutant form of Dab1 in which putative proline-directed phosphorylation sites were changed to Ala. Four major phosphopeptide spots were consistently detected with Dab1 phosphorylated in vitro by Cdk5–p35 (Fig. 1D, in vitro), as well as Cdk5-p39 (Supplementary Fig. S1B). Of these, three spots (1–3) corresponded with those detected in maps of phospho-Dab1 immunoprecipitated from transfected HEK293 cells (Fig. 1D, HEK) and cultured neurons (Fig. 1D, Neuron). Ser was determined as the phosphorylated residue in all three spots by phospho-amino acid analysis (data not shown). Therefore, various Dab1 mutants were generated with Ser residues that precede Pro (SP) (Fig. 1A, black arrowhead) replaced by Ala. Mutation of Ser400 to Ala (Dab1-S400A) resulted in the disappearance of spot 3 (Fig. 1E,F, S400A). Changing Ser491 to Ala (Dab1-S491A) resulted in loss of spot 2 and substantial reduction of spot 1 (Fig. 1E,F, S491A). Other mutations at Ser260, Ser481, Ser515 and Ser550 had no effect on the phosphopeptide map pattern of these four phosphopeptides (data not shown). Spots 1–3 vanished almost completely with the double Ala mutant of Ser400 and Ser491 (2A) (Figs 1E,F and 2A). These results indicate that Ser400 and Ser491 are the two major phosphorylation sites of Dab1 both in vitro and in cultured neurons. Dab1 was apparently phosphorylated at an additional site (spot 4 in Fig. 1E and arrow in Fig. 1F), and phosphorylation of this site may correspond to a shift upward in electrophoretic mobility (see Fig. 2A,B). Furthermore, phosphorylation of this site was shown by phospho-amino acid analysis to be phospho-Thr (data not shown).
|
To further study Cdk5-dependent phosphorylation of Dab1 in vivo, a phosphorylation state-specific antibody to phospho-Ser400 or phospho-Ser491 was generated. The anti-pS400 or anti-pS491 antibody selectively recognized Dab1 phosphorylated in vitro and cultured cells (Fig. 2A,B, and see also Fig. 5D). The phosphorylation of Dab1 in brains was studied with the anti-pS400 antibody because our anti-pS491 did not detect phosphorylated Dab1 in brain extracts. Phosphorylation at Ser400 was clearly detected when Dab1 immunoprecipitated from whole brain extracts of post-natal day 2 (P2) mice (Fig. 2C). When brains of yotari mice with a deletion mutation of Dab1 were used, no phospho-Ser400 was detected (Fig. 2D). Phosphorylation of Ser400 occurred predominantly at perinatal periods during brain development (Fig. 2E), as found with Ser491 (Keshvara et al. 2002). These results suggest that both phosphorylations occur synchronously in developing brains.
|
|
Identification of CIN85, CD2AP, CP and CPβ as Dab1-binding proteins in a Cdk5 phosphorylation-dependent manner
Because Dab1 is known to be an adaptor protein that transmits signals through protein–protein interactions, we hypothesized that phosphorylation of Dab1 by Cdk5 could regulate its interactions with other proteins, as is the case for Tyr phosphorylation (Ballif et al. 2004), and we searched comprehensively for such proteins. To screen the binding partners, the C-terminal fragment (CT) of Dab1 (amino acids 233–555) or its Ala mutant at both Ser400 and Ser491, was tagged with GST at the N-terminus and His at the C-terminus (GST-Dab1-CT or GST-Dab1-CT-2A) and used in an affinity matrix. GST-Dab1-CT or GST-Dab1-CT-2A coupled to glutathione–Sepharose beads was phosphorylated by Cdk5–p35. Completion of phosphorylation was monitored by immunoblotting of GST-Dab1-CT on beads with anti-pS400 and anti-pS491 (Fig. 4A). Rat brain was applied to columns of the respective Sepharose beads described above, and bound proteins were eluted sequentially with 1 M NaCl and 1 M MgCl2 buffers. Two distinct bands (bands 1 and 2) were specifically detected in the NaCl elution from the column of unphosphorylated GST-Dab1-CT and GST-Dab1-CT-2A, but not from the column of phosphorylated GST-Dab1-CT (Fig. 4B). One band (band 3) was also found in the MgCl2 elution of the unphosphorylated GST-Dab1-CT and GST-Dab1-CT-2A (Fig. 4C).
|
and CPβ, respectively. Band 3 contained two proteins, CIN85/Ruk/SETA and CD2AP/CMS. CIN85 and CD2AP are closely related adaptor proteins and will be referred to as CIN85/CD2AP in some cases hereafter. Binding of these proteins to unphosphorylated Dab1-CT was confirmed by immunoblotting. CP and CPβ antibodies specifically detected bands 1 and 2, respectively, in the NaCl eluate of unphosphorylated Dab1-CT (Fig. 4B, right panels). A CIN85 antibody specifically detected band 3. A CD2AP antibody detected a protein band that occurred at a slightly lower molecular weight region than band 3 (Fig. 4C, right panels). Whether CP or CIN85/CD2AP can bind to full length Dab1 was also examined by pulldown assay. Both CP and CIN85/CD2AP bound to full-length Dab1 (Dab1-p80) with similar efficiency to that observed for Dab1-CT binding (Supplementary Fig. S2). However, neither CP nor CIN85/CD2AP was found to bind Dab1-p36 or Dab1-p45. These results indicate that the C-terminal tail region, but not the N-terminal PTB domain or the phospho-Tyr region, is the only binding region in Dab1 for CP and CIN85/CD2AP.
The phosphoserine-dependent binding of Dab1 to CIN85
CP is known to bind to CIN85/CD2AP (Hutchings et al. 2003), raising the possibility that CIN85/CD2AP functions as an adaptor to link CP to Dab1. Therefore, we focused on the interaction of Dab1 with CIN85. First, we confirmed the binding in brains by immunoprecipitation of Dab1 from rat brain extracts. CIN85 was detected only in Dab1 immunoprecipitate but not in control immunoprecipitates (Fig. 5A). We used the transient transfection of Myc-tagged Dab1 and Flag-tagged CIN85 in HEK293 cells for further characterization of their binding. CIN85 was also co-immunoprecipitated with Dab1 in HEK293 cells (Fig. 5B). We next examined whether the binding in cultured cells was phosphorylation-dependent. First, the ability of double Ala (2A) or double Glu (2E) mutants of Dab1 to bind with CIN85 was assessed. Compared with wild-type Dab1 (Fig. 5C, WT) there was a marked increase in binding with the 2A mutant (Figs 2A and 5C), but binding was similar with mutant 2E (Figs 2E and 5C). These results strongly suggest that Dab1–CIN85 interactions are regulated by Cdk5-dependent phosphorylation of Dab1 and that this phosphorylation attenuates the association.
Dab1–CIN85 interactions were further investigated by immunofluorescent staining (Fig. 5D). Expression of CIN85 in COS-7 cells induced the formation of large vesicles, around the surface of which CIN85 appeared to localize, with accompanying diffuse cytoplasmic staining, as reported by Watanabe et al. (2000). Dab1 in all forms was distributed diffusely throughout the cytoplasm (data not shown). Co-expression of CIN85 did not change the overall distribution of Dab1 or Dab1-2E (Fig. 5D, a–c and g–i). On the other hand, Dab1-2A showed marked accumulation in large vesicles formed by CIN85 (Fig. 5D, d–f). We investigated the co-localization of Dab1 with CIN85 in cultured neurons by immunofluorescence. Unfortunately, however, the lack of the available antibodies for the immunostaining and/or the low expression levels of proteins did not allow us to reach a conclusive result.
We next asked whether the binding of Dab1 to CIN85 is affected by Cdk5 activity. Co-transfection of p35 in HEK293 cells increased phosphorylation of Dab1, as evidenced by immunoblotting with anti-pS400 and anti-pS491 as well as the electrophoretic mobility shift (Fig. 5E, lane 2). Concomitantly, the amount of CIN85 co-immunoprecipitated with Dab1 decreased considerably. Inhibition of Cdk5 with roscovitine attenuated the inhibitory effect induced by co-transfection of p35 (Fig. 5E, compare lanes 2 and 3 of top panel). On the other hand, co-immunoprecipitation of CIN85 with Dab1-2A was not affected by co-expression with p35 (Fig. 5E, lane 5). These results indicate that cellular Dab1–CIN85 interaction and co-localization are also dependent on phosphorylation by Cdk5–p35.
The PTPAPR sequence in Dab1 is the binding site for the SH3 domains of CIN85
To further characterize Dab1–CIN85 interactions, deletion mutant forms of CIN85 were derived. One form consisted of the N-terminal domain (NT), which included the three SH3 domains while another comprised only the C-terminal domain (CT) composed of proline-rich and coiled-coil regions (Fig. 6A, upper panel). Immunoprecipitation of these forms of CIN85 revealed that Dab1 bound to the N-terminal fragment as well as to full length CIN85 (FL) (Fig. 6A).
|
PR). As expected, neither Dab1-R488A nor Dab1-PR bound to CIN85 (Fig. 6B, lanes 1–3). Although the Ala mutations at both S400 and S491 (Dab1-2A) enhanced the binding of Dab1 to CIN85, the mutations failed to induce CIN85 binding to the R488A or PR mutant form of Dab1 (Fig. 6B, lanes 4–6). Whether the co-localization is dependent on PxxxPR motif in Dab1 was also examined using Dab1-R488A or Dab1-PR mutant expressed in COS-7 cells by immunofluorescent staining. Co-localization with CIN85 was lost as a result of these mutations (Supplementary Fig. S3). Ser491 is the critical phosphorylation site in Dab1 regulating binding to CIN85/CD2AP
Using a single Ala mutant of Ser400 (S400A) or Ser491 (S491A), we next determined which phosphorylation site is responsible for inhibition of binding. Both mutants exhibited CIN85 binding when not phosphorylated, as did wild-type Dab1 (Fig. 6C, lanes 6 and 8). Cdk5–p35-dependent phosphorylation of Dab1-S400A markedly reduced this binding (Fig. 6C, lane 7). In contrast, Dab1-S491A and Dab1-2A did not affect binding ability even after phosphorylation (Fig. 6C, lanes 9 and 11). These results clearly indicate that Ser491 is the critical phosphorylation site in the regulation of the association with CIN85. In addition, CD2AP also showed the same binding properties to the Dab1 single Ala mutant as CIN85 (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
, and CPβ as proteins that bind specifically to the unphosphorylated form of Dab1. Ser491 was the critical phosphorylation site that suppressed the binding of CIN85. Ser491 locates at the C-terminal end close to the PTPAPR sequence, conforming to the PxxxPR or Px(P/A)xxR atypical SH3 binding motif for the SH3 domains found in CIN85/CD2AP family proteins. These results indicate that CIN85 is a Dab1 C-terminal domain binding protein, and the interaction is regulated by phosphorylation of Dab1 at Ser491 by Cdk5. The present study raises the possibility that Cdk5 participates in Reelin signaling from Dab1 to the actin cytoskeleton by regulating the interaction between Dab1 and CIN85. Cdk5 can phosphorylate Dab1 at multiple sites in vitro (Ohshima et al. 2007), but only Ser491 has been identified as the in vivo phosphorylation site (Keshvara et al. 2002). Thus, phosphorylation of Dab1 by Cdk5 has not been fully investigated. We used a two-dimensional phosphopeptide map to determine the number of phosphorylation sites. Four phosphorylation spots were detected in in vitro phosphorylation, and three were observed to be involved in cellular phosphorylation. The same sites were phosphorylated by either Cdk5–p35 or Cdk5–p39. The difference in phenotype observed in p35- or p39-deficient mice (Ko et al. 2001) may not be explained by their substrate specificity but rather by expression stages in development (Takahashi et al. 2003). Other SP or TP sites could be phosphorylated in HEK293 cells since the double mutant at Ser400 and Ser491 was still shifted upward when co-transfected with Cdk5–p35. However, taking into consideration that two-dimensional phosphopeptide map can allow the detection of most of phosphorylation sites at a time and that our two-dimensional phosphopeptide map of neuronal phosphorylation was almost identical to that reported by Ballif et al. (2004) for neurons (see their supplementary fig. S4), it is highly probable that Ser400 and Ser491 are two major in vivo phosphorylation sites in Dab1.
Dab1 is an adaptor protein that transmits signals through protein–protein interactions. It is well established that the N-terminal PTB domain of Dab1 binds to ApoER2 or VLDLR Reelin receptors in response to Reelin stimulation, inducing Tyr-phosphorylation of Dab1 by Src family kinases. Phospho-Tyr residues in the central region of the Dab1 molecule then interact with proteins, including Lis1 and several SH3 domain-containing proteins, to activate downstream pathways (Assadi et al. 2003; Bock et al. 2003; Pramatarova et al. 2003; Ballif et al. 2004; Suetsugu et al. 2004). However, the function of the C-terminal region is poorly understood. Mutant mice with only a single copy of the Dab-p45 gene show abnormal development of the neocortex and hippocampus (Herrick & Cooper 2002), suggesting that the C-terminal domain of Dab1-p80 is necessary for correct functioning of Reelin–Dab1 signaling. Consequently, we hypothesized that Cdk5 phosphorylation in the C-terminal tail domain also modulates the Reelin–Dab1 pathway through regulation of interactions with other proteins and we searched comprehensively for such proteins. We applied a method used for identification of Tyr-phosphorylated Dab1-binding proteins (Ballif et al. 2004). Proteins newly identified here were CIN85, CD2AP, and subunits of CP, CP and CPβ. CIN85 and CD2AP are a family of SH3 domain-containing adaptor proteins. CP is an actin filament-capping protein at the barbed end involved in stabilizing the filaments.
The present results provide the first experimental evidence of CIN85, CD2AP, and subunits of CP, CP and CPβ as Dab1 binding proteins. After identification of these proteins, however, we found the paper (Kowanetz et al. 2003b) that shows the interaction between Dab1 and CIN85 in vitro. Kowanetz et al. (2003b) have indicated the binding of Dab1 expressed in cultured cells to GST-SH3 domain of CIN85 by pulldown assay. The SH3 domains of CIN85 and CD2AP are unique in having a binding preference for a PxxxPR or Px(P/A)xxR motif, but not for the classical SH3-binding motif, PxxP (Kowanetz et al. 2003a; Kurakin et al. 2003). Dab1 possesses the PTPAPR sequence at amino acids 483–488 in the C-terminal tail domain. Further, the SH3 domains of CIN85 bind to Dab2, another mammalian Dab homologue, at the PKPAPR sequence in the C-terminal region (Kowanetz et al. 2003b). Dab2 has the N-terminal PTB domain with a high homology to Dab1 and is capable of interacting with lipoprotein receptors. Although the overall C-terminal region differs between Dab1 and Dab2, they share a similar proline-rich region encompassing about 100 amino acids containing the P(T or K)PAPR sequence. With respect to CP, the retention of CP and CPβ in the Dab1–Sepharose beads is explained by reports that CP binds to CIN85/CD2AP (Hutchings et al. 2003; Bruck et al. 2006). The CP binding site is upstream of the C-terminal coiled-coil region of CIN85. It is likely that CP binds to Dab1–Sepharose beads through interaction with CIN85/CD2AP. This is consistent with the observation that CP and CPβ were eluted from the column with 1 M NaCl before CIN85/CD2AP elution with 1 M MgCl2, although we have not shown experimentally the binding is indirect.
Regulation by Tyr phosphorylation has been reported for the interaction of CIN85 with several PxxxPR motifs containing proteins. For example, the interaction of CIN85 with Cbl is enhanced by receptor tyrosine kinase-induced tyrosine phosphorylation of Cbl (Soubeyran et al. 2002). Cbl is an E3 ligase involved in receptor endocytosis, which was first used for isolation of CIN85 (Cbl interacting protein 85) in a yeast two-hybrid system (Take et al. 2000). CFBP (a CIN85/CD2AP family binding protein) was recently reported to interact with CIN85/CD2AP in a Tyr phosphorylation-dependent manner (Konishi et al. 2006). The phospho-Tyr residues are located at a distance from the PxxxPR sequence, and protein conformation may play a critical role in the binding.
The Cdk5-dependent Ser phosphorylation found here might involve more direct regulation, depending on amino acid sequences around the PxxxPR binding motif. Ser491 is located three amino acids downstream from the PTPAPR sequence (see Dab1 in Table 1). Introduction of a negative charge into Ser491 after phosphorylation could decrease the binding strength between Dab1 and CIN85, electrically and/or conformationally. Examination of the amino acid sequences around the PxxxPR motif of known PxxxPR motif-containing proteins reveals that many have the possible Cdk consensus phosphorylation sites, TP or SP, around or within the PxxxPR motif (Table 1). It was recently reported that interaction of Xp95 (a Xenopus orthologue of AIP1/Alix) with CIN85 is regulated by mitotic phosphorylation at Thr residue in the PxxxPR motif (Dejournett et al. 2007; see Xp95 in Table 1). These results suggest that Ser/Thr phosphorylation could play an inhibitory role in the interactions between the SH3 domains of CIN85/CD2AP family proteins and their binding partners within the PxxxPR motif.
|
The Dab1–CIN85 interaction may represent a process of the ubiquitin-dependent Dab1 degradation by proteasome. Dab1 is degraded by proteasome when Dab1 is phosphorylated at Tyr by Reelin stimulation (Bock et al. 2004). Dab1 is ubiquitinated in a Cbl-dependent manner when Tyr phosphorylation of Dab1 is enhanced by co-expression with Fyn kinase (Suetsugu et al. 2004). Together, CIN85 may facilitate the ubiquitination of Dab1 with Cbl by forming a ternary complex transiently.
In summary, although several signaling pathways are now considered to be activated in downstream of Reelin signaling, the present results suggest another pathway that directly links Dab1 and the endocytosis machinery and/or actin cytoskeleton through CIN85 and CP. In the pathway, Cdk5 may regulate the interaction of Dab1with CIN85 by phosphorylation of Dab1 at Ser491.
|
Experimental procedures |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Chemicals and antibodies used included roscovitine and bacterial alkaline phosphatase (Wako Chemicals, Osaka, Japan); anti-Cdk5 (DC17), anti-Myc (9E10), anti-p35 (C-19) and anti-CD2AP (H-290) antibodies (Santa Cruz Biotechnology, Santa Cruz, CA); anti-β-tubulin (TUB 2.1), anti-Flag (M2) and anti-actin (A2066) antibodies (Sigma, St Louis, MO); anti-CIN85 antibody (Calbiochem, San Diego, CA); anti-phosphotyrosine antibody (4G10) (Upstate Biotechnology, Lake Placid, NY); goat anti-Dab1 antibody (AB9012, Chemicon, Temecula, CA); anti-CP1 and anti-CPβ2 antibodies (The Development Studies Hybridoma Bank, The University of Iowa, Department of Biological Sciences, Iowa City, IA); anti-Dab1-C-terminal domain (CT) antibody, produced by exposure of rabbits to a KLH-conjugated Dab1-C-terminal peptide (C-SGEPSGDNISPQDGS); anti-phospho-Ser400 (pS400) and phospho-Ser491 (pS491) antibodies, by exposure of rabbits to the phosphopeptides, DSARSpSPQSDK-C-KLH (for pS400) and KLH-C-PAPRQSpSPSKSS (for pS491), respectively.
Plasmid construction
The primers used for plasmid construction are described in Supplementary Table S1. cDNA of each Dab1 isoform was obtained by PCR amplification of the cDNA library prepared from adult mouse brain total RNAs, and subcloned into a pCR2.1 vector (Invitrogen, Carlsbad, CA). For construction of GST-Dab1-His bacterial expression plasmids, each Dab1 isoform cDNA in pCR2.1 was amplified by PCR. The PCR products were ligated in pET23a(+)-GST (GST-His plasmid), which was prior constructed from pET23a(+) (Novagen, Madison, WI). A Myc-tagged Dab1-p80 plasmid for expression in mammalian cells was constructed in pcDNA3 by PCR amplification using Dab1-p80 cDNA in pCR2.1 as a template. Site-directed or deletion mutagenesis was carried out using the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). pCMV-Cdk5 and pCMV-p35 were provided by L-H. Tsai, and pcDNA3-Reelin was obtained from T. Curran. CIN85 full length (FL, amino acids 2–665), the N-terminal fragment (NT, amino acids 2–327) or C-terminal fragment (CT, amino acids 328–665) tagged with Flag at the N-terminal end, was amplified by PCR using a CIN85 cDNA (Invitrogen) as a template and inserted into the HindIII/BamHI site of a pFLAG-CMV2 vector. All cDNA constructs were confirmed by DNA sequence.
Cell culture, transfection, immunoprecipitation and immunofluorescent staining
HEK293 cell culture, plasmid transfections and immunoprecipitations were performed as previously described (Kamei et al. 2007). Rat brain cortical neurons were cultured as described previously (Kamei et al. 2007).
The cellular localization of Dab1 proteins was observed in COS-7 cells co-transfected with pCMV2-Flag-CIN85 using Lipofectamine 2000 (Invitrogen). Twenty-four hours after transfection, cells were stained doubly with anti-Dab1-CT and anti-Flag antibody, followed by anti-rabbit and anti-mouse secondary antibodies conjugated with Alexa488 and Alexa546, respectively. Fluorescent images were captured by a laser-scanning confocal microscope LSM5PASCAL (Carl Zeiss, Oberkochen, Germany).
In vitro phosphorylation, metabolic labeling of Dab1 in HEK293T cells and cultured neurons, and two-dimensional phosphopeptide map analysis
Dab1 isoforms tagged with GST and His were expressed in Escherichia coli BL21 (DE3) and purified with a Ni-NTA column (Qiagen, Hilden, Germany). In vitro phosphorylation of Dab1 by Cdk5–p35 and metabolic labeling of Dab1 in HEK293 cells or cultured neurons were conducted essentially as described previously (Kamei et al. 2007). Dab1 was immunoprecipitated from the HEK293 cell extracts with anti-Myc antibody 9E10 or from neuron extracts with anti-Dab1-CT antibody. Two-dimensional phosphopeptide map analysis was carried out as described previously (Kamei et al. 2007).
Preparation of rat or mouse brain extracts and identification of Dab1-binding proteins
Rat whole brains at post-natal day 3 (P3) or mouse cortex were homogenized at 4 °C in lysis buffer (30 mM Hepes, pH 7.5, 100 mM NaCl, 0.5% Triton X-100, 10% glycerol, 5 mM MgCl2, 1 mM EGTA, 25 mM NaF, 10 mM β-glycerophosphate, 1 mM sodium orthovanadate, 1 mM DTT and protease inhibitors) using a Polytron homogenizer. The brain extracts was obtained by centrifugation at 20 000 g for 60 min at 4 °C.
Dab1 binding proteins were isolated according to the method described by Ballif et al. (2004) with modifications. GST-Dab1-CT or GST-Dab1-CT-S400/491A (2A), which was prepared from an Escherichia coli extracts, was bound to glutathione–Sepharose beads (GSH-beads, GE Healthcare Bio-Sciences) and then phosphorylated by Cdk5–p35. The rat brain extracts, pre-cleared by GSH-beads and then GSH-beads bound to GST-His, were mixed with GSH-beads conjugated with either GST-Dab1-CT, phosphorylated GST-Dab1-CT or phosphorylated GST-Dab1-CT-2A for 6 h at 4 °C. The beads were washed with lysis buffer, followed by sequential elution with 1 M NaCl buffer (30 mM Hepes, pH 7.5, 1 M NaCl, 0.5% Triton X-100), 1 M MgCl2 buffer (30 mM Hepes, pH 7.5, 1 M MgCl2, 0.5% Triton X-100), and then Laemmli's SDS-PAGE sample buffer (LSB). Proteins specifically bound to unphosphorylated Dab1 were analyzed by LC-MS/MS after trypsin digestion (Taoka et al. 2003).
SDS-PAGE and immunoblotting
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting were performed as described previously (Kamei et al. 2007). Primary antibodies were diluted with CanGet Signal solutions (TOYOBO, Osaka, Japan). Horseradish peroxidase-conjugated or alkaline phosphatase-conjugated secondary antibody (Dako, Glostrup, Denmark) was used for detection of reactions.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
* Correspondence: E-mail: hisanaga-shinichi{at}tmu.ac.jp
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Ballif, B.A., Arnaud, L., Arthur, W.T., Guris, D., Imamoto, A. & Cooper, J.A. (2004) Activation of a Dab1/CrkL/C3G/Rap1 pathway in Reelin-stimulated neurons. Curr. Biol. 14, 606–610.[CrossRef][Medline]
Beffert, U., Weeber, E.J., Morfini, G., Ko, J., Brady, S.T., Tsai, L.H., Sweatt, J.D. & Herz, J. (2004) Reelin and cyclin-dependent kinase 5-dependent signals cooperate in regulating neuronal migration and synaptic transmission. J. Neurosci. 24, 1897–1906.
Bock, H.H., Jossin, Y., Liu, P., Forster, E., May, P., Goffinet, A.M. & Herz, J. (2003) Phosphatidylinositol 3-kinase interacts with the adaptor protein Dab1 in response to Reelin signaling and is required for normal cortical lamination. J. Biol. Chem. 278, 38772–38779.
Bock, H.H., Jossin, Y., May, P., Bergner, O. & Herz, J. (2004) Apolipoprotein E receptors are required for reelin-induced proteasomal degradation of the neuronal adaptor protein Disabled-1. J. Biol. Chem. 279, 33471–33479.
Bruck, S., Huber, T.B., Ingham, R.J., Kim, K., Niederstrasser, H., Allen, P.M., Pawson, T., Cooper, J.A. & Shaw, A.S. (2006) Identification of a novel inhibitory actin-capping protein binding motif in CD2-associated protein. J. Biol. Chem. 281, 19196–19203.
Chae, T., Kwon, Y.T., Bronson, R., Dikkes, P., Li, E. & Tsai, L.H. (1997) Mice lacking p35, a neuronal specific activator of Cdk5, display cortical lamination defects, seizures, and adult lethality. Neuron 18, 29–42.[CrossRef][Medline]
Cuitino, L., Matute, R., Retamal, C., Bu, G., Inestrosa, N.C. & Marzolo, M.P. (2005) ApoER2 is endocytosed by a clathrin-mediated process involving the adaptor protein Dab2 independent of its Rafts’ association. Traffic 6, 820–838.[CrossRef][Medline]
D’Arcangelo, G., Homayouni, R., Keshvara, L., Rice, D.S., Sheldon, M. & Curran, T. (1999) Reelin is a ligand for lipoprotein receptors. Neuron 24, 471–479.[CrossRef][Medline]
Dejournett, R.E., Kobayashi, R., Pan, S., Wu, C., Etkin, L.D., Clark, R.B., Bogler, O. & Kuang, J. (2007) Phosphorylation of the proline-rich domain of Xp95 modulates Xp95 interaction with partner proteins. Biochem. J. 401, 521–531.[CrossRef][Medline]
Dhavan, R. & Tsai, L.H. (2001) A decade of CDK5. Nat. Rev. Mol. Cell Biol. 2, 749–759.[CrossRef][Medline]
Dikic, I. (2002) CIN85/CMS family of adaptor molecules. FEBS Lett. 529, 110–115.[CrossRef][Medline]
Herrick, T.M. & Cooper, J.A. (2002) A hypomorphic allele of dab1 reveals regional differences in reelin-Dab1 signaling during brain development. Development 129, 787–796.
Hiesberger, T., Trommsdorff, M., Howell, B.W., Goffinet, A., Mumby, M.C., Cooper, J.A. & Herz, J. (1999) Direct binding of Reelin to VLDL receptor and ApoE receptor 2 induces tyrosine phosphorylation of disabled-1 and modulates tau phosphorylation. Neuron 24, 481–489.[CrossRef][Medline]
Hisanaga, S. & Saito, T. (2003) The regulation of cyclin-dependent kinase 5 activity through the metabolism of p35 or p39 Cdk5 activator. Neurosignals 12, 221–229.[CrossRef][Medline]
Howell, B.W., Gertler, F.B. & Cooper, J.A. (1997) Mouse disabled (mDab1): a Src binding protein implicated in neuronal development. EMBO J. 16, 121–132.[CrossRef][Medline]
Hutchings, N.J., Clarkson, N., Chalkley, R., Barclay, A.N. & Brown, M.H. (2003) Linking the T cell surface protein CD2 to the actin-capping protein CAPZ via CMS and CIN85. J. Biol. Chem. 278, 22396–22403.
Kamei, H., Saito, T., Ozawa, M., Fujita, Y., Asada, A., Bibb, J.A., Saido, T.C., Sorimachi, H. & Hisanaga, S. (2007) Suppression of calpain-dependent cleavage of the CDK5 activator p35 to p25 by site-specific phosphorylation. J. Biol. Chem. 282, 1687–1694.
Keshvara, L., Magdaleno, S., Benhayon, D. & Curran, T. (2002) Cyclin-dependent kinase 5 phosphorylates disabled 1 independently of Reelin signaling. J. Neurosci. 22, 4869–4877.
Ko, J., Humbert, S., Bronson, R.T., Takahashi, S., Kulkarni, A.B., Li, E. & Tsai, L.H. (2001) p35 and p39 are essential for cyclin-dependent kinase 5 function during neurodevelopment. J. Neurosci. 21, 6758–6771.
Konishi, H., Tashiro, K., Murata, Y., Nabeshi, H., Yamauchi, E. & Taniguchi, H. (2006) CFBP is a novel tyrosine-phosphorylated protein that might function as a regulator of CIN85/CD2AP. J. Biol. Chem. 281, 28919–28931.
Kowanetz, K., Szymkiewicz, I., Haglund, K., Kowanetz, M., Husnjak, K., Taylor, J.D., Soubeyran, P., Engstrom, U., Ladbury, J.E. & Dikic, I. (2003a) Identification of a novel proline-arginine motif involved in CIN85-dependent clustering of Cbl and down-regulation of epidermal growth factor receptors. J. Biol. Chem. 278, 39735–39746.
Kowanetz, K., Terzic, J. & Dikic, I. (2003b) Dab2 links CIN85 with clathrin-mediated receptor internalization. FEBS Lett. 554, 81–87.[CrossRef][Medline]
Kurakin, A.V., Wu, S. & Bredesen, D.E. (2003) Atypical recognition consensus of CIN85/SETA/Ruk SH3 domains revealed by target-assisted iterative screening. J. Biol. Chem. 278, 34102–34109.
Lynch, D.K., Winata, S.C., Lyons, R.J., Hughes, W.E., Lehrbach, G.M., Wasinger, V., Corthals, G., Cordwell, S. & Daly, R.J. (2003) A Cortactin-CD2-associated protein (CD2AP) complex provides a novel link between epidermal growth factor receptor endocytosis and the actin cytoskeleton. J. Biol. Chem. 278, 21805–21813.
Mejillano, M.R., Kojima, S., Applewhite, D.A., Gertler, F.B., Svitkina, T.M. & Borisy, G.G. (2004) Lamellipodial versus filopodial mode of the actin nanomachinery: pivotal role of the filament barbed end. Cell 118, 363–373.[CrossRef][Medline]
Morimura, T., Hattori, M., Ogawa, M. & Mikoshiba, K. (2005) Disabled1 regulates the intracellular trafficking of reelin receptors. J. Biol. Chem. 280, 16901–16908.
Ohshima, T., Ogawa, M., Veeranna, Hirasawa, M., Longenecker, G., Ishiguro, K., Pant, H.C., Brady, R.O., Kulkarni, A.B. & Mikoshiba, K. (2001) Synergistic contributions of cyclin-dependant kinase 5/p35 and Reelin/Dab1 to the positioning of cortical neurons in the developing mouse brain. Proc. Natl. Acad. Sci. USA 98, 2764–2769.
Ohshima, T., Suzuki, H., Morimura, T., Ogawa, M. & Mikoshiba, K. (2007) Modulation of Reelin signaling by Cyclin-dependent kinase 5. Brain Res. 1140, 84–95.
Ohshima, T., Ward, J.M., Huh, C.G., Longenecker, G., Veeranna, Pant, H.C., Brady, R.O., Martin, L.J. & Kulkarni, A.B. (1996) Targeted disruption of the cyclin-dependent kinase 5 gene results in abnormal corticogenesis, neuronal pathology and perinatal death. Proc. Natl. Acad. Sci. USA 93, 11173–11178.
Pramatarova, A., Ochalski, P.G., Chen, K., Gropman, A., Myers, S., Min, K.T. & Howell, B.W. (2003) Nck β interacts with tyrosine-phosphorylated disabled 1 and redistributes in Reelin-stimulated neurons. Mol. Cell. Biol. 23, 7210–7221.
Rice, D.S., Sheldon, M., D’Arcangelo, G., Nakajima, K., Goldowitz, D. & Curran, T. (1998) Disabled-1 acts downstream of Reelin in a signaling pathway that controls laminar organization in the mammalian brain. Development 125, 3719–3729.[Abstract]
Sasaki, Y., Cheng, C., Uchida, Y., Nakajima, O., Ohshima, T., Yagi, T., Taniguchi, M., Nakayama, T., Kishida, R., Kudo, Y., Ohno, S., Nakamura, F. & Goshima, Y. (2002) Fyn and Cdk5 mediate semaphorin-3A signaling, which is involved in regulation of dendrite orientation in cerebral cortex. Neuron 35, 907–920.[CrossRef][Medline]
Soubeyran, P., Kowanetz, K., Szymkiewicz, I., Langdon, W.Y. & Dikic, I. (2002) Cbl-CIN85-endophilin complex mediates ligand-induced downregulation of EGF receptors. Nature 416, 183–187.[CrossRef][Medline]
Stolt, P.C. & Bock, H.H. (2006) Modulation of lipoprotein receptor functions by intracellular adaptor proteins. Cell. Signal. 18, 1560–1571.[CrossRef][Medline]
Suetsugu, S., Tezuka, T., Morimura, T., Hattori, M., Mikoshiba, K., Yamamoto, T. & Takenawa, T. (2004) Regulation of actin cytoskeleton by mDab1 through N-WASP and ubiquitination of mDab1. Biochem. J. 384, 1–8.[CrossRef][Medline]
Takahashi, S., Saito, T., Hisanaga, S., Pant, H.C. & Kulkarni, A.B. (2003) Tau phosphorylation by cyclin-dependent kinase 5/p39 during brain development reduces its affinity for microtubules. J. Biol. Chem. 278, 10506–10515.
Take, H., Watanabe, S., Takeda, K., Yu, Z.X., Iwata, N. & Kajigaya, S. (2000) Cloning and characterization of a novel adaptor protein, CIN85, that interacts with c-Cbl. Biochem. Biophys. Res. Commun. 268, 321–328.[CrossRef][Medline]
Taoka, M., Ichimura, T., Wakamiya-Tsuruta, A., Kubota, Y., Araki, T., Obinata, T. & Isobe, T. (2003) V-1, a protein expressed transiently during murine cerebellar development, regulates actin polymerization via interaction with capping protein. J. Biol. Chem. 278, 5864–5870.
Tibaldi, E.V. & Reinherz, E.L. (2003) CD2BP3, CIN85 and the structurally related adaptor protein CMS bind to the same CD2 cytoplasmic segment, but elicit divergent functional activities. Int. Immunol. 15, 313–329.
Watanabe, S., Take, H., Takeda, K., Yu, Z.X., Iwata, N. & Kajigaya, S. (2000) Characterization of the CIN85 adaptor protein and identification of components involved in CIN85 complexes. Biochem. Biophys. Res. Commun. 278, 167–174.[CrossRef][Medline]
Accepted: 15 August 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
