GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2009) 14, 627-641. doi:10.1111/j.1365-2443.2009.01296.x
© 2009 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sato, J.
Right arrow Articles by Nojima, H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Sato, J.
Right arrow Articles by Nojima, H.

GAK, a regulator of clathrin-mediated membrane trafficking, localizes not only in the cytoplasm but also in the nucleus

 Jun Sato, Hiroyuki Shimizu, Takashi Kasama, Norikazu Yabuta and Hiroshi Nojima*

Department of Molecular Genetics, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita City, Osaka 565-0871, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
The ubiquitously expressed Cyclin G-associated kinase (GAK) regulates clathrin-mediated membrane trafficking in the cytoplasm. However, the association of GAK with a nuclear protein Cyclin G1 that is unrelated to membrane trafficking suggests an unidentified role of GAK in the nucleus. Indeed, we report here that GAK localizes in both cytoplasm and nucleus by immunostaining, ectopic expression of GFP-GAK and pull-down assays using dissected GAK fragments. GAK forms complexes not only with cyclin G1 but also with other nuclear proteins such as p53, clathrin heavy chain (CHC) and protein phosphatase 2A (PP2A) B'{alpha}1. Moreover, CHC associates with GAK via a different domain depending on whether it is in the cytoplasm or nucleus. Immunostaining revealed that about 20~30% of B'{alpha}1, cyclin G1 and p53 complex with nuclear GAK. CHC also displayed dots in the nucleus and almost all nuclear CHC signals colocalized with GAK. These observations together suggest an important function of GAK in the nucleus.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
GAK (cyclin G-associated kinase) is a serine/threonine kinase that was initially identified as an association partner of cyclin G (Kanaoka et al. 1997). As suggested by its strong homology (outside of its Ser/Thr kinase domain) to the neuronal-specific protein auxilin, the ubiquitously expressed GAK regulates clathrin-mediated membrane trafficking as it is an essential cofactor for the Hsc70-dependent uncoating of clathrin-coated vesicles (Greener et al. 2000). Furthermore, the knockdown of GAK by vector-based small hairpin RNA revealed that GAK not only induces clathrin exchange on clathrin-coated pits, but also mediates the binding of clathrin and adaptors to the plasma membrane and the trans-Golgi network (Lee et al. 2005; Zhang et al. 2005). Total internal reflectance microscopy revealed the dynamic behavior of GAK that, following transient dynamin recruitment, is transiently recruited to the clathrin puncta; this recruitment is dependent on the PTEN-like domain of GAK that binds to the phospholipids (Lee et al. 2006). GAK also phosphorylates the mu2-subunit of the AP2 adaptor complex, which suggests GAK plays a pivotal role in the assembly of clathrin-coated vesicles (Korolchuk & Banting 2002). Furthermore, GAK knockout mice showed that GAK deletion blocks development and causes lethality in adult animals by disrupting clathrin-mediated endocytosis (Lee et al. 2008). Thus, it is now well established that GAK is an essential player in membrane trafficking (Eisenberg & Greene 2007).

However, the fact that GAK associates with a nuclear protein Cyclin G1 that plays no role in membrane trafficking suggests that GAK may play other, as yet unknown, roles in cellular events other than membrane trafficking. Supporting this is the fact that the knockdown of GAK by small hairpin RNA dramatically enhanced EGFR expression and its tyrosine kinase activity, which significantly altered the spectrum of downstream EGFR signaling; it was suggested that these effects may be due to altered receptor trafficking (Zhang et al. 2004). Moreover, mass spectrometric analysis (MS/MS) identified GAK as a protein that interacts with nucleophosmin-anaplastic lymphoma kinase (NPM-ALK), whose constitutive overexpression is a key oncogenic event in anaplastic large-cell lymphomas (Crockett et al. 2004). In addition, GAK was found to interact with androgen receptor (AR) and to enhance AR activity in a ligand-dependent manner; GAK expression was also found to be significantly increased in prostate cancers that had progressed to androgen independence (Ray et al. 2006). The latter report suggests that GAK acts not only in the cytoplasm but also in the nucleus.

Another putative novel role of GAK appears to be played by its association partners, cyclin G and protein phosphatase 2A (PP2A) B’{alpha} subunit that are partially localized in the nucleus. It has been shown that PP2A is recruited to its target proteins by cyclin G1 (Okamoto et al. 1996), a member of the cyclin family (Tamura et al. 1993). The cyclin G1 gene is a target of the p53 tumor suppressor protein (Okamoto & Beach 1994) and is induced in a p53-dependent manner in response to DNA damage (Kimura & Nojima 2002). Cyclin G1 was found to complex with and regulate the active PP2A holoenzymes and control MDM2 by dephosphorylating its T216 and S166 residues; this leads to the destabilization of p53 (Okamoto et al. 2002). Moreover, cyclin G1 interacts directly with MDM2 after DNA damage and promotes ARF/MDM2 complex formation, thereby downregulating p53 levels; indeed, the accumulation of p53 that is observed after DNA damage was enhanced in cyclin G1–/– cells (Kimura & Nojima 2002), even though the cyclin G1 null mouse is normal at birth and remains healthy in adulthood (Kimura et al. 2001). These findings suggest that cyclin G1 is a key regulator of the p53-MDM2 network that acts in part in the nucleus by regulating PP2A (Chen 2002).

PP2A is one of the highly abundant and ubiquitously expressed Ser/Thr phosphatases that regulate multiple signaling pathways in eukaryotic cells (Janssens et al. 2005; Westermarck & Hahn 2008). PP2A consists of three subunits, namely, invariable catalytic (C) and structural (A) subunits and a variable regulatory (B) subunit. Three unrelated PP2A regulatory subunit families denoted as B, B' and B'' have been identified. With regard to the B' family, these are encoded by five distinct mammalian genes. Moreover, many isoforms are generated from these genes. Of relevance to this paper is the B'{alpha} (alternatively called B56gamma) subunit subfamily, which includes three alternative splicing variants with differing subcellular localizations (Ito et al. 2000), namely, either in the nucleus (B'{alpha}1 and B'{alpha}2) or in the cytoplasm (B'{alpha}3). Hereafter, we will sometimes refer to PP2A B'{alpha} simply as B'{alpha}.

Thus, it is apparent that GAK, at least partly, plays a role in the nucleus. Nonetheless, no direct evidence for the nuclear activity of GAK has been reported to date. In the present study, we show that GAK localizes not only in the cytoplasm but also in the nucleus, where it complexes with nuclear proteins such as p53, clathrin heavy chain (CHC) and B'{alpha}1, a nuclear form of the B'{alpha} subunit. Moreover, we show here that association domains between GAK and CHC are distinct from cytoplasm and nucleus.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
GAK localizes in both the cytoplasm and nucleus

Reflecting its role as a regulator of clathrin-mediated membrane trafficking, GAK localizes in the cytoplasm (Greener et al. 2000), in particular at the trans-Golgi network (TGN; Kametaka et al. 2007). When we examined the localization of GAK by immunofluorescence with a polyclonal anti-GAK antibody (pGAK) or a monoclonal anti-GAK antibody (1E1) that we prepared previously (Kanaoka et al. 1997) using laser scanning microscope (Fig. 1A,B), we could show that at least some immunostained images are detected as the filamentous structure in human cells such as HEF (human embryonic fibroblasts), KD (a primary culture of human lip fibroblasts) and HeLa (cervical cancer). Moreover, we always observed not only the filamentous images in the cytoplasm, but also the dotted images in the nucleus (Fig. 1A,B); indeed, the nuclear signals detected here were stronger than the cytoplasmic signals when 1E1 monoclonal antibody was used (Fig. 1B).


Figure 1
View larger version (41K):
[in this window]
[in a new window]

  		Figure 1  Subcellular localization of GAK, as detected by polyclonal and monoclonal antibodies. (A) Immunostaining of human GAK in HEF, KD and HeLa cells with the polyclonal anti-GAK antibody pGAK. (B) Immunostaining of HEF cells with the monoclonal anti-GAK antibody 1E1. (C) Western blot analysis of endogenous GAK in HeLa cell extract by using the monoclonal anti-GAK monoclonal antibodies 9–10 and 9–13. As a control, the purified GST-GAK (kinase domain) fusion protein was probed. (D) Immunostaining of TIG-1 cells with the monoclonal anti-GAK antibodies 9–10 and 9–13 (middle panels). As a control, the cells were stained with the pGAK polyclonal antibody (left panels). (E) Immunostaining of TIG-1 cells with pGAK polyclonal antibody and the ER staining marker ER-TrackerTM Red. Arrows indicate the colocalized regions. (F) Immunostaining of TIG-1 cells with pGAK polyclonal antibody and the mitochondrial staining marker MitoTrackerTM Red. Arrowheads indicate the colocalized regions. Immunofluorescence micrographs were obtained by laser scanning microscope (Zeiss, LSM410 or LSM510) and dual color images were obtained and analyzed by co-localization image processing software. The merged images are shown in the right panels (D, E and F). Bars represent 10 µm.

 
Because the 1E1 monoclonal antibody is not useful for Western blot analysis due to its low sensitivity, we prepared new monoclonal antibodies against GAK and found that two of the clones denoted as 9–10 and 9–13 detected purified GAK specifically and with high sensitivity (Fig. 1C). They also both detected a doublet of bands in HeLA S3 cell extracts around 170 kDa (Fig. 1C). When we immunostained TIG-1 cells with these monoclonal antibodies, we found that clone 9–10 only detected nuclear GAK while clone 9–13 detected GAK in both the nucleus and cytoplasm (Fig. 1D). As expected, most of the dotted images in the nucleus and the filamentous images in the cytoplasm that were obtained by immunostaining with the polyclonal antibody pGAK and the 9–10 and 9–13 monoclonal antibodies overlapped almost completely (Fig. 1D). These results confirmed by one polyclonal antibody and three distinct monoclonal antibodies indicate that GAK also localizes in the nucleus, even though GAK does not appear to have a nuclear localization signal (NLS). When we performed siRNA-mediated knockdown experiment to examine if these nuclear signals were reduced by depletion of GAK mRNA, we found that the GAK-knockdown cells showed the cell cycle arrest at metaphase when nuclear membrane was lost (our unpublished result); thus, it was unable to analyze the status of nuclear localization of GAK under these conditions.

Because the filamentous images of GAK in the cytoplasm (Fig. 1A,B,D) resembles those of endoplasmic reticulum (ER) or mitochondria, we stained the human TIG-1 cells (human fetal lung fibroblasts) with the ER (or TGN) staining marker ER-TrackerTM Red and the mitochondrial staining marker MitoTrackerTM Red to examine whether at least some of the cytoplasmic GAK colocalize with these cytoplasmic structures. We first confirmed that the cytoplasmic GAK partly localized at the TGN (Kametaka et al. 2007) or ER in TIG-1 cells (Fig. 1E, arrows). Moreover, we found that at least some of the immunostained images of GAK also coincided with those of mitochondria (Fig. 1F, arrowheads). Notably, no nuclear dot was observed when we skipped the permeabilization process with 0.1% Triton X-100 after the cell fixation (see Experimental procedures) to highlight the ER or TGN-staining with ER-TrackerTM Red (Fig. 1E); this is because the nuclear membrane remained intact to inhibit the entrance of anti-GAK antibody into the nucleus. Thus, Triton-treatment is essential to visualize the nuclear localization of GAK.

To examine whether these nuclear dots of GAK colocalize with nuclear bodies (NBs) that are recognized by antibodies against SF2 (nuclear stress bodies; Biamonti 2004; Chiodi et al. 2004), promyelocytic leukaemia (PML) protein (PML-NB; Bernardi & Pandolfi 2007), or coilin (Cajal body; Stanek & Neugebauer 2006), we immunostained TIG-1 cells using the relevant antibodies; however, we could detect no remarkable colocalization of GAK with these NBs (Supporting Fig. S1).

GAK is phosphorylated in the nucleus

To determine if GAK is modified by phosphorylation, we used human embryonic kidney 293T cells because they yield comparatively large amounts of cell extract per culture dish. Upon Western blot analysis of the 293T extract, the 9–10 and 9–13 antibodies revealed the same doublet of bands around 170 kDa (Fig. 2A) that were detected by these antibodies in HeLa S3 cell extracts (Fig. 1C). When we added {lambda}-phosphatase to the 293T cell extract in the presence or absence of phosphatase inhibitors, the upper band that was detected by both the 9–10 and 9–13 antibodies migrated slightly faster; this effect was blocked when phosphatase inhibitors were also present (Fig. 2A). Ectopically expressed 6Myc-GAK showed similar band shifts, which indicates that exogenous GAK is also subjected to phosphorylation (Fig. 2B). Significantly, the kinase-dead form of GAK (M-GAK-KD) also showed similar band shifts (Fig. 2C). This indicates that GAK phosphorylation is not due to self-phosphorylation. Notably, the lower band of the doublets shown by asterisks in Fig. 2A disappeared when the insoluble materials were removed by centrifugation before the extract was dissolved in sample buffer for electrophoresis (Fig. 2B,C); this band may thus be a nonspecific background artifact or another modified form of GAK that is bound to the insoluble materials.


Figure 2
View larger version (54K):
[in this window]
[in a new window]

  		Figure 2  Western blot analysis of GAK and subcellular localization of GFP-GAK. (A–D) Western blot analysis of phosphorylation modifications of endogenous GAK (A) or ectopically expressed Myc-tagged wild-type GAK (M-GAK; B) or kinase-dead GAK (M-GAK-KD; C) in 293T cells by examining the effect of {lambda}-phosphatase (PPase) and/or phosphatase inhibitors. GAK was detected by the 9–10 or 9–13 anti-GAK monoclonal antibodies (A) or the anti-Myc monoclonal antibody (B, C). The arrowsheads indicate the shifting of the band due to (de)phosphorylation. The asterisk in (A) denotes the lower band of the doublet bands (see Figs 1C and 2G). (D) Western blot analysis of cytoplasmic and nuclear fractions of 293T cell extracts using anti-Orc2 (nuclear control), anti-{alpha} tubulin (cytoplasmic control), or anti-GAPDH (loading control) antibodies (i) or 9–10 (ii), 9–13 (iii), or pGAK (iv) anti-GAK antibodies. Cyto, cytoplasmic fraction; Nuc, nuclear fraction. The arrows and small arrowheads indicate the bands detected specifically in the nuclear fraction alone and primarily in the cytoplasmic fraction, respectively. The large arrowhead and the asterisk denote the upper and lower bands of the doublet bands, respectively (see Figs 1C and 2A). The position of the 170 kDa sizemarker is shown by the horizontal bars. (E–G) Subcellular localization of GFP-GAK after its ectopic expression in HeLa cells. (E) Schematic depiction of plasmid constructs expressing intact GAK (GAK full), GAK-N (which contains the kinase domain), and GAK-C (which lacks the kinase domain). (F) Fluorescent images of GAK full, GAK-N, GAK-C or vector alone, as observed under a fluorescence microscope. Bar = 10 µm. (G) Western blot analysis of HeLa cell extracts expressing GAK full, GAK-N, GAK-C or vector alone by using anti-GFP antibody. Arrows indicate the intact, appropriately sized, GFP-fusion proteins.

 
To confirm the nuclear localization of GAK biochemically, we isolated nuclear and cytoplasmic proteins from 293T cells and subjected them to Western blot analysis using 9–10, 9–13 and pGAK anti-GAK antibodies. The successful fractionation of the cell extract was confirmed by probing the Western blots with an antibody against the cytoplasmic protein {alpha}-tubulin or the nuclear protein origin recognition complex 2 (Orc2), although it should be noted there was slight contamination of the cytoplasmic fraction by nuclear proteins and vice versa (Fig. 2D-i). We also detected GAPDH as a loading control. Close examination of the area of the SDS-polyacrylamide gel around the molecular weight of 170 kDa revealed that all anti-GAK antibodies recognized multiple bands in both the cyoplasmic and nuclear fractions (Fig. 2D-ii, -iii and -iv). Because some of these multiple bands (denoted by arrows or arrowheads) were observed by using distinct antibodies, they probably showed the GAK bands but not the nonspecific background. For example, the lower band of the doublet bands (asterisk) was detected in the nuclear but not the cytoplasmic fraction, as were three slowly-migrating bands and a weak band at 170 kDa (denoted by arrows). It also seems that the putative phosphorylated band of the doublet bands (large arrowhead) is detected primarily in the nucleus by the monoclonal antibodies. At least two additional bands (small arrowheads) were detected in both fractions but in larger amounts in the cytoplasmic fraction. These observations suggest that the cytoplasmic and nuclear GAK molecules are modified differently.

GFP-GAK localizes in both cytoplasm and nucleus

To examine which domain of GAK is required for its nuclear localization, we dissected GAK into two parts and fused them to green fluorescent protein (GFP) for in vivo visualization under a microscope (Fig. 2E). When we ectopically expressed the constructs in HeLa cells, we found that full length GAK was distributed primarily in the cytoplasm and sparsely in the nucleus (Fig. 2F). In contrast, the N-terminal construct bearing the kinase domain was observed exclusively in the cytoplasm while the C-terminal construct that lacked the kinase domain was found in both the cytoplasm and nucleus, although at a higher proportion in the nucleus than full length GAK. We confirmed by Western blot analysis that the transfected cells expressed intact GFP-GAK proteins of the expected size (Fig. 2G, arrows). These results indicate that the C-terminal part of GAK is required for its nuclear localization. Notably, the ectopically expressed GFP-GAK proteins failed to form dots in the nucleus, and were instead homogeneously distributed (Fig. 2F). This is probably because the ectopically expressed GFP-GAK escaped from modifications and failed to form a complex with its nuclear association partners. Indeed, the full size GFP-GAK protein, like the ectopically expressed full size Myc-GAK protein shown in modification, was observed upon Western blot analysis as a single band (Fig. 2G, see arrow in the leftmost lane) without any other slowly migrating bands. These putative modifications of GAK and their physiological significance will be more rigorously investigated in the future.

GAK complexes with cyclin G1, p53 and PP2A B'{alpha} in the nucleus

It is known that cyclin G1 forms a complex with PP2A B' and p53 (Okamoto et al. 2002). To determine whether GAK is included in this complex, we first dissected GAK into four parts (Fig. 3A, left), tagged them with 6Myc, and examined their association with cyclin G1 in vivo in 293T cells expressing FLAG-tagged cyclin G1. For this, the 293T cell extracts were immunoprecipitated with anti-Myc antibody and the resulting GAK immunoprecipitates were subjected to Western blot analysis using anti-FLAG antibodies. As expected, Cyclin G1 co-immunoprecipitated with full length GAK (M-GAK; Fig. 3B, lane 4). Moreover, we found that cyclin G1 associates with the non-kinase domain (M-{Delta}k; Fig. 3B, lane 6) but not the kinase domain (M-k; Fig. 3B, lane 8) of GAK. Reciprocal immunoprecipitation (Ip) using anti-FLAG antibody and Western analysis confirmed that cyclin G1 associates with M-GAK, M-{Delta}j and M-{Delta}k (Fig. 3C, lanes 4, 8 and 12), but not M-j and M-k (Fig. 3C, lanes 16 and 20) of GAK. The result indicates that cyclin G1 associates with the non-kinase domain of GAK.


Figure 3
View larger version (38K):
[in this window]
[in a new window]

  		Figure 3  GAK associates in vivo with the nuclear proteins cyclin G1 and p53. (A) Schematic depiction of the GAK fragments (left panel) and a summary of their ability to associate with cyclin G1, p53, and B’{alpha} (right panel). Colored domains are kinase domain (orange), linker domain (yellow), PTP domain (purple), Proline rich region (blue), PEST sequence (light blue), C-terminal region (green) and J domain (red). Minus, no association; plus/minus, weak association; plus, association. (B, C) GAK associates with cyclin G1 via its non-kinase domain. 293T cells were transfected with FLAG-cyclin G1 and various 6-Myc-tagged GAK fragments. The 6-Myc-tagged GAK fragments used in this study are depicted in Fig. 2A and include full length GAK (M-GAK), GAK lacking the J domain (M-{Delta}J), GAK lacking the kinase domain (M-{Delta}k), the J domain of GAK (M-J) and the GAK kinase domain (M-k). As controls, empty 6Myc-expressing (M-vec) and FLAG-expressing (F-vec) vectors were used. The cell extracts were immunoprecipitated (IP) with anti-Myc antibody and the immunoprecipitates were subjected to Western blot analysis with the same antibodies. F-G1, the anti-FLAG-cyclin G1 antibody. (D, E) GAK associates weakly with p53 via its non-kinase domain. 293T cells expressing 6Myc-tagged GAK fragments and endogenous p53 were immunoprecipitated with anti-Myc (D) or anti-p53 (E) antibodies and subjected to Western blotting with the same antibodies. Asterisks indicate nonspecific bands. TE, total cell extract. IgG, a non-immune IgG control for the p53 antibody.

 
We next examined whether GAK associates with p53 in vivo. For this, we co-transfected 293T cells with the 6Myc-GAK fragments described above. Immunoprecipitation of the GAK fragments and Western blotting with anti-p53 antibodies revealed that p53 does associate with almost full-length (see Fig. 3A) J-domain minus GAK (M-{Delta}j), albeit weakly (Fig. 3D; arrowhead in lane 6). Moreover, we showed that the nonkinase domain ({Delta}k) of GAK also associates rather strongly with p53 (Fig. 3D; arrowhead in lane 8). When we performed the reciprocal experiment using anti-p53 antibody as the immunoprecipitating antibody, only very weak GAK bands were detected for M-{Delta}j alone (Fig. 3E, arrow in lane 9). That 6Myc-GAK binds rather strongly to p53 when anti-Myc antibody is used for immunoprecipitation but binds only weakly when anti-p53 antibody is used for immunoprecipitation suggests that only a small proportion of the p53 molecules in the cell bind to GAK in vivo. It was very difficult to show the interaction of these nuclear proteins including GAK, Cyclin G1, PP2A and p53 at endogenous level primarily due to the low sensitivity of the available antibodies.

Nuclear form of PP2A B'{alpha} subunit complexes with GAK

We next determined whether PP2A B'{alpha}1 (depicted in the left panel of Fig. 4A) associates with GAK by co-transfecting 293T cells with FLAG-tagged B'{alpha}1 and the 6Myc-GAK fragments described above. Note that B'{alpha}1, B'{alpha}2 and B'{alpha}3 are generated from the same gene by alternative splicing and only differ from each other in their C-terminal portions (Fig. 4A). Immunoprecipitation of the GAK fragments with anti-Myc antibody and Western blotting with anti-FLAG antibodies revealed that B'{alpha}1 binds to M-GAK, M-{Delta}j, M-{Delta}k and M-k (Fig. 4B, lanes 4, 6, 8 and 12) but not M-j alone (Fig. 4B, lane 10), unlike cyclin G1 and p53, which do not associate with the kinase domain of GAK (Fig. 3). The ability of cyclin G1, p53 and B'{alpha}1 to associate with the various GAK fragments are summarized in the right panel of Fig. 3A.


Figure 4
View larger version (37K):
[in this window]
[in a new window]

  		Figure 4  GAK associates with all three B’{alpha} (1–3) subtypes. (A) Schematic depiction of the PP2A B'{alpha} subtypes (left panels) and a summary of their ability to associate with GAK (right panel). Two or three pluses signify a stronger association. (B) Western blots to show that GAK associates with B'{alpha}1 via its kinase and non-kinase domains. 293T cells were co-transfected with 3FLAG-tagged B'{alpha}1 (F-{alpha}1) and various 6Myc-tagged GAK fragments, immunoprecipitated (IP) with anti-Myc antibody and subjected to Western blot analysis (WB) with anti-Myc and anti-FLAG antibodies. (C, D) Western blots to show that GAK associates most strongly with B'{alpha}1, weakly with B'{alpha}3, and very faintly with B'{alpha}2. 293T cells were co-transfected with FLAG-tagged full-length GAK and Myc-tagged B' isoforms and the total cell extract (TE) was directly subjected to Western blot analysis with anti-Myc and anti-FLAG antibodies (left panels). Alternatively, the extracts were immunoprecipitated (IP) with anti-FLAG antibody (C) or anti-Myc antibody (D) and the immunoprecipitates were subjected to Western blot analysis with anti-Myc and anti-FLAG antibodies (right panels). Asterisks indicate nonspecific bands. Arrowhead in (C) indicates a band for IgG used for immunoprecipitation that served as a loading control.

 
We previously showed that B'{alpha}1 and B'{alpha}2, but not B'{alpha}3, localize predominantly in the nucleus (Ito et al. 2000). Thus, we next asked which B'{alpha} subtype preferentially associates with GAK. For this purpose, we co-transformed 293T cells with FLAG-tagged full-length GAK and N-terminally Myc-tagged B'{alpha}1, B'{alpha}2 or B'{alpha}3. Immunoprecipitation of GAK and Western blot analysis of the immunoprecipitates using anti-Myc antibodies revealed that GAK associated most strongly with B'{alpha}1 and recognized B'{alpha}2 and B'{alpha}3 very faintly and weakly, respectively (Fig. 4C, lanes 8, 10 and 12). This was confirmed by reciprocal immunoprecipitation/Western blot analysis (Fig. 4D, lanes 6, 7 and 8). It was very difficult to show the interaction of these nuclear proteins at the endogenous level because of low sensitivity of our antibodies. To cover this problem, we performed IP/Western analysis under over expression condition.

GAK associates with CHC in the nucleus

Because GAK regulates the uncoating of clathrin-coated vesicles (Eisenberg & Greene 2007), it is likely that GAK associates with CHC directly. To gain insights into this interaction, we dissected CHC into five parts according to Edeling et al. (2006) and N-terminally FLAG-tagged them (Fig. 5A, denoted as CHC1~5 in left panel). Because the CHC5 fragment was too small to be detected by FLAG-tagging, we tagged it with Myc instead. 293T cells were co-transformed with each CHC fragment and full-length GAK reciprocally tagged with Myc or FLAG (M-GAK or F-GAK, respectively). The CHC fragments were then immunoprecipitated with anti-FLAG antibody (CHC1~4) or anti-Myc antibody (CHC5) and subjected to Western blot analysis using either antibody. This revealed that GAK associates strongly with CHC1, 2 (Fig. 5B-i, lanes 6 and 7), 4 (Fig. 5B-ii, lane 4) and 5 (Fig. 5B-iii, lane 4) but only weakly with CHC3 (Fig. 5B-i, lane 8). This was confirmed by reciprocal immunoprecipitation of GAK and Western analysis of the CHC fragments contained in these immunoprecipitates as follows: CHC1 (Fig. 5C-i, lane 4), CHC2 (Fig. 5C-ii, lane 4), CHC3 (Fig. 5C-iii, lane 4), CHC4 (Fig. 5C-iv, lane 4) and CHC5 (Fig. 5C-v, lane 4).


Figure 5
View larger version (34K):
[in this window]
[in a new window]

  		Figure 5  GAK associates with CHC through distinct domains in cytoplasm and nucleus. (A) Schematic depiction of the CHC fragments used in this study (left panels) and a summary of their ability to associate with GAK (right panel). CHC was dissected into five parts according to Edeling et al. (2006). GAK strongly associates with CHC2 (green) and CHC4 (yellow) in cytoplasm, while it associates with CHC1 (red) and CHC5 (blue) in nucleus. GAK associates only weakly with CHC3 (purple) in both cytoplasm and nucleus. Two or three pluses signify a stronger association. CHC1~4 were tagged (light blue ovals) with 3FLAG (F-CHC1~4) while CHC5 was tagged with 6Myc (M-CHC5). (B, C) GAK coimmunoprecipitates with CHC1, 2, 4 and 5. 293T cells were cotransfected with Myc- or FLAG-tagged GAK along with FLAG-CHC1~4 or Myc-CHC5. The total cell extract (TE) was directly subjected to Western blot analysis with anti-Myc and anti-FLAG antibodies (left panels). Alternatively, the extracts were immunoprecipitated (IP) with anti-FLAG or anti-Myc antibody and the immunoprecipitates were subjected to Western blot analysis with anti-Myc and anti-FLAG antibodies (right panels). In (B), the CHC immunoprecipitates were examined for the presence of GAK, while (C) is the reciprocal experiment, where the GAK immunoprecipitates were examined for the presence of CHC fragments. Asterisks indicate nonspecific bands. Putative degradation products (#) are also indicated. An arrowhead (in B, lane 8) indicates F-CHC2 band that is partly overlapped with IgG band.

 
CHC associates with GAK via a different domain in the nucleus

To examine if this association pattern differs between the cytoplasm and nucleus, we fractionated the cell extracts as described above (see Fig. 2D) and performed Ip/Western analyses similar to those described above in relation to Fig. 5. We found that GAK predominantly associated with the CHC1 and 5 fragments in the nucleus (Fig. 6A) but in the cytoplasm predominantly associated with CHC2 and 4. Again, we found negligible association between GAK and the CHC3 fragment in either compartment. Successful fractionation was confirmed by subjecting the fractionated extracts to Western blot analysis using anti-Orc2 (nuclear control), anti-{alpha} tubulin (cytoplasmic control) antibodies (Fig. 6B). These observations together suggest that CHC associates with GAK via a different domain depending on whether it is in the nucleus or cytoplasm as schematically depicted in Fig. 6C (see Discussion). This in turn suggests that the nuclear function(s) of GAK and clathrin differs from their role in cytoplasmic vesicle trafficking (see Discussion). The ability of GAK to associate with each of the CHC fragments is summarized in the right panel of Fig. 5A (right panel).


Figure 6
View larger version (49K):
[in this window]
[in a new window]

  		Figure 6  CHC associates with GAK via a different domain depending on whether it is in the cytoplasm or nucleus. 293T cells were co-transfected with CHC fragments and GAK and their lysates were biochemically separated into cytoplasmic and nuclear fractions. (A) The two fractions were directly subjected to Western blot analysis with anti-Myc and anti-FLAG antibodies (left panels). Alternatively, the extracts were immunoprecipitated (IP) with anti-Myc antibody and the immunoprecipitates were subjected to Western blot analysis with both antibodies to determine the presence of FLAG-CHC fragments in the Myc-GAK immunoprecipitates (right panels). Arrowheads and asterisks indicate the CHC fragment bands and nonspecific bands, respectively. (B) As controls, the fractions were subjected to Western blot analysis using anti-Orc2 (nuclear control), anti-{alpha} tubulin (cytoplasmic control) antibodies. Cyto, cytoplasmic fraction; Nuc, nuclear fraction. (C) Schematic depiction to show how CHC associates with GAK via a different domain in the cytoplasm or nucleus. Possible interactions of a single GAK molecule to CHC (upper panels), and three GAK molecules to CHC triskelion (lower panels) are shown (see Discussion for details).

 
Subcellular colocalization of GAK with nuclear proteins

To investigate where in the cell the biochemical associations between GAK and PP2A B'{alpha}, cyclin G1, p53, and CHC occur, we subjected TIG-1 cells to immunohistochemical analysis using specific antibodies against each protein. When we examined whether GAK colocalized with PP2A B'{alpha}, we found that B'{alpha} also displayed strong nuclear dots, and that about 25% of these merged with the nuclear GAK signals (yellow; Fig. 7A). In contrast, the cytoplasmic signals colocalized poorly. Although the anti-B'{alpha} antibody cannot distinguish between the different subtypes (Ito et al. 2000), we concluded that it was predominantly the B'{alpha}1 isoform that colocalizes with GAK as nuclear dots given that B'{alpha}1 and B'{alpha}2, but not B'{alpha}3, localize in the nucleus and that GAK associates more strongly with B'{alpha}1 than with B'{alpha}2 (Fig. 4). Cyclin G1 is known to show marked immunostaining in the nucleus and sparse staining throughout the cytoplasm (Reimer et al. 1999). Indeed, about 20% and 30% of the cyclin G1 and p53 signals colocalized with GAK, respectively (Fig. 7B,C). These results suggest that about 20~30% of B'{alpha}1, cyclin G1 and p53 complex with nuclear GAK.


Figure 7
View larger version (47K):
[in this window]
[in a new window]

  		Figure 7  Nuclear co-localization of GAK with its association partners. TIG-1 cells were immunostained with the 9–13 anti-GAK monoclonal antibody (A, B) or pGAK polyclonal antibody (C, D) together with antibodies against PP2A B'{alpha} (A), cyclin G1 (B), p53 (C) or CHC (D). The bar graphs show the percentages of yellow signals relative to the total number of nuclear PP2A/cyclin G1/p53/CHC signals in the merged images. The standard deviations in the graph were calculated from 10 cells for PP2A B'{alpha} and 3 cells for each of the others. Black bars = 10 µm. White bar = 2 µm.

 
Most of the CHC signals were detected in the cytoplasm and very few of these merged with the GAK signals (Fig. 7D, upper panels). However, CHC also displayed dots in the nucleus. This was observed only when the CHC signal was highlighted because the nuclear CHC signals represent only about 5% of the cytoplasmic CHC signals (Enari et al. 2006). Surprisingly, almost all nuclear CHC dots colocalized with GAK (Fig. 7D, lower panels), which suggests that most of the nuclear CHC functions together with GAK.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
The ubiquitously expressed GAK protein is known to regulate endocytosis in the cytoplasm (Eisenberg & Greene 2007). In the present study, we showed that GAK localizes not only in the cytoplasm but also in the nucleus. This nuclear localization of GAK was confirmed by immunostaining HeLa, HEF, KD and TIG-1 cells using both polyclonal and monoclonal antibodies specific for the GAK protein (Fig. 1A,B,D), performing Western blotting with anti-GAK antibodies on the nuclear and cytoplasmic fractions of 293T cells (Fig. 2D), and observing GFP-GAK in transfected HeLa cells under a microscope (Fig. 2E–F). That the 9–10 monoclonal antibody only recognized the nuclear form of GAK while the 9–13 monoclonal antibody recognized both cytoplasmic and nuclear GAK suggests that nuclear GAK bears an epitope that is not present in cytoplasmic GAK. This in turn suggests GAK is modified somehow in either the cytoplasm or nucleus. Indeed, we found that nuclear GAK is phosphorylated (Fig. 2A–D), although we cannot yet exclude the possibility that cytoplasmic GAK is also phosphorylated (Fig. 2D). We also showed that this phosphorylation event was mediated by another, as yet unknown protein kinase rather than by self-activity because GAK-KD was phosphorylated to the same degree as wild-type GAK (Fig. 2B,C). The identification of this kinase will be the subject of our future study. The nuclear localization of GAK was independent of p53 activity because it was observed in both HEF, KD (Fig. 1A) and TIG-1 (Fig. 7) cells which harbor wild type p53, in HeLa cells (Fig. 1A) which lack p53, and in 293T cells (Fig. 6), in which p53 is present but is inactivated by large T antigen, a major early gene product encoded by SV40 (Lilyestrom et al. 2006). The results presented here thus indicate that GAK is not only important in the cytoplasm as a regulator of vesicle trafficking, but it also plays an important role in the nucleus.

Like its cytoplasmic function in vesicle trafficking, one of the nuclear functions of GAK may also relate to CHC since we showed that CHC is also present in the nucleus, albeit at much lower levels than in the cytoplasm (Fig. 7D). This nuclear localization of CHC was also reported by Enari et al. (2006). Moreover, we found that nearly all of the nuclear CHC molecules associated with GAK (Fig. 7D, see merged image). Notably, the portion of CHC that associates with GAK differs depending on whether the two proteins occur in the cytoplasm or the nucleus (Fig. 6A). Thus, while GAK associates with the middle part of CHC (CHC2 and CHC4) in the cytoplasm, it interacts with its N- and C-termini (CHC1 and CHC5) in the nucleus (Fig. 6C). The binding of GAK to the termini of CHC means that the CHC junction is occupied by GAK (see Fig. 5), which is likely to prohibit the formation of clathrin-coated vesicles in the nucleus. Moreover, it may also affect the interaction between p53 and CHC because p53 associates with the C-terminus of CHC (Enari et al. 2006). This explains why GAK only colocalized with CHC in the nucleus (Fig. 7D). Notably, it was recently reported that the p53-binding site of CHC behaved as a monomer in cells and had a higher ability to transactivate p53 (Ohmori et al. 2008). Thus, it is likely that CHC adopts a distinct conformation in the nucleus (e.g. as a single triskelion or fibrous structure). These observations together with our own suggest that GAK and CHC may collaborate not only in the cytoplasm as vesicle transporters but also in the nucleus as regulators of p53-mediated transcription.

Another apparent nuclear function of GAK is to regulate the activity of B'{alpha}1 when it is recruited by cyclin G1 to its dephosphorylation target proteins (Okamoto et al. 1996). This is shown by the following observations. First, GAK preferentially complexes with cyclin G1 (Fig. 3), which is shown to be predominantly localized in the nucleus (Fig. 7B). Second, B'{alpha}1 localized in the nucleus, although the anti-PP2A B'{alpha} antibody we used here also recognized B'{alpha}3 subunit that is present in the cytoplasm as well (Fig. 7A). Third, Ip/Western analysis indicated that GAK bound preferentially to B'{alpha}1 (Fig. 4), which is the nuclear form of B'{alpha}. Finally, immunostaining revealed the nuclear signals, but not the cytoplasmic signals, of B'{alpha} colocalized with GAK (Fig. 7A, see merged image and right panel). PP2A is now considered to be a tumor suppressor because inactivation of PP2A by viral oncoproteins, mutation of specific subunits, or overexpression of endogenous inhibitors contributes to cell transformation by regulating specific phosphorylation events (reviewed in Westermarck & Hahn 2008). Indeed, we previously found a truncated mutation of B'{alpha} subunit ({Delta}B'{alpha}) in highly metastatic mouse melanoma BL6 cells (Ito et al. 2000). This {Delta}B'{alpha} variant lacks N-terminal 65 amino acids due to retrotransposon insertion and its endogenous overexpression enhanced cell motility through hyper-phosphorylation of paxillin. Moreover, overexpression of {Delta}B'{alpha} also damaged the cell cycle checkpoint and enhanced the genetic instability of tumors, which promotes tumor progression from the nonmetastatic to the metastatic state (Ito et al. 2003). Thus, B'{alpha}, in collaboration with GAK, may also be involved in tumor progression.

GAK was recently reported to be an AR-interacting transcriptional coactivator (Ray et al. 2006). Given that the signal transduction pathway that leads to AR activity is also regulated by the phosphorylation and dephosphorylation of particular proteins, it is possible that GAK is also involved in this hormone signaling pathway. It should be remembered here that transcription occurs in the nucleus. Because GAK was first identified as an AR-interacting protein by using the Tup1-repressed transactivator system, it is possible that GAK will also be detected in other transcriptional complexes. This is supported by the recent report of Enari et al. (2006) mentioned above, which suggests that GAK also acts as a transcriptional coactivator in collaboration with CHC. We also showed that B'{alpha} is concentrated in intranuclear structures known as nuclear speckles (Fig. 7A) that are macromolecular structures that accumulate transcription and splicing factors in cardiomyocytes (Gigena et al. 2005). It remains to be determined by our future experiments whether any GAK-mediated events are actually involved in these transcription regulatory events.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Cell culture

The HEF line is a gift from Dr M. Yutsudo. HeLa cells were provided by the Japanese Cancer Research Resources Bank (JCRB). The other human cells that were used were purchased from the American Type Culture Collection. The cells were maintained in 5% CO2 at 35 °C in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% fetal bovine serum (Hyclone Laboratory, Logan, UT), penicillin (100 U/mL), and streptomycin (100 µg/mL).

Antibodies

Anti-GAK monoclonal antibodies were produced in our laboratory by immunizing mice with rat GST-GAK fusion protein according to standard methods. The monoclonal antibodies that recognized GST were excluded during screening. Polyclonal antibodies against rat GAK, human cyclin G1 or Orc2 were generated by immunizing New Zealand White rabbits with GST-rat GAK, GST-human cyclin G1 or GST-human Orc2 via standard techniques. Antibodies that recognized the GST portion were removed by affinity chromatography using GST-bound resin. An anti-PP2A-B'{alpha} polyclonal antibody that does not distinguish between the B'{alpha}1, -2 or -3 subtypes was prepared as reported previously (Ito et al. 2000). Antibodies against the following antigens were purchased: Myc (Oncogene Science, Uniondale, NY), actin, GFP (MBL, Nagoya, Japan), p53 (DO-1; Santa Crutz Biotechnology, Santa Cruz, CA), CHC (BD bioscience, San Jose, CA), GAPDH (Fitzgerald Industries International, Inc., Concord, MA) and FLAG (Zymed, San Francisco, CA).

Fluoroimmunostaining

Cells (synchronized or in log phase) were cultured on coverslips immersed in a culture dish ({phi} = 6 cm) and fixed by sequential 10 min treatments at 20 °C with 3.7% formamide, 0.1% Triton X-100, and 0.05% Tween 20 in PBS. After removing the medium, the cells on coverslips were rinsed for 5 min three times with 2 mL TBST buffer (20 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.05% Tween 20) supplemented with 2% bovine serum albumin (BSA; Sigma) at room temperature. Subsequently, 100 µL of TBST containing the relevant antibody was spotted on parafilm and the coverslips were lifted and laid cell-side down on the liquid. After incubation at room temperature for 1 h, the coverslips were rinsed in culture dishes cell-side up with TBST/BSA as described above. FITC- or Texas Red-linked anti-rabbit Ig (from donkey; Amersham) or Texas Red-linked anti-mouse Ig (from sheep; Amersham) served as secondary probes and were incubated with the cells attached to the coverslips for 1 h and rinsed three times as described above. MitoTrackerTM Red and ER-TrackerTM Red was purchased from Molecular Probes (Invitrogen) Inc. (Eugene, OR). Photographs were taken and the images were recorded by a inverted laser scan microscope (Zeiss LSM410 or LSM510).

Immunoprecipitation and Western blot analysis

Immunoprecipitation was performed essentially as described previously (Kanaoka et al. 1997). Briefly, cells were collected and lysed in lysis buffer (50 mM Tris-HCl [pH 7.5], 250 mM NaCl, 0.1% Nonidet P-40) supplemented with protease inhibitors (2 µg/mL aprotinin, 2 µg/mL of leupeptin, 1 µg/mL pepstatin A, 50 µg/mL PMSF, 1 mM EGTA). After clarifying the extract by centrifugation at 10 000 x g for 5 min; aliquots of the supernatant were first immunoprecipitated by using protein A-Sepharose alone. The clarified lysates were subsequently immunoprecipitated by using the relevant antibodies. Thereafter, equal quantities of fresh or immunoprecipitated cell extract were adsorbed to protein A-Sepharose, pelleted and subjected to 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE). For Western blotting, total cellular proteins and immunoprecipitates resolved on gels were transferred to nitrocellulose filters and probed with the relevant antibodies. Immunoreactive protein bands were visualized by using RenaissanceTM chemiluminescence reagents (DuPont NEN, Boston, MA).

Preparation of GFP-, GST-, FLAG- or Myc-fusion constructs

To fuse genes in-frame to GFP (green fluorescent protein), a synthetic linker was inserted into the pEGFP-C1 vector (Clontech, Palo Alto, CA); this linker was designed to allow cDNA inserts to be inserted in-frame via AscI-NotI sites. To obtain cDNA inserts for human GAK carrying in-frame AscI-NotI sites, oligonucleotides with sequences around the initiation codon and the termination codon that contained an AscI site and a NotI site, respectively, were synthesized and used as PCR primers for PCR with the relevant cDNA substrate. The identity of each gene was confirmed by DNA sequencing of four independent clones, and the plasmid DNA without a mismatched DNA sequence was selected and cut with AscI and NotI. The resulting cDNA inserts were incorporated into the GFP-vectors.

A similar strategy was used to create other plasmid constructs that express GST, FLAG or Myc-fusion proteins. The primer pairs used for PCR amplification during the process of these manipulations are listed in Supplementary Table S1. All plasmid constructs were transfected into HeLa or 293T cells by using TransITTM polyamine transfection reagents (Pan Vera Corporation, Madison, WI) according to the manufacturer's protocol. The expressed GFP-fusion proteins were visualized by using a Zeiss LSM510 confocal photomicroscope.


   Acknowledgements
 
We are obliged to Dr Takahiro Nagase (Kazusa DNA Research Institute, Japan) for the CHC plasmid (KIAA0034) and Dr M. Yutsudo (RIMD, Osaka University) for the gift of HEF cells. We thank Dr P. Hughes for critically reading the manuscript. This work was supported in part by Innovation Plaza Osaka of the Japan Science and Technology Agency (JST), and by grants-in-aid for Scientific Research on Priority Areas ‘Applied Genomics’, Scientific Research (S), Exploratory Research, and the Science and Technology Incubation Program in Advanced Regions, from the Ministry of Education, Culture, Sports, Science and Technology of Japan to Hiroshi Nojima.


   Footnotes
 
Communicated by: Fumio Hanaoka

* Correspondence: snj-0212{at}biken.osaka-u.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Bernardi, R. & Pandolfi, P.P. (2007) Structure, dynamics and functions of promyelocytic leukaemia nuclear bodies. Nat. Rev. Mol. Cell Biol. 8, 1006–1016.[CrossRef][Medline]

Biamonti, G. (2004) Nuclear stress bodies: a heterochromatin affair? Nat. Rev. Mol. Cell Biol. 5, 493–498.[CrossRef][Medline]

Chen, X. (2002) Cyclin G: a regulator of the p53-Mdm2 network. Dev. Cell 2, 518–519.[CrossRef][Medline]

Chiodi, I., Corioni, M., Giordano, M., Valgardsdottir, R., Ghigna, C., Cobianchi, F., Xu, R.M., Riva, S. & Biamonti, G. (2004) RNA recognition motif 2 directs the recruitment of SF2/ASF to nuclear stress bodies. Nucleic Acids Res. 32, 4127–4136.[Abstract/Free Full Text]

Crockett, D.K., Lin, Z., Elenitoba-Johnson, K.S. & Lim, M.S. (2004) Identification of NPM-ALK interacting proteins by tandem mass spectrometry. Oncogene 23, 2617–2629.[CrossRef][Medline]

Edeling, M.A., Smith, C. & Owen, D. (2006) Life of a clathrin coat: insights from clathrin and AP structures. Nat. Rev. Mol. Cell Biol. 7, 32–44.[Medline]

Eisenberg, E. & Greene, L.E. (2007) Multiple roles of auxilin and hsc70 in clathrin-mediated endocytosis. Traffic 8, 640–646.[CrossRef][Medline]

Enari, M., Ohmori, K., Kitabayashi, I. & Taya, Y. (2006) Requirement of clathrin heavy chain for p53-mediated transcription. Genes Dev. 20, 1087–1099.[Abstract/Free Full Text]

Gigena, M.S., Ito, A., Nojima, H. & Rogers, T.B. (2005) A B56 regulatory subunit of protein phosphatase 2A localizes to nuclear speckles in cardiomyocytes. Am. J. Physiol. Heart Circ. Physiol. 289, H285–H294.[Abstract/Free Full Text]

Greener, T., Zhao, X., Nojima, H., Eisenberg, E. & Greene, L.E. (2000) Role of cyclin G-associated kinase in uncoating clathrin-coated vesicles from non-neuronal cells. J. Biol. Chem. 275, 1365–1370.[Abstract/Free Full Text]

Ito, A., Kataoka, T.R., Watanabe, M., Nishiyama, K., Mazaki, Y., Sabe, H., Kitamura, Y. & Nojima, H. (2000) A truncated isoform of the PP2A B56 subunit promotes cell motility through paxillin phosphorylation. EMBO J. 19, 562–571.[CrossRef][Medline]

Ito, A., Koma, Y., Watabe, K., Nagano, T., Endo, Y., Nojima, H. & Kitamura, Y.A. (2003) A truncated isoform of the protein phosphatase 2A B56gamma regulatory subunit may promote genetic instability and cause tumor progression. Am. J. Pathol. 162, 81–91.[Abstract/Free Full Text]

Janssens, V., Goris, J. & Van Hoof, C. (2005) PP2A: the expected tumor suppressor. Curr. Opin. Genet. Dev. 15, 34–41.[CrossRef][Medline]

Kametaka, S., Moriyama, K., Burgos, P.V., Eisenberg, E., Greene, L.E., Mattera, R. & Bonifacino, J.S. (2007) Canonical interaction of cyclin G associated kinase with adaptor protein 1 regulates lysosomal enzyme sorting. Mol. Biol. Cell 18, 2991–3001.[Abstract/Free Full Text]

Kanaoka, Y., Kimura, S.H., Okazaki, I., Ikeda, M. & Nojima, H. (1997) GAK: a cyclin G associated kinase contains a tensin/auxilin-like domain. FEBS Lett. 402, 73–80.[CrossRef][Medline]

Kimura, S.H., Ikawa, M., Ito, A., Okabe, M. & Nojima, H. (2001) Cyclin G1 is involved in G2/M arrest in response to DNA damage and in growth control after damage recovery. Oncogene 20, 3290–3300.[CrossRef][Medline]

Kimura, S.H. & Nojima, H. (2002) Cyclin G1 associates with MDM2 and regulates accumulation and degradation of p53 protein. Genes Cells 7, 869–880.[Abstract]

Korolchuk, V.I. & Banting, G. (2002) CK2 and GAK/auxilin2 are major protein kinases in clathrin-coated vesicles. Traffic 3, 428–439.[CrossRef][Medline]

Lee, D.W., Wu, X., Eisenberg, E. & Greene, L.E. (2006) Recruitment dynamics of GAK and auxilin to clathrin-coated pits during endocytosis. J. Cell Sci. 119, 3502–3512.[Abstract/Free Full Text]

Lee, D.W., Zhao, X., Yim, Y.I., Eisenberg, E. & Greene, L.E. (2008) Essential Role of GAK (auxilin-2) in Developing and Mature Mice. Mol. Biol. Cell 19, 2766–2776.[Abstract/Free Full Text]

Lee, D.W., Zhao, X., Zhang, F., Eisenberg, E. & Greene, L.E. (2005) Depletion of GAK/auxilin 2 inhibits receptor-mediated endocytosis and recruitment of both clathrin and clathrin adaptors. J. Cell Sci. 118, 4311–4321.[Abstract/Free Full Text]

Lilyestrom, W., Klein, M.G., Zhang, R., Joachimiak, A. & Chen, X.S. (2006) Crystal structure of SV40 large T-antigen bound to p53: interplay between a viral oncoprotein and a cellular tumor suppressor. Genes Dev. 20, 2373–2382.[Abstract/Free Full Text]

Ohmori, K., Endo, Y., Yoshida, Y., Ohata, H., Taya, Y. & Enari, M. (2008) Monomeric but not trimeric clathrin heavy chain regulates p53-mediated transcription. Oncogene 27, 2215–2227.[CrossRef][Medline]

Okamoto, K. & Beach, D. (1994) Cyclin G is a transcriptional target of the p53 tumor suppressor protein. EMBO J. 13, 4816–4822.[Medline]

Okamoto, K., Kamibayashi, C., Serrano, M., Prives, C., Mumby, M.C. & Beach, D. (1996) p53-dependent association between cyclin G and the B' subunit of protein phosphatase 2A. Mol. Cell. Biol. 16, 6593–6602.[Abstract/Free Full Text]

Okamoto, K., Li, H., Jensen, M.R., Zhang, T., Taya, Y., Thorgeirsson, S.S. & Prives, C. (2002) Cyclin G recruits PP2A to dephosphorylate Mdm2. Mol. Cell 9, 761–771.[CrossRef][Medline]

Ray, M.R., Wafa, L.A., Cheng, H., Snoek, R., Fazli, L., Gleave, M. & Rennie, P.S. (2006) Cyclin G-associated kinase: A novel androgen receptor-interacting transcriptional coactivator that is overexpressed in hormone refractory prostate cancer. Int. J. Cancer 118, 1108–1119.[CrossRef][Medline]

Reimer, C.L., Borras, A.M., Kurdistani, S.K., Garreau, J.R., Chung, M., Aaronson, S.A. & Lee, S.W. (1999) Altered regulation of cyclin G in human breast cancer and its specific localization at replication foci in response to DNA damage in p53+/+ cells. J. Biol. Chem. 274, 11022–11029.[Abstract/Free Full Text]

Stanek, D. & Neugebauer, K.M. (2006) The Cajal body: a meeting place for spliceosomal snRNPs in the nuclear maze. Chromosoma 115, 343–354.[CrossRef]

Tamura, K., Kanaoka, Y., Jinno, S., Nagata, A., Ogiso, Y., Shimizu, K., Hayakawa, T., Nojima, H. & Okayama, H. (1993) Cyclin G: a new mammalian cyclin with homology to fission yeast Cig1. Oncogene 8, 2113–8.[Medline]

Westermarck, J, & Hahn. W.C. (2008) Multiple pathways regulated by the tumor suppressor PP2A in transformation. Trends Mol. Med. 14, 152–160.[CrossRef][Medline]

Zhang, C.X., Engqvist-Goldstein, A.E., Carreno, S., Owen, D.J., Smythe, E. & Drubin, D.G. (2005) Multiple roles for cyclin G-associated kinase in clathrin-mediated sorting events. Traffic 6, 1103–1113.[CrossRef][Medline]

Zhang, L., Gjoerup, O. & Roberts, T.M. (2004) The serine/threonine kinase cyclin G-associated kinase regulates epidermal growth factor receptor signaling. Proc. Natl. Acad. Sci. USA 101, 10296–10301.[Abstract/Free Full Text]

Received: 11 October 2008
Accepted: 19 February 2009




This article has been cited by other articles:


Home page
 J. Cell Sci.Home page
H. Shimizu, I. Nagamori, N. Yabuta, and H. Nojima
GAK, a regulator of clathrin-mediated membrane traffic, also controls centrosome integrity and chromosome congression
J. Cell Sci., September 1, 2009; 122(17): 3145 - 3152.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sato, J.
Right arrow Articles by Nojima, H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Sato, J.
Right arrow Articles by Nojima, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS