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


     

Genes to Cells (2009) 14, 227-241. doi:10.1111/j.1365-2443.2008.01265.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 Kawai, K.
Right arrow Articles by Yagisawa, H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kawai, K.
Right arrow Articles by Yagisawa, H.

Focal adhesion-localization of START-GAP1/DLC1 is essential for cell motility and morphology

 Katsuhisa Kawai1, Yui Iwamae1, Masaki Yamaga1a, Minoru Kiyota1, Hiroko Ishii1, Hajime Hirata1, Yoshimi Homma2 and Hitoshi Yagisawa1,*

1 Graduate School of Life Science, University of Hyogo, Harima Science Garden City, Hyogo-ken 978-1297, Japan
2 Department of Biomolecular Sciences, Institute of Biomedical Sciences, Fukushima Medical University, Hikariga-oka, Fukushima 960-1295, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
There is a class of GTPase activating proteins for the Rho family GTPases (RhoGAPs) that contain the steroidogenic acute regulatory protein (STAR)-related lipid transfer (START) domain. In mammals three genes encode such proteins and they are designated START-GAP1–3 or deleted in liver cancer 1–3 (DLC1–3). In this study, we examined the intracellular localization and roles of START-GAP1/DLC1 in cell motility. Immunofluorescence microscopic analysis of NRK cells and HeLa cells revealed that START-GAP1 was localized in focal adhesions. Amino acid residues 265–459 of START-GAP1 were found to be necessary for focal adhesion targeting and we name the region "the focal adhesion-targeting (FAT) domain." It was previously known that ectopic expression of START-GAP1 induced cell rounding. We demonstrated that the FAT domain of START-GAP1 was partially required for this morphological change. Furthermore, expression of this domain in HeLa cells resulted in dissociation of endogenous START-GAP1 from focal adhesions as a dominant negative modulator, reducing cell migration and spreading. Taken together, START-GAP1 is targeted to focal adhesions via the FAT domain and regulates actin rearrangement through down-regulation of active RhoA and Cdc42. Its absence from focal adhesions could, therefore, cause abnormal cell motility and spreading.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Cell motility is crucial for various cellular events including early development, wound healing, immune response and tumorigenesis. Many factors including adhesion molecules, receptors, small GTPases, MAP kinases, cytoskeleton regulatory proteins have been implicated in cell motility (Huttenlocher et al. 1995; Fukata et al. 2003; Huang et al. 2004). Among them it is well-known that the regulation of actin cytoskeleton via Rho family GTPases is mainly responsible for cell motility (Etienne-Manneville & Hall 2002). Rho family GTPases serve as molecular switches in various cellular functions, including cell motility, intracellular transport, and gene transcription leading to proliferation, differentiation or cell death. At present, the Rho family GTPases consist of more than 20 members (Wherlock & Mellor 2002). Among these, RhoA, Rac1 and Cdc42, are the most thoroughly characterized. These proteins regulate cell motility through actin rearrangement and cell polarity (Etienne-Manneville & Hall 2002). Rac1 and Cdc42 are involved in the formation of lamellipodia and filopodia, respectively (Ridley et al. 1992). Active RhoA induces the formation of actin stress fibers and focal adhesions (Ridley & Hall 1992). As cell movement involves formation and disruption of these structures, the Rho family GTPases should be under strict spatio-temporal regulations.

Two factors are mainly involved in these regulations. Guanine nucleotide exchange factors replace GDP with GTP to activate Rho family GTPases (Nobes & Hall 1994). The Rho family GTPases then return to the inactive state upon hydrolysis of GTP to GDP by intrinsic GTPase activity, a process facilitated by GTPase-activating proteins (GAPs) (Bourne et al. 1990). Although many molecules have been identified as Guanine nucleotide exchange factors or GAPs, it is not well-known where, when or how these regulatory molecules act on Rho family GTPases in cellular signaling. GAPs for Rho family GTPases (RhoGAPs) inactivate regulators of the actin cytoskeleton. RhoGAPs are the largest known group of regulators of Rho family GTPases and significantly outnumber RhoGTPases. They have been implicated in many aspects of tumorigenesis (Sahai & Marshall 2002; Moon & Zheng 2003).

Steroidogenic acute regulatory protein (STAR)-related lipid transfer (START)-GAP1 was originally cloned from rat cDNA library (Homma & Emori 1995). START-GAP1 enhances the activity of phospholipase C-{delta}1 (PLC{delta}1), which generates two second-messengers containing inositol 1,4,5-trisphospate (IP3) and diacylglycerol (DAG) via hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP2). IP3 is accepted by IP3 receptor on ER, resulting in the elevation of intracellular calcium concentration, whereas DAG acts as the activator for PKC (Nishizuka 1995; Rebecchi & Pentyala 2000). Indeed, microinjection of START-GAP1 into the cytosol elevated intracellular calcium concentration (Sekimata et al. 1999). START-GAP1 was also demonstrated to acts as a GAP for RhoA and over-expression of START-GAP1 induces morphological change along with disruption of actin stress fibers (Sekimata et al. 1999). START-GAP was therefore the originally designated adaptor for Rho and PLC (ARP), but later the name p122RhoGAP (or just simply p122) was used to avoid confusion with another ARP (actin related proteins) (Sekimata et al. 1999; Kawai et al. 2004; Yamaga et al. 2004). Moreover, START-GAP1 has also been known as a tumor suppressor gene product. It is frequently under-expressed or not expressed in several tumor cells and inhibits cell growth, invasion and metastasis (Yuan et al. 1998, 2003; Zhou et al. 2004). These phenotypes require the GAP activity of START-GAP1 (Zhou et al. 2004). Because the START-GAP1 gene was originally found deleted in human liver cancer cells, it was also designated deleted in liver cancer-1 (DLC1). START-GAP1 (DLC1) consists of, from N-terminus to C-terminus, the sterile {alpha} motif (SAM), GAP domain and START domain, which is known as a lipid binding or transfer domain (Alpy & Tomasetto 2005).

Although substantial work has been done to dissect the physiological role of START-GAP1, its localization and function are still not clear. In this study, we examined the subcellular distribution of START-GAP1 using an anti-START-GAP1 antibody and a series of deletion and point mutants of GFP-fused START-GAP1 and characterized the effects of domains on the cell morphology and motility. We found that START-GAP1 is localized in focal adhesions through a region covering amino acid residues 265–459. Expression of this region in cells affects cell attachment and cell motility, suggesting that START-GAP1 regulates cell motility at the site of cell contact, at least partially, by controlling the activation and inactivation cycle of Rho GTPases that control the actin organization.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
START-GAP1 is localized in focal adhesions

In the previous published article (Kawai et al. 2004), we reported that endogenous START-GAP1 is localized in focal adhesions in NRK cells (Fig. 1A, a–d). To examine whether START-GAP1 is also localized in focal adhesions in other cell lines, we performed immunofluorescence microscopy using an anti-START-GAP1 antibody in HeLa cells. Endogenous START-GAP1 was distributed in punctated structures mainly at the periphery of spreading cells (Fig. 1B, e). F-actin staining (Fig. 1B, f) revealed that START-GAP1 was localized at the tips of actin stress fibers (Fig. 1B, g,h). The distribution of the dot-like structure also coincided with that of vinculin by immunofluorescence detection (Fig. 1C,D). It was therefore suggested that START-GAP1 was also localized in focal adhesions in HeLa cells. Similar localization of endogenous START-GAP1 was observed in PC12 and HepG2 cells (data not shown).


Figure 1
Figure 1
View larger version (132K):
[in this window]
[in a new window]

  		Figure 1  Localization of endogenous START-GAP1. (A) Endogenous START-GAP1 in NRK cells was stained with a rabbit polyclonal anti-START-GAP1 antibody (a). F-actin was stained with Texas-Red-X Phalloidin (b). The merged image is shown in the right panel (c) and the enlarged merged image (threefold magnification of the original) is shown underneath (d). (B) Localization of START-GAP1 in HeLa cells was detected using a mouse monoclonal anti-START-GAP1 antibody (e). F-actin was stained with Texas-Red-X Phalloidin (f). The merged image is shown in the right panel (g) and the enlarged merged image (threefold magnification of the original) is shown underneath (h). (C) Endogenous START-GAP1 and vinculin were stained with a rabbit polyclonal anti-START-GAP1 antibody (i) and a mouse monoclonal anti-vinculin antibody (j), respectively. The merged image is shown in the right panel (k). In the lower panels threefold magnified images are shown (l, m, n). (D) Endogenous START-GAP1 was stained with a mouse monoclonal anti-START-GAP1 antibody (o) Vinculin was stained with an Alexa594-labeled mouse monoclonal anti-vinculin antibody using Zenon mouse IgG labeling kit (Molecular Probes) (p). The merged image is shown in the right panel (q). Cells were visualized with a confocal microscope.

 
Region responsible for the focal adhesion localization

To determine the region essential for the focal adhesion targeting of START-GAP1, we constructed various deletion and truncated mutants of START-GAP1 fused with GFP (Fig. 2A), and observed subcellular distribution of GFP fluorescence in NRK or HeLa cells using confocal microscopy. When over-expressed, GFP alone showed the diffused cytoplasmic and nuclear distribution in both cells (not shown). First, the contribution of known protein domains was examined. With regard to the GAP domain, two mutants, GFP-START-GAP1(R669E), which has a mutation in the arginine loop of the GAP domain so that both the GAP activity and the RhoA-binding ability are lost, and GFP-START-GAP1({Delta}GAP), which lacks the entire GAP domain, were expressed in NRK cells. These two mutants were found in punctated structures similar to those of endogenous START1 (Fig. 2B, a,b), indicating that neither the GAP function nor the overall structure of the domain is necessary for the focal adhesion targeting of START-GAP1. We then investigated whether the START domain contributes to the focal adhesion localization using a mutant GFP-START-GAP1(1–533). This mutant was also found localized in focal adhesions (Fig. 2B, c). To examine whether the SAM domain is responsible for the focal adhesion-targeting of START-GAP1, GFP-START-GAP1(1–117), which covers the amino-terminal to the SAM domain, was expressed in NRK cells. Diffuse fluorescence was observed in the cytoplasm and the nucleus (Fig. 2B, d), suggesting that the SAM domain is not sufficient for the targeting of START-GAP1 and that the rest of the amino-terminal half, a region in 117–533, is likely to be responsible.


Figure 2
Figure 2
View larger version (49K):
[in this window]
[in a new window]

  		Figure 2  Intracellular distribution of point and truncated mutants of START-GAP1. (A) Schematic representation of GFP-fused START-GAP1 mutants. START-GAP1 contains sterile {alpha} motif (SAM), GTPase-activating protein (GAP) domain and steroidogenic acute regulatory protein-related lipid transfer (START) domains. The numbers above each construct refer to the amino acid numbers. (B) Confocal laser microscopic images of GFP-START-GAP1(R669E) (a), GFP-START-GAP1 ({Delta}GAP) (b), GFP-START-GAP1(1–533) (c) or START-GAP1(1–117) (d) transiently expressed in NRK cells. (C) Intracellular distribution of transiently-expressed mutants of START-GAP1 in HeLa cells. Fluorescence images of GFP-START-GAP1(1–533) (a), START-GAP1(117–533) (b), GFP-START-GAP1(1–459) (c), GFP-START-GAP1(265–533) (d), GFP-START-GAP1 (163–467) (e), GFP-START-GAP1(10–396) (f), GFP-START-GAP1(117–459) (g), GFP-START-GAP1(265–459) (h), GFP-START-GAP1(1–533({Delta}163–467)) (i) and GFP-START-GAP1(396–459) (j) were taken using a confocal laser microscopy system.

 
We then narrowed down the sequence essential for the localization in focal adhesions. Plasmids encoding GFP-fused START-GAP1 mutants: GFP-START-GAP1(1–459), GFP-START-GAP1(117–459), GFP-START-GAP1 (163–467), GFP-START-GAP1(265–533), GFP-START-GAP1(265–459), GFP-START-GAP1(10–396), GFP-START-GAP1(396–459) and GFP-START-GAP1(1–533, {Delta}163–467) were constructed and each of them was expressed in HeLa cells. GFP-START-GAP1(1–459), GFP-START-GAP1(117–459), GFP-START-GAP1(163–467), GFP-START-GAP1(265–533) and GFP-START-GAP1(265–459) were localized in focal adhesions, while GFP-START-GAP1(10–396), GFP-START-GAP1(396–459) and GFP-START-GAP1(1–533, {Delta}163–467) were not (Fig. 2C, a–j). These results, taken together, indicate that the region corresponding to amino acid residues 265–459 of START-GAP1 contains a minimal signal sequence for focal adhesion-targeting (FAT) and the region could be designated the FAT domain.

The distribution of GFP-START-GAP1(265–459) was compared with that of F-actin. Similar to the distribution of endogenous START-GAP1, the GFP signal was found at the tips of actin fibers (Fig. 3, a–c), suggesting that the FAT domain was targeted to focal adhesions.


Figure 3
View larger version (32K):
[in this window]
[in a new window]

  		Figure 3  Localization of the GFP-START-GAP1 FATdomain and F-actin. GFP-START-GAP1(265–459) was transiently expressed in HeLa cells. After fixation, cells were stained with Texas-Red-X Phalloidin. Images were taken using a confocal laser microscopy system. The merged image is shown in the right panel.

 
Cell rounding caused by ectopic expression of START-GAP1 is dependent on both GAP and FAT domains

It was previously reported that ectopic expression of START-GAP1 induces a morphological change (cell rounding) (Sekimata et al. 1999). Indeed, HeLa cells expressing GFP-fused START-GAP1 induced cell rounding and theses cells were bared with long protrusions (Fig. 4A, a–c). Nevertheless, HeLa cells expressing the R669E mutant did not show such a change (Fig. 4A, d–f). Because in vitro GAP assay demonstrated that the R669E mutant completely lost its GAP activity (Fig. 4B), it was concluded that the morphological changes induced by ectopic expression of START-GAP1 was dependent on its GAP activity. To assess whether the GAP activity alone is sufficient for the morphological change, we expressed GFP-fused GAP domain in HeLa cells (Fig. 4A, g–i). Although isolated GAP domain can exert the GAP activity for RhoA and Cdc42 almost as strong as full-length START-GAP does in vitro (Fig. 4B), expression of this domain alone were not able to induce cell rounding (Fig. 4A, g–i). This result indicates that the GAP activity alone is insufficient to induce the morphological changes.


Figure 4
Figure 4
View larger version (76K):
[in this window]
[in a new window]

  		Figure 4  Effect of expression of START-GAP1 on cell morphology. (A) HeLa cells were transiently transfected with plasmids encoding GFP-START-GAP1(full-length) (a–c), GFP-START-GAP1(R669E) (d–f), or GFP-START-GAP1(GAP) (g–i). After 24 h, formaldehyde-fixed cells were stained with Texas-Red-X Phalloidin and observed under fluorescence microscopy. Fluorescence images of GFP are shown in (a), (d) and (g). Those of Texas-Red-X Phalloidin are in (b), (e) and (h). Merged images are shown in (c), (f) and (i). (B) In vitro GAP assay for Rho proteins. [{gamma}32P]GTP bound to RhoA, Rac1, or Cdc42 was incubated with recombinant GST, GST-START-GAP(full-length), GST-START-GAP1(R669E), or GST-START-GAP1(GAP). After incubation (0, 2, 5 min), the hydrolysis of [{gamma}32P]GTP bound to Cdc42, Rac1, or RhoA was assayed by measuring the radioactivity of [{gamma}32P]GTP bound to each small G protein using a nitrocellulose filtration method. The results shown are representative of three independent experiments. (C) Schematic representation of GFP-fused START-GAP1 mutant. The numbers above each construct refer to the amino acid numbers. (D) HeLa cells were grown on coverslips were transfected with each GFP-fused START-GAP1 mutants. Cells were visualized with a confocal microscope. The frequency of cell rounding is shown with the number of round cells by the number of the GFP-positive cells. A value is the average of more than five independent experiments. About 50 GFP-positive cells were counted for each experiment. The asterisk (* or **) indicates that the difference was considered significant (P < 0.001 or P < 0.0001) by Student's t-test, respectively.

 
We then explored the inter-relationship among the domains in START-GAP1 on the morphological changes. As far as the GFP florescent intensity was concerned, expression levels of constructs transfected in HeLa cells were almost equal. Because expression of GFP-START-GAP(1–862) in HeLa cells also induced cell rounding (Fig. 4D), it was considered that N-terminal region of START-GAP1 is necessary for the morphological changes in addition to the GAP activity. We considered that the presence of the FAT domain might have some influence on cell morphology. It is possible to assume that the GAP activity is responsible for the morphological change only when the domain is recruited to focal adhesions. To assess this idea, we constructed two mutants, GFP-START-GAP1(1–862{Delta}FAT), which is a FAT domain deleted version of GFP-START-GAP1(1–862), and GFP-START-GAP1(GAP-FAT), which is the GAP domain fused with the FAT domain (Fig. 4C). As shown in Fig. 4D, expression of GAP-FAT increased the number of HeLa cells with rounded shapes compared with cells expressing the GAP domain alone (Fig. 4D, right two columns). Conversely, cells expressing 1–862-{Delta}FAT contain less rounded cells compared with cells expressing full-length GFP-START-GAP1 or 1–862 (Fig. 4D, middle two columns). These results strongly suggest that recruitment of the active GAP domain to focal adhesions is essential for cell rounding.

Effects of expression of the FAT domain on the function of endogenous START-GAP1

We then explored if expression of the FAT domain affected the localization of endogenous START-GAP1. HeLa cells transiently expressing the FAT domain were observed by immunofluorescence analysis using an anti-START-GAP1 antibody. As described above, endogenous START-GAP1 is localized in focal adhesions in control unexpressed HeLa cells, whereas in cells expressing the GFP-FAT domain, endogenous START-GAP1 was no longer localized in focal adhesions and showed a diffused localization pattern (Fig. 5A). This result suggests that the ectopically expressed FAT domain competitively binds to target molecules of endogenous START-GAP1 in focal adhesion complexes and induces dissociation of endogenous START-GAP1 from focal adhesions. The FAT domain, therefore, can be used as a dominant negative regulator of START-GAP1 in focal adhesions.


Figure 5
Figure 5
View larger version (80K):
[in this window]
[in a new window]

  		Figure 5  Effects of ectopic expression of the FAT region. (A) The effect of transient expression of GFP-START-GAP1(265–459) on the distribution of endogenous START-GAP1 in HeLa cells was examined. Fluorescence images of GFP (a) and endogenous START-GAP1 (b) were taken using a confocal laser microscopy system. Endogenous START-GAP1 was detected by immunostaining with an anti-START-GAP1 antibody. The merged image is shown in (c). START-GAP1 are diffusely distributed in FAT-expressed cells (closed triangles), while it is localized as focal adhesion like pattern in unexpressed cells (open triangles). (B) The effect of ectopic expression of the START-GAP1 FAT domain on cell motility. HeLa cells transiently expressing GFP only (right panel) or GFP-START-GAP1(265–459) (left panel) were used in a wound healing assay. The black outline shows the cell edge (made by a scratch) at the beginning of the experiment. The gray outline shows the cell edge after 10 h. The data are from three independent experiments with 2–4 dishes. (C) The effect of ectopic expression of the FAT domain on cell spreading. HeLa cells transiently expressing GFP or GFP-START-GAP1(FAT) were used in a spreading assay. Harvested cells were replated onto a fibronectin-coated coverslips, fixed after 60 min of incubation at 37 °C, and stained with Texas-Red-X Phalloidin. Cell images were taken using a confocal laser microscope. Upper panel showed as control GFP and lower panel showed as GFP-START-GAP1(265–459). The maximum cell area was estimated by NIH Image (D). The result of a typical experiment is shown as the mean ± SD. The asterisk indicates that the difference was considered significant (n = 38, P < 0.0001) by Student's t-test. The experiment was repeated three times and gave similar results. About 30 cells were counted for each experiment.

 
One important biological function of START-GAP1 is the ability to negatively regulate Rho GTPases via its catalytic RhoGAP domain. During cell migration and/or cell spreading, the activities of Rho GTPases are differently regulated in the leading edge, where the RhoA activity is down-regulated while the Rac1 activity is up-regulated (Ridley et al. 2003). In focal adhesions endogenous START-GAP1 may regulate actin rearrangements mediated by Rho GTPases and could therefore be involved in cell migration and/or cell spreading. Inhibition of endogenous START-GAP1 may therefore modulate cell motility or cell spreading. Indeed, HeLa cells expressing GFP-START-GAP1(265–459) showed a slower recovery from the wound compared to control cells expressing GFP only: the mean ± SD of % recovery was 45 ± 9 from three independent experiments. (Fig. 4B, a), compared with cells expressing GFP alone (Fig. 4B, b) that migrate as fast as untransfected cells (not shown). During migration of untransfected cells, endogenous START-GAP1 as well as vinculin was enriched at the leading edge of moving cells (see Figure S1 in Supporting Information). Moreover, we assessed the effect of the FAT domain on cell spreading (Fig. 5C,D). Initial rates of cell adherence after re-plating were much slower (about 50% of control) in cells expressing GFP-START-GAP1(265–459) compared with cells expressing GFP alone. These two results, taken together, suggest that sequestration of START-GAP1-binding molecule(s) by the FAT domain causes dissociation of endogenous START-GAP1 from focal adhesions, blocking the turnover of focal adhesions required for normal cell movements via the interaction between the basal membranes and the matrix. Modulation of the RhoA or Cdc42 activity by START-GAP1 may therefore play a key role in cell motility through the regulation of assembly and disassembly of focal adhesion complexes.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
Using a newly raised anti-START-GAP1 polyclonal antibody, we found START-GAP1 is a focal adhesion-localized protein in several cell lines. Focal adhesions are plasma membrane sites where integrins accumulate and play an important role for maintenance of cell morphology (Critchley 2000). In focal adhesions integrins link to actin fibers via several adaptor proteins for example talin and tensin (Sechi & Wehland 2000). Rho family GTPases regulate these structures mediated by activation of its effectors. We show that START-GAP1 is a negative regulator of RhoA and Cdc42 at least in vitro (Fig. 4B). It is therefore very intriguing that such a negative regulator is localized in focal adhesions where actin remodeling extensively takes place with the aid of Rho family GTPases. START-GAP1 could regulate cell morphology via inhibition of RhoA and Cdc42 in focal adhesions.

As shown in Fig. 2, the region responsible for focal adhesion-localization of START-GAP1 was narrowed down to amino acid residues 265–459 between SAM and GAP domains. We named the region the FAT domain. We have previously reported that ectopic expression of the N-terminal half of START-GAP1 was diffusely distributed in BHK cells (Yamaga et al. 2004). Although the subcellular distribution of focal adhesions is similar in both BHK and NRK cells, the difference in subcellular localization of the N-terminal START-GAP1 in this study may come from the difference in cell types. It is possible that START-GAP1 cannot target focal adhesions in BHK cells in a similar mechanism that is present in NRK cells. Presumably, cellular components necessary for focal adhesion-localization of START-GAP1 are lacking in BHK cells. Further analyses of the difference in START-GAP1 localization between these two cell lines could bring about insights into the targeting mechanisms of START-GAP1 subcellular localization.

Sekimata et al. reported that over-expression of full length START-GAP1 induces cell rounding (Sekimata et al. 1999). We confirmed this observation in HeLa cells (Fig. 4A). This morphological change by START-GAP1 is dependent on its GAP activity (Sekimata et al. 1999). We demonstrated in this study that recruitment of START-GAP1 to focal adhesions in addition to the presence of active GAP activity is necessary for this morphological change (Fig. 4C,D). Furthermore, as indicated in Fig. 5, dispersion of endogenous START-GAP1 from focal adhesions by expressing the isolated FAT domain, reduced cell motility and spreading. These results suggest that not only the cytoplasmic localization but also the targeting to focal adhesions is essential to exert proper function of START-GAP1.

Generally, Rho GTPases are involved in the maintenance of focal adhesion (Ridley & Hall 1992). Especially RhoA regulates formation of actin stress fibers and focal adhesions. It is thereby considered that START-GAP1 negatively regulates formation of focal adhesions through down-regulation of RhoA on focal adhesions via GAP domain. Nevertheless, the level of cell rounding caused by expression of the chimera mutant, GAP-FAT, was less than that caused by expression of full-length START-GAP1 or 1–862 (Fig. 4C,D), indicating that some other factors in addition to the GAP and FAT domains are necessary for induction of full cell rounding. The START domain may not participate in the morphological changes, since cell rounding induced by 1–862 expression was about the same level as that induced by expression of full-length START-GAP1. The mutant 1–862{Delta}FAT, which is not localized in focal adhesions, caused more severe morphological changes than GAP-FAT, suggesting that the N-terminal region franking the FAT domain contains region(s) stimulating the GAP activity to disturb the maintenance of normal cell morphology.

Recently, three groups have reported that START-GAP1 interacts with tensin family (Yam et al. 2006; Liao et al. 2007; Qian et al. 2007). Tensin family consists of four members including tensin1, tensin2, tensin3 and cten (Lo 2004). All of these are reported to be in focal adhesions. Tensin family proteins function as adaptor proteins via two protein-interacting domains, a Src homology 2 (SH2) domain and a phosphotyrosine-binding (PTB) domain (Lo 2004). Interestingly, the FAT domain of START-GAP1 contains a cten-binding site, suggesting that interaction with tensin family would at least partly involved in localization of START-GAP1 in focal adhesions. Liao et al. reported that the latter half of the FAT domain was sufficient for interaction with cten (Liao et al. 2007). In our hands, however, the latter half of the FAT domain was insufficient for targeting to focal adhesions (Fig. 2C, j), suggesting that focal adhesion-localization of START-GAP1 is not completely determined by interaction with tensin family proteins but also by other interactions mediated through the former half of the FAT domain. The FAT domain actually contains no canonical functional or structural protein domains. However, it is noteworthy that the N-terminal half of the domain is rich in serine residues. Among them Ser322 is phosphorylated by PKB/Akt upon stimulation with insulin in primary adipocytes and CHO cells (Hers et al. 2006), although physiological meaning of this phosphorylation remains elusive.

START-GAP1 is a multidomain protein containing SAM, RhoGAP domain and START domain. In mammals there are two isoforms of START-GAP1, namely START-GAP2 (DLC2) and START-GAP3 (DLC3), with the same domain structure. We classified these proteins as the START-GAP family (Kawai et al. 2007). It was reported that all START-GAPs possess tumor suppression activity and expression of these proteins are impaired in several tumor cells by gene deletion or inhibition of the promoter activity (Ching et al. 2003; Wong et al. 2003; Yuan et al. 2004; Durkin et al. 2007). Liao et al. claimed that at least the RhoGAP activity and interaction with the tensin family are involved in these growth suppression (Liao et al. 2007). Roles of the SAM domain and the START domain are not well-known.

Rho GTPases are important in regulation of focal adhesions (Ridley & Hall 1992). There are more than 50 RhoGAPs in mammals and they act as negative regulators of Rho GTPases (Moon & Zheng 2003). In addition to START-GAP1, several RhoGAPs such as RC-GAP and GRAF are localized in focal adhesions (Hildebrand et al. 1996; Lavelin & Geiger 2005). The differences among these focal adhesion-localized RhoGAPs are not clear. One possibility is the difference in the substrate specificity. RC-GAP acts as a RhoGAP of Rac1 and Cdc42, but not of RhoA (Lavelin & Geiger 2005), whereas START-GAP1 acts on RhoA and Cdc42, but not on Rac1 as shown in this study. GRAF has substrate specificity similar to START-GAP1 (Hildebrand et al. 1996). These facts suggest that each focal adhesion-localized RhoGAP possesses a distinct function besides the substrate specificity. It is conceivable that upstream regulation may be different among these RhoGAPs, since binding partners of these RhoGAPs are reported to be different. START-GAP1 binds to tensin, while GRAF binds to FAK, a tyrosine kinase localized in focal adhesions (Hildebrand et al. 1996). Analyses of regulation of RhoGAPs including START-GAPs, GRAF and RC-GAP would lead us to understand regulation mechanisms of focal adhesion turnover by Rho GTPases.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Materials and chemicals

The following were obtained commercially: a mouse monoclonal anti-vinculin from Sigma (St. Louis, MO), anti-DLC1 (START-GAP1) antibody from Transduction Laboratories (Lexington, KY), Cy3-conjugated anti-mouse IgG from Kirkeggad & Perry Laboratories (Gaithersburg, MD), FITC-conjugated anti-rabbit IgG, pGEX vector, glutathione Sepharose 4B and cyanogen bromide-activated Sepharose 4B from GE Life Science (Piscataway, NJ), Texas-Red-X Phalloidin from Molecular Probes (Eugene, OR), Lipofectin and OPTIMEM from Invitrogen (Carlsbad, CA), pEGFP vector from Clontech (Palo Alto, CA). [{gamma}32P]-GTP from Perkin Elmer (Salem, MA).

Cell culture and transfection

NRK cells or HeLa cells were grown in Dulbecco's modified Eagle's medium (DMEM) containing 5% or 10% fetal bovine serum (FBS), respectively. NRK cells were transfected by plasmid microinjection. NRK cells were seeded onto a 13-mm glass coverslip coated with fibronectin (50 µg/mL). At 24 h after seeding, plasmid DNA (100 ng/µL) was microinjected (injection pressure: 120 hPa, compensation pressure: 30 hPa, injection duration: 0.3 s) into the cell nuclei using sterile Femtotips (Eppendorf, Hamburg, Germany) held in a Micromanipulator 5171 (Eppendorf, Hamburg, Germany) with a pressure supply from a Trancejector 5246 (Eppendorf). Two hours after microinjection, the cells were fixed. HeLa cells were transfected using a Lipofection method in OPTIMEM. Both NRK and HeLa cells were fixed with 3% formaldehyde in phosphate buffered saline (PBS) for 10 min at room temperature.

Plasmid construction

Mammalian expression plasmids for GFP fusion proteins, pEGFP-START-GAP1(wt), pEGFP-START-GAP1(R669E), pEGFP-START-GAP1(1–862), pEGFP-START-GAP1(1–533) and pEGFP-START-GAP1(1–117) were produced as described previously (Sekimata et al. 1999). pEGFP-START-GAP1({Delta}GAP) was constructed by removing the EcoRV/EcoRI fragment from pEGFP-START-GAP1(wt). pEGFP-START-GAP1(1–533({Delta}163–467)) and pEGFP-START-GAP1(1–862{Delta}FAT) were produced by removing the PmacI fragment from pEGFP-START-GAP1(1–533) and pEGFP-C1-START-GAP1(1–862), respectively. pEGFP-START-GAP1(117–533), pEGFP-START-GAP1(117–459), pEGFP-START-GAP1(10–396), pEGFP-START-GAP1(396–459), pEGFP-START-GAP1(265–533), pEGFP-START-GAP1(265–459), pEGFP-START-GAP1(1–459) and pEGFP-START-GAP1 (163–467) were generated from pEGFP-START-GAP1(1–533) by inserting the BamHI, BamHI/SacI, HindIII, HindIII/SacI PvuII/BamHI, PvuII/SacI, XhoI/SacI and PmacI fragments into the pEGFP vector, respectively. pEGFP-START-GAP1(GAP) was constructed from pEGFP-START-GAP1(1–862) and pEGFP-START-GAP1 (618–1083) (Sekimata et al. 1999). pEGFP-START-GAP1(GAP-FAT) was constructed from pEGFP-START-GAP1(117–533) and pEGFP-START-GAP1(GAP). pGEX4T-1-RhoA, pGEX4T-1-Rac1 and pGEX4T-1-Cdc42 were constructed by insertion of the BamHI fragment obtained from pEF-BOS-Myc-RhoA, pEF-BOS-Myc-Rac1 and pEF-BOS-Myc-Cdc42, respectively (kind gifts from Dr Yoshimi Takai) (Yamada et al. 2005), into the pGEX4T-1 vector.

Protein purification

All proteins were used as GST fusion proteins. pGEX plasmids were transformed in E. coli BL21 strain and GST-fusion protein were produced and purified with standard procedures.

Production of antibody

A rabbit polyclonal anti-rat START-GAP1 antibody was raised against a GST fusion protein of the amino-terminal half of rat START-GAP1, GST-START-GAP1(1–533). Immunized serum was first passed through a column of GST-bound Sepharose 4B to remove anti-GST antibodies. The anti-p122 antibody was then affinity-purified on a column of the immunogen covalently bound to cyanogen bromide-activated Sepharose 4B using an elution buffer (0.1 M glycine/HCl, pH 3.5) and stored in Tris/HCl (pH 8.0) after neutralization. The antibody was suitable for immunoblotting and immunocytochemistry.

Immunofluorescence analysis

NRK or HeLa cells were seeded onto a glass coverslip and transfected with plasmids encoding START-GAP1-mutants as described above. Cells were then fixed with 3% formaldehyde, and permeabilized with a permeabilizing and blocking buffer (0.1% Triton X-100, 2% FBS in PBS) for 10 min at room temperature. The permeabilized cells were incubated with a mouse monoclonal anti-vinculin or the rabbit polyclonal anti-START-GAP1, followed by Cy3-conjugated anti-mouse IgG or FITC-conjugated anti-rabbit IgG, respectively. Actin filaments were stained with Texas-Red-X Phalloidin for 20 min at room temperature. In the case of double-labeled HeLa cells with START-GAP1 and vinculin, Zenon mouse IgG labeling kit (Molecular Probes) was used for vinculin staining. Fluorescence images were taken using a confocal laser microscopy system (Carl Zeiss LSM 510) built around a Zeiss Axioplan 2 microscope (Carl Zeiss, Oberkochen, Germany).

In vitro GAP assay

GAP assay was performed as described (Fukui et al. 1997). Rho proteins were purified using the bacterial expression system for the GST fusion protein. RhoA, Rac1, or Cdc42 (3 pmol) was incubated at 30 °C for 10 min in a reaction mixture containing 25 mM Tris/HCl (pH 8.0), 10 mM EDTA, 5 mM MgCl2, 0.5 mM dithiothreitol, 0.3% CHAPS, 10.5 nM [{gamma}-32P]GTP and 1 µM GTP. The reaction was terminated by adjusting the MgCl2 concentration to 20 mM. To this mixture, 0.1 pmol of GST-START-GAP1, GST-START-GAP1 (R669E), GST-START-GAP1(GAP) or GST alone in 25 mM Tris/HCl (pH 8.0) was added to a total volume of 50 µL and further incubated at 30 °C for the indicated periods of time. The mixture was recovered on a nitrocellulose filter, and the radioactivity retained on the filter was measured by Cerenkov counting using a liquid scintillation counter.

Cell rounding assay

HeLa cells were seeded onto coverslips in 24-well plate. After 24 h, the cells were transfected as described above with GFP-fused START-GAP1 mutants. After fixation with 3% formaldehyde, cells were observed by confocal microscopy.

Wound-healing assay

HeLa cells transiently expressing GFP or GFP-START-GAP1(265–459) were seeded at a concentration of 1 x 105 cells per 0.5 mL per well in a 24-well culture plate. At 24 h after seeding, a portion of the cell layer was partially scratched with a rubber policeman and the edges of the scratched area were recorded. The cells were incubated for 10 h before they were fixed and cell invasion into the scratched area was assessed by phase-contrast microscopy (Carl Zeiss Axiovert 135, Oberkochen, Germany).

Cell-spreading assay

Subconfluent HeLa cells were transfected overnight with pEGFP or pEGFP-START-GAP1(265–459). Cells were then harvested with 0.5 mM EDTA in PBS, resuspended in DMEM containing 10% FBS, and counted. Cells were replated onto fibronectin-coated glass coverslips placed in a 24-well culture plate at a concentration of 2.5 x 104 cells per 0.5 mL per well. After 60 min at 37 ºC, cells were fixed with 3% formaldehyde, permeabilized and stained for F-actin with Texas-Red-X Phalloidin. Cell images were taken using a confocal laser microscope. The cell area was estimated using NIH Image (http://rsb.info.nih.gov/nih-image/).


   Acknowledgements
 
This work was supported by grants from Ministry of Education, Culture, Sports, Science and Technology of Japan (#13033035), the Japan Society of Promotion of Science (Grant-in-Aid for Scientific Research: #19570184), Hyogo Science and Technology Association, Kawanishi Memorial-Shinmeiwa Foundation, and a Special Research Grant of University of Hyogo to HY. The Graduate Fellowship of the Japan Society of Promotion of Science to MY and the Sasakawa Fellowship of the Nippon Foundation to KK were also appreciated.


   Footnotes
 
Communicated by: Eisuke Nishida

aPresent address: Department of Biochemistry, University of Wisconsin–Madison, 433 Babcock Dr, Madison, WI 53706, USA Back

* Correspondence: yagisawa{at}sci.u-hyogo.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Alpy, F. & Tomasetto, C. (2005) Give lipids a START: the StAR-related lipid transfer (START) domain in mammals. J. Cell Sci. 118, 2791–2801.[Abstract/Free Full Text]

Bourne, H.R., Sanders, D.A. & McCormick, F. (1990) The GTPase superfamily: a conserved switch for diverse cell functions. Nature 348, 125–132.[CrossRef][Medline]

Ching, Y.P., Wong, C.M., Chan, S.F., Leung, T.H., Ng, D.C., Jin, D.Y. & Ng, I.O. (2003) Deleted in liver cancer (DLC) 2 encodes a RhoGAP protein with growth suppressor function and is underexpressed in hepatocellular carcinoma. J. Biol. Chem. 278, 10824–10830.[Abstract/Free Full Text]

Critchley, D.R. (2000) Focal adhesions—the cytoskeletal connection. Curr. Opin. Cell Biol. 12, 133–139.[CrossRef][Medline]

Durkin, M.E., Ullmannova, V., Guan, M. & Popescu, N.C. (2007) Deleted in liver cancer 3 (DLC-3), a novel Rho GTPase-activating protein, is downregulated in cancer and inhibits tumor cell growth. Oncogene 26, 4580–4589.[CrossRef][Medline]

Etienne-Manneville, S. & Hall, A. (2002) Rho GTPases in cell biology. Nature 420, 629–635.[CrossRef][Medline]

Fukata, M., Nakagawa, M. & Kaibuchi, K. (2003) Roles of Rho-family GTPases in cell polarisation and directional migration. Curr. Opin. Cell Biol. 15, 590–597.[CrossRef][Medline]

Fukui, K., Sasaki, T., Imazumi, K., Matsuura, Y., Nakanishi, H. & Takai, Y. (1997) Isolation and characterization of a GTPase activating protein specific for the Rab3 subfamily of small G proteins. J. Biol. Chem. 272, 4655–4658.[Abstract/Free Full Text]

Hers, I., Wherlock, M., Homma, Y., Yagisawa, H. & Tavare, J.M. (2006) Identification of p122RhoGAP (deleted in liver cancer-1) Serine 322 as a substrate for protein kinase B and ribosomal S6 kinase in insulin-stimulated cells. J. Biol. Chem. 281, 4762–4770.[Abstract/Free Full Text]

Hildebrand, J.D., Taylor, J.M. & Parsons, J.T. (1996) An SH3 domain-containing GTPase-activating protein for Rho and Cdc42 associates with focal adhesion kinase. Mol. Cell. Biol. 16, 3169–3178.[Abstract/Free Full Text]

Homma, Y. & Emori, Y. (1995) A dual functional signal mediator showing RhoGAP and phospholipase C-{delta} stimulating activities. EMBO J. 14, 286–291.[Medline]

Huang, C., Jacobson, K. & Schaller, M.D. (2004) MAP kinases and cell migration. J. Cell Sci. 117, 4619–4628.[Abstract/Free Full Text]

Huttenlocher, A., Sandborg, R.R. & Horwitz, A.F. (1995) Adhesion in cell migration. Curr. Opin. Cell Biol. 7, 697–706.[CrossRef][Medline]

Kawai, K., Kiyota, M., Seike, J., Deki, Y. & Yagisawa, H. (2007) START-GAP3/DLC3 is a GAP for RhoA and Cdc42 and is localized in focal adhesions regulating cell morphology. Biochem. Biophys. Res. Commun. 364, 783–789.[CrossRef][Medline]

Kawai, K., Yamaga, M., Iwamae, Y., Kiyota, M., Kamata, H., Hirata, H., Homma, Y. & Yagisawa, H. (2004) A PLC{delta}1-binding protein, p122RhoGAP, is localized in focal adhesions. Biochem. Soc. Trans. 32, 1107–1109.[CrossRef][Medline]

Lavelin, I. & Geiger, B. (2005) Characterization of a novel GTPase-activating protein associated with focal adhesions and the actin cytoskeleton. J. Biol. Chem. 280, 7178–7185.[Abstract/Free Full Text]

Liao, Y.C., Si, L., deVere White, R.W. & Lo, S.H. (2007) The phosphotyrosine-independent interaction of DLC-1 and the SH2 domain of cten regulates focal adhesion localization and growth suppression activity of DLC-1. J. Cell Biol. 176, 43–49.[Abstract/Free Full Text]

Lo, S.H. (2004) Tensin. Int. J. Biochem. Cell Biol. 36, 31–34.[CrossRef][Medline]

Moon, S.Y. & Zheng, Y. (2003) Rho GTPase-activating proteins in cell regulation. Trends Cell Biol. 13, 13–22.[CrossRef][Medline]

Nishizuka, Y. (1995) Protein kinase C and lipid signaling for sustained cellular responses. FASEB J. 9, 484–496.[Abstract]

Nobes, C. & Hall, A. (1994) Regulation and function of the Rho subfamily of small GTPases. Curr. Opin. Genet. Dev. 4, 77–81.[CrossRef][Medline]

Qian, X., Li, G., Asmussen, H.K., Asnaghi, L., Vass, W.C., Braverman, R., Yamada, K.M., Popescu, N.C., Papageorge, A.G. & Lowy, D.R. (2007) Oncogenic inhibition by a deleted in liver cancer gene requires cooperation between tensin binding and Rho-specific GTPase-activating protein activities. Proc. Natl Acad. Sci. USA 104, 9012–9017.[Abstract/Free Full Text]

Rebecchi, M.J. & Pentyala, S.N. (2000) Structure, function, and control of phosphoinositide-specific phospholipase C. Physiol. Rev. 80, 1291–1335.[Abstract/Free Full Text]

Ridley, A.J. & Hall, A. (1992) The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70, 389–399.[CrossRef][Medline]

Ridley, A.J., Paterson, H.F., Johnston, C.L., Diekmann, D. & Hall, A. (1992) The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell 70, 401–410.[CrossRef][Medline]

Ridley, A.J., Schwartz, M.A., Burridge, K., Firtel, R.A., Ginsberg, M.H., Borisy, G., Parsons, J.T. & Horwitz, A.R. (2003) Cell migration: integrating signals from front to back. Science 302, 1704–1709.[Abstract/Free Full Text]

Sahai, E. & Marshall, C.J. (2002) RHO-GTPases and cancer. Nat. Rev. Cancer 2, 133–142.[CrossRef][Medline]

Sechi, A.S. & Wehland, J. (2000) The actin cytoskeleton and plasma membrane connection: PtdIns(4,5)P2 influences cytoskeletal protein activity at the plasma membrane. J. Cell Sci. 113(Pt 21), 3685–3695.[Abstract]

Sekimata, M., Kabuyama, Y., Emori, Y. & Homma, Y. (1999) Morphological changes and detachment of adherent cells induced by p122, a GTPase-activating protein for Rho. J. Biol. Chem. 274, 17757–17762.[Abstract/Free Full Text]

Wherlock, M. & Mellor, H. (2002) The Rho GTPase family: a Racs to Wrchs story. J. Cell Sci. 115, 239–240.[Free Full Text]

Wong, C.M., Lee, J.M., Ching, Y.P., Jin, D.Y. & Ng, I.O. (2003) Genetic and epigenetic alterations of DLC-1 gene in hepatocellular carcinoma. Cancer Res. 63, 7646–7651.[Abstract/Free Full Text]

Yam, J.W., Ko, F.C., Chan, C.Y., Jin, D.Y. & Ng, I.O. (2006) Interaction of deleted in liver cancer 1 with tensin2 in caveolae and implications in tumor suppression. Cancer Res. 66, 8367–8372.[Abstract/Free Full Text]

Yamada, T., Sakisaka, T., Hisata, S., Baba, T. & Takai, Y. (2005) RA-RhoGAP, Rap-activated Rho GTPase-activating protein implicated in neurite outgrowth through Rho. J. Biol. Chem. 280, 33026–33034.[Abstract/Free Full Text]

Yamaga, M., Sekimata, M., Fujii, M., Kawai, K., Kamata, H., Hirata, H., Homma, Y. & Yagisawa, H. (2004) A PLC{delta}1-binding protein, p122/RhoGAP, is localized in caveolin-enriched membrane domains and regulates caveolin internalization. Genes Cells 9, 25–37.[Abstract/Free Full Text]

Yuan, B.Z., Jefferson, A.M., Baldwin, K.T., Thorgeirsson, S.S., Popescu, N.C. & Reynolds, S.H. (2004) DLC-1 operates as a tumor suppressor gene in human non-small cell lung carcinomas. Oncogene 23, 1405–1411.[CrossRef][Medline]

Yuan, B.Z., Miller, M.J., Keck, C.L., Zimonjic, D.B., Thorgeirsson, S.S. & Popescu, N.C. (1998) Cloning, characterization, and chromosomal localization of a gene frequently deleted in human liver cancer (DLC-1) homologous to rat RhoGAP. Cancer Res. 58, 2196–2199.[Abstract/Free Full Text]

Yuan, B.Z., Zhou, X., Durkin, M.E., Zimonjic, D.B., Gumundsdottir, K., Eyfjord, J.E., Thorgeirsson, S.S. & Popescu, N.C. (2003) DLC-1 gene inhibits human breast cancer cell growth and in vivo tumorigenicity. Oncogene 22, 445–450.[CrossRef][Medline]

Zhou, X., Thorgeirsson, S.S. & Popescu, N.C. (2004) Restoration of DLC-1 gene expression induces apoptosis and inhibits both cell growth and tumorigenicity in human hepatocellular carcinoma cells. Oncogene 23, 1308–1313.[CrossRef][Medline]

Received: 9 October 2008
Accepted: 9 November 2008




This article has been cited by other articles:


Home page
 Annals of Clinical & Laboratory ScienceHome page
P.-p. Wu, Y.-l. Jin, Y.-f. Shang, Z. Jin, P. Wu, and P.-l. Huang
Restoration of DLC1 Gene Inhibits Proliferation and Migration of Human Colon Cancer HT29 Cells
Ann. Clin. Lab. Sci., January 1, 2009; 39(3): 263 - 269.
[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 Kawai, K.
Right arrow Articles by Yagisawa, H.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kawai, K.
Right arrow Articles by Yagisawa, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS