Skip Navigation


Carcinogenesis Advance Access originally published online on January 18, 2007
Carcinogenesis 2007 28(6):1145-1152; doi:10.1093/carcin/bgm008
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/6/1145    most recent
bgm008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Patel, V.
Right arrow Articles by Gutkind, J.S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Patel, V.
Right arrow Articles by Gutkind, J.S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Persistent activation of Rac1 in squamous carcinomas of the head and neck: evidence for an EGFR/Vav2 signaling axis involved in cell invasion

Vyomesh Patel1, Hans M. Rosenfeldt1, Ruth Lyons1,3, Joan-Marc Servitja1,4, Xosé R. Bustelo2, Mary Siroff1 and J.Silvio Gutkind1,*

1 Oral and Pharyngeal Cancer Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD 20892, USA
2 Centro de Investigación del Cáncer and Instituto de Biología Molecular y Celular del Cáncer, University of Salamanca-CSIC, Campus Unamuno, E-37007 Salamanca, Spain
3 Present address: Cancer Biology Program, Garvan Institute of Medical Research, St Vincent's Hospital, Sydney, Australia
4 Present address: Endocrinology, Hospital Clinic de Barcelona, Institut d'Investigacions Biomediques August Pi i Sunyer, Barcelona, Spain

* To whom correspondence should be addressed. Tel: +1 301 496 6259; Fax: +1 301 402 0823; Email: sg39v{at}nih.gov


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
The poor prognosis associated with head and neck squamous cell carcinoma (HNSCC) is primarily due to both local invasion and the regional and/or distant metastatic spread. Recent findings have provided evidence that the acquisition of a motile and invasive phenotype by cancer cells involves the dysregulated function of key intracellular molecular mechanisms together with aberrant signaling events initiated by the surrounding microenvironment. These intrinsic and extrinsic biochemical pathways in turn often converge to stimulate the activity of members of the Rho family of Ras-related guanosine triphosphate (GTP)-binding proteins, including RhoA, Rac and Cdc42, which control the organization of the actin cytoskeleton thereby regulating cell adhesion, polarity and motility. In this study, we examined the status of activation of these GTPases in a representative collection of HNSCC cell lines. Surprisingly, we found that most HNSCC cells exhibit remarkably high levels of GTP-bound Rac1. Further analysis revealed that the activation of Rac1 in these HNSCC cells could be due to two independent signaling events, an epidermal growth factor receptor (EGFR)-based autocrine loop that leads to the activation of the Rac1 exchange factor Vav2 and an EGFR/Vav2-independent pathway that arises as a consequence of the oncogenic mutation of the H-ras proto-oncogene. Indeed, we provide evidence that the EGFR/Vav2/Rac1 axis is a crucial pathway for the acquisition of motile and invasive properties of most HNSCC cells. These findings shed light onto the molecular mechanisms involved in HNSCC cell invasion, and may reveal new therapeutic opportunities to halt the metastatic spread of these aggressive malignancies.

Abbreviations: BSA, bovine serum albumin; DMEM, Dulbecco's modified Eagle's medium; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; GTP, guanosine triphosphate; HNSCC, head and neck squamous cell carcinoma


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
The overall decline in cancer death rates is largely attributable to our increased understanding of disease pathogenesis, thus affording new opportunities for the prevention, early detection and molecular-targeted treatment of some of the most highly prevalent cancer types. In contrast, the molecular architecture of head and neck squamous cell carcinoma (HNSCC), a tumor type that has been responsible for ~12 000 deaths in the USA during 2004 alone (1), still remains not fully understood (2,3). Although various therapeutic options are now becoming available to treat this disease, the 5-year survival rate for these tumors has improved only marginally for the last 30 years (1). The poor prognosis associated with this disease is largely due to both local invasion and the regional and/or distant metastatic spread of tumor cells. Indeed, patients with signs of metastatic disease have very poor prognosis (17% 5-year survival rate) (4,5). Arguably, a better understanding of the molecular mechanisms underlying the invasive and metastatic properties of HNSCC cells is required to help improving patient treatment and survival.

The acquisition of a motile and invasive phenotype is a crucial event in the progression of localized malignancies to a widespread metastatic disease [reviewed in ref. 6]. Recent findings have provided evidence that this step involves both intrinsic pathways activated in cancer cells as well as extrinsic cues emanating from the surrounding microenvironment (6–8). These intrinsic and extrinsic signaling pathways in turn often converge to stimulate the activity of members of the Rho family of Ras-related GTP-binding proteins, including RhoA, Rac and Cdc42, which control the organization of the actin cytoskeleton thereby regulating cell adhesion, polarity and motility (9–11). It is now known that these biological activities are coordinated by the concerted action of different Rho/Rac subfamily members. For instance, RhoA function is required for the initial formation of the lamellipodia and the establishment of focal adhesions and stress fibers, Rac1 promotes membrane ruffles and lamellipodia and Cdc42 trigger filopodia in the leading edge of migrating cells (10,11).

Based on this previous evidence, we decided to test the activation status of some of these GTPases in a large collection of HNSCC cell lines derived from well-defined clinical stage tumors (T2–T4) and displaying known genetic and molecular alterations that are typical of HNSCC (12–17) in order to begin dissecting the molecular events responsible for the migration and invasiveness of this tumor type. This experimental strategy revealed that most HNSCC cells exhibit remarkably high levels of GTP-bound Rac1. Instead, the levels of activation of Ras, RhoA and Cdc42 varied more widely in the same cell types. Further analysis revealed that the activation of Rac1 in these HNSCC cells could be due to two independent signaling events, an epidermal growth factor receptor (EGFR)-initiated pathway that leads to the activation of the Rac1 exchange factor Vav2 and an EGFR/Vav2-independent pathway utilized by oncogenic mutants of the H-ras proto-oncogene. Finally, we provide evidence indicating that the EGFR/Vav2/Rac1 axis is a crucial pathway for the acquisition of motile and invasive properties of most HNSCC cells. These findings shed light onto the molecular mechanisms by which HNSCC cells acquire the ability to invade their surrounding tissues, and suggest that inhibiting Rac1 or its downstream targets may provide novel therapeutic opportunities to prevent HNSCC tumor dissemination and metastasis.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Cell lines and culture conditions
Culture conditions of the HNSCC cell panel used in this study have been described in detail elsewhere (14,15). Briefly, cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum at 37°C in 95% air/5% CO2. HaCaT (spontaneously immortalized and non-tumorigenic human skin epidermal keratinocytes) and HEK293T (human embryonic kidney) epithelial cells were cultured as described above.

Cell stimulation
Epidermal growth factor (EGF) purchased from Sigma–Aldrich, St Louis (MO) was prepared as 100 µg/ml stock solution in 10 mM acetic acid containing 0.1% bovine serum albumin (BSA), according to manufacturer's instruction, and used at 100 ng/ml for 10 min unless indicated. N-(3-chlorophenyl)-6,7-dimethoxy-4-quinazolinamine (AG1478; EMD Biosciences, La Jolla, CA), an inhibitor of the EGFR was prepared as stock solution in dimethyl sulfoxide and diluted in media for the required concentration (10–50 µM).

Assessment of GTP levels of Ras superfamily proteins
Relative activity of the small GTPases (Ras, Rac1 and RhoA) was assessed by virtue of the ability to affinity precipitate their active GTP-bound forms upon their binding to effector domains that have been fused to glutathione-S-transferase bound to glutathione–Sepharose 4B beads, essentially as reported previously (18,19). Briefly, HNSCC and HaCaT cells were grown to 70–80% confluency, serum starved for 24 h and left untreated or followed by the indicated treatment (EGF stimulation, AG4178 inhibition and inhibition prior to stimulation) and cells were subsequently lysed with ice-cold buffer containing 10 mM Tris, 100 mM NaCl, 1% Triton X-100, 0.5 mM ethylenediaminetetraacetic acid, 40 mM ß-glycerophosphate, 10 mM MgCl2, 1 mM Na3VO4, 10 µg/ml aprotinin, 10 µg/ml leupeptin and 1 mM phenylmethylsulfonyl fluoride. Cellular debris was removed by centrifugation followed by protein determination of the supernatants by a colorimetric assay method (Bio-Rad, Hercules, CA). Equivalent (0.8–1.0 mg protein) amounts of total cellular lysates were then incubated at 4°C for precisely 30 min with 20–40 µg of purified, bacterially expressed (pGEX) glutathione-S-transferase fusion protein, containing the GTP–RhoA-binding domain of rhotekin–RhoA-binding domain, the Rac1–Cdc42-binding domain of p21/Cdc42/Rac1-activated kinase 1 (PAK1) or the RAS-binding domain of c-RAF1, previously immobilized to glutathione–Sepharose 4B beads (GE Healthcare, Piscataway, NJ). Next, after three washes in lysis buffer, the beads were boiled in 1x protein loading buffer, and the released proteins were resolved on 12% polyacrylamide–sodium dodecyl sulfate gels, transferred onto Immobilon-P membrane (Millipore Corporation, Bedford, MA), and the active form of the GTPases were detected using the indicated primary antibodies (Rac1, 1:1000, R56220 [GenBank] , BD Biosciences, San Jose, CA; RhoA, 1:500, sc-26C4, Santa Cruz Biotechnology, Santa Cruz, CA; Pan–Ras, 1:1000, Ab-4, Calbiochem) and horseradish peroxidase-conjugated goat anti-rabbit or goat anti-mouse secondary antibodies (Santa Cruz), followed by enhanced chemiluminescence (Pierce, RockFord, IL) and autoradiography. In parallel, 50 µg of total cellular lysates were similarly resolved and subsequently assessed for total levels of Rac1, RhoA and Ras proteins. The same lysates were further utilized to monitor EGFR levels and activity, using antibodies recognizing total and phosphorylated forms of the protein (1:1000; 610017 and E12120 [GenBank] , respectively; BD Biosciences).

Lysates recovered from pull-down experiments were further used to immunoprecipitate (1 h at 4°C) total tyrosine-phosphorylated proteins using a mixture of antibodies recognizing the phosphotyrosines (PY99, Santa Cruz; 4G10, Upstate Millipore, Charlottesville, VA; PY20, ICN Pharmaceuticals: Costa Mesa, CA; PY100, Cell Signaling, Beverly MA), and the captured complexes (GammaBind; GE Healthcare), after washing, were resolved and detected for the indicated protein with the appropriate antibodies.

Pre-designed siRNA oligonucleotides were utilized to target Vav2 and H-Ras for knock-down experiments. Two Vav2 sequences (#1: 5'-GGAACAGCGAGCUGUUUGAtt-3'; #2: 5'-GGGAUCAGGCCUUUUCCCUtt-3') and a single H-Ras sequence (5'-GCAGAUCAAACGGGUGAAGtt-3') together with control siRNA (all from Ambion, Austin, TX) were used to transfect (20 nM) into exponentially growing HNSCC cells using HiPerFect Transfection reagent (Qiagen, Valencia, CA). Cells were incubated for 48 h prior to serum starvation (12 h), and harvested for the presence of the active form of Rac1. Where indicated, H-Ras and ERK2 (C-20 and C-14, respectively; Santa Cruz) and ß-tubulin (556321, BD Biosciences) were used to monitor protein loading. Anti-Vav2 rabbit polyclonal antibodies were generated by us using a synthetic peptide corresponding to the Vav2 acidic region.

Cell migration assays
Serum-free DMEM containing 0.1% BSA was placed in the bottom wells of a Boyden chamber whereas similar medium containing 5 x 104 HNSCC cells was added to the top chamber. The two chambers were separated by a poretics polycarbonate polyvinylpyrrolidone free 8 µ pore size membrane (Osmonics, GE Osmonics Labstore, Minnetonka, MN) pre-coated with 50 µg/ml collagen IV (Trevigen, Gaithersburg, MD). After incubation for 9–12 h, the chamber was disassembled ensuring minimal disruption of migrated cells, and the membrane stained with Diff-Quick Stain (Dade Behring Inc., Deerfiled, IL), placed on a glass slide and the positively stained cells were quantified microscopically by counting 20 random high-power fields per filter. EGF (10–100 ng/ml) in serum-free DMEM containing 0.1% BSA was placed in the bottom wells and served as a positive chemoattractant for cell migration. One-way analysis of the variance of the migration data was performed followed by the Tukey's test to examine the differences among cell lines and test groups. A value of P < 0.05 was considered to be statistically significant.

Matrigel invasion assays
To monitor the invasion potential of HNSCC cells, BD MatrigelTM Matrix suspension (BD Biosciences) was used to coat 12 µ pore Transwell inserts (Corning Inc, Lowell, MA) and subsequently left to dry. Next, cells were removed from culture plates with Hanks' balanced salt solution/5 mM ethylenediaminetetraacetic acid/25 mM N-[2-Hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid] pH 7.2, washed in DMEM/0.1% BSA, counted and 5 x 104 cells seeded on top of the dried Matrigel in the inserts. DMEM/BSA containing EGF (100 ng/ml), or DMEM/BSA alone as control, was added to the lower chamber, and plates were incubated at 37°C for 24 h. Filters were fixed (4% formaldehyde/phosphate-buffered saline), stained with crystal violet (0.2% crystal violet in 2% ethanol) and after removing the contents of the upper membrane surface, invasive cells were microscopically counted in 20 random high-power fields per filter. For Vav2 knock down, exponentially growing cells were transfected, using HiPerFect Transfection reagent (Qiagen), with 20 nM of either control or Vav2 (#1) siRNA and after that incubated for 48 h. Cells were then serum starved and processed for invasion assay as detailed above. Statistical analysis of the data was performed as described above for migration assays.

Construction of expression libraries from HNSCC cells
Total RNA was extracted from exponentially growing HNSCC cells using Trizol (Invitrogen, Carlsbad, CA) and followed by the purification of mRNA following provided instructions (mRNA Purification Kit, GE Healthcare). Next, the mRNA was used as template to construct cDNA expression libraries in the pEAK8 cloning vector essentially following the provided guidelines (Edge BioSystems, Gaithersburg, MD),

Selection of transforming clones and focus formation assays
To identify transforming sequences from HNSCC cells, bacteria transformed with the primary cDNA expression library (~2 x 106 individual clones) were plated in 20 independent pools (105 bacterial clones each), and DNA plasmids and bacterial stocks generated from each pool. Pools containing transforming sequences were identified by transfecting NIH 3T3 cells and counting cell foci after 2 weeks of culture, as described previously (20). Only two pools contained focus-forming sequences, and each was divided into 20 subsequent pools (~5000 colonies per plate). Secondary pools containing the highest numbers of transforming plasmids were sequentially divided into 20 pools for two more rounds, until individual transforming plasmids were identified and sequenced.

Amplification, subcloning and sequencing of exon 1 of the H-ras locus
Genomic DNA isolated from HNSCC cells was used as template to amplify exon 1 of the human proto-oncogene c-H-Ras using the standard polymerase chain reaction method and the following primers: 5'-AGACCCTGTAGGAGGACCCC; 3'-CTGGGCTCGCCCGCAGCAGCTGCTGG. Next, the expected 206 bp polymerase chain reaction products were excised from the gel, purified and subcloned into the pGEM-T Easy Vector System (Promega, Madison, WI) for subsequent sequence analysis using T7 (forward) and SP6 (reverse) primers.

Expression plasmids
Wild-type human H-Ras and the corresponding V12 mutant were subcloned into the pCEFL-KZ AU5 expression vector (19). For the D12 mutant, the DNA was initially amplified from the pEAK8 expression library and the 570 bp, and inserted into the pCEFL-KZ AU5 vector. All clones were sequence verified prior to further biochemical analysis. Next, the ability of c-H-Ras and related mutants to influence GTPase activiy was assessed in vitro. Briefly, 293T cells were transfected with the indicated plasmids, and incubated for a further 12–16 h prior to serum starvation for an additional 12–16 h cells. Cells were lysed at 4°C and processed to detect GTP–Rac1 activity essentially as described above. AU5 (Covance, Denver, PA) tag antibody was used to monitor the expression efficiency of the transfected plasmids. Expression plasmids for Vav2 were used as positive controls (19).

Generation of adenoviral particles and viral infection of eukaryotic cells
Production of high-titer recombinant adenovirus using the AdEasy system has been described previously (21). Adenoviruses encoding the myc-epitope-tagged dominant-negative Rac1 cDNA containing a substitution at position 17 (Rac1N17; provided by Dr Toren Finkel, National Institutes of Health) was essentially prepared and amplified as described (22). Briefly, the high-titer viral stock was initially used to transfect 911 cells, and subsequently amplified using fresh 911 cells. The control virus expressed green fluorescent protein (GFP) (Adeno-GFP). Amplified virus underwent purification by cesium chloride centrifugation and subsequent titration was performed by plaque assay in low-passage 293 cells using the Adeno-X Rapid Titer kit and instructions provided in the manual (Invitrogen). A series of preliminary infections were performed in a subset of the cell lines each to determine the optimal dose of virus based on its limited cytotoxicity and effects on the growth rate when compared with the mock-infected cells. All subsequent infections were performed at a multiplicity of infection of 50 for 12 h on cells that had been maintained in culture for 24 h (serum free). Assessment of gene transfer efficiency after infection was by western blot as well as by examination of high-efficiency GFP expression.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Detection of activation levels of endogenous GTPases in epithelial cells
As Rho/Rac GTPases are only active during a short window of time upon cell stimulation, we initially wished to set up experimental conditions that could be compatible with the reproducible detection of the activated levels of endogenous Ras superfamily proteins in tumor cells. For this purpose, we first evaluated whether variations in the GTP levels of a number of GTPases could be detected in HaCaT, an immortalized, non-tumorigenic epithelial cell line (23). To this end, we performed pull-down experiments with appropriate glutathione-S-transferase fusion protein probes to detect GTP-loaded GTPases in quiescent and stimulated cells (see Materials and Methods). As shown in Figure 1, these conditions allowed the detection of the rapid and transient stimulation of Ras, Rac1 and RhoA upon stimulation of quiescent cells with either EGF or serum (Figure 1 and data not shown, respectively), indicating that they were probably optimal for the detection of active GTPases in tumor cells.


Figure 1
View larger version (43K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. EGF activates Ras and Rho GTPases in HaCaT cells. HaCaT cells were serum starved followed by EGF stimulation (100 ng/ml) for the indicated times. Cells were lysed and equivalent amounts of the recovered protein were subsequently incubated with the appropriate Sepharose-bound glutathione-S-transferase fusion proteins including the Ras-, Rac1- and RhoA-binding region of Raf-1 (RAF), PAK1 (PAK) and rhotekin, respectively. GTP-bound forms of these GTPases were washed and resolved together with total extracts, by sodium dodecyl sulfate–polyacrylamide gel electrophoresis followed by western blot analysis using antibodies recognizing Ras (top panel), Rac1 (middle) and RhoA (bottom). Data shown are representative of those obtained from three independent experiments. AP, affinity precipitation; TCL, total cell lysate.

 
Activation of Rac1 is a common feature in HNSCC cell lines
We next used these pull-down experiments to monitor the activation status of Rac1, RhoA and Ras in a representative panel of nine HNSCC cell lines (HN8, HN12, HN13, HN17, HN19, HN22, HN26, HN30 and HN31). These cell lines were selected based on their known pattern of genetic aberrations and typical alterations of signaling pathways (12–17,24). As a control, we used the EGF-stimulated HaCaT cells to demonstrate the detection of GTP-loaded GTPases under these conditions. In addition, we used immunoblot analysis of total cellular lysates using antibodies to Ras proteins and ß-tubulin to demonstrate the equal loading of all samples. These experiments indicated that Ras was activated in more than half of the selected HNSCC cell lines (Figure 2, top panel, see lanes 3, 4, 6, 7, 8 and 9). These levels of activity were, however, lower than those found in EGF-stimulated HaCaT cells (Figure 2, top panel, lane 11). RhoA showed a more restricted pattern of activation, since only two cell lines of the panel (HN30 and HN31) displayed high levels of GTP–RhoA (Figure 2, fifth panel from top, lanes 8 and 9). In parallel experiments, we also observed that most HNSCC cells exhibit a similar level of active Cdc42 (data not shown). In contrast with the above results, we found that Rac1 is activated in a wide range of HNSCC cells (Figure 2, third panel from top, lanes 2–4 and 6–9). These levels of activity were similar, or even higher, than those induced by EGF in HaCaT cells (Figure 2, third panel from top, lane 11). Immunoblot analysis confirmed that all GTPases were expressed at similar levels with the only exception of the loss of RhoA in the HN13 cell line (Figure 2, second, fourth and sixth panels from top). This was not due to unequal loading, since ß-tubulin and the rest of GTPases were detected in that sample (Figure 2, bottom panel). These data indicate that constitutive activity of Rac1 is a common and specific feature of most HNSCC cells.


Figure 2
View larger version (40K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Detection of active GTPases in HNSCC. Exponentially growing HNSCC cells were serum starved and lysed and equivalent amounts of the recovered protein extracts were incubated with the appropriate Sepharose-bound glutathione-S-transferase fusion protein. GTP-bound GTPases were affinity precipitated (AP) and analyzed in parallel with total cell lysates (TCLs) as above. In each case, HaCaT cells were included in the analysis to serve as negative (unstimulated) and positive (EGF stimulated) controls. Results shown are representative of three independent experiments. Levels of ß-tubulin were used as loading control (lower panel).

 
A role for Rac1 in HNSCC migration and invasion
We initially assessed HNSCC cell lines displaying contrasting levels of GTP–Rac1 for cell migration using Boyden chamber assays. Consistent with the role of Rac1 in cell migration, cells with elevated levels of GTP–Rac1 (HN12, HN13, HN17 and HN26) displayed a high basal migratory activity, whereas those cells with low (HN8 and HN19) to medium (HN22) Rac1 activity, migrated less efficiently (Figure 3A). The migratory activity of most HNSCC cells was further enhanced upon treatment of cells with chemoattractants such as serum or EGF. The only exception was HN30 cells that had elevated basal migratory activity but impaired chemotaxis toward EGF (Figure 3A). To examine whether this motility was Rac1 dependent, we evaluated whether the inactivation of the endogenous Rac1 protein could lead to the reduction of the migratory activity of HNSCC cells. To this end, we expressed in the indicated cell lines the dominant-negative version of Rac1 (Rac1N17 mutant) using an adenoviral gene delivery system (22). This strategy allowed the expression of Rac1N17 at significantly higher levels than the endogenous wild-type Rac1 protein both in HaCaT and HNSCC cells (Figure 3B, insert; data not shown). Under these conditions, Rac1N17 caused a significant reduction in both the basal and EGF-stimulated migratory activity of HaCaT, HN12 and HN13 cells, as an example. In contrast, it was totally ineffective when over-expressed in HN30 cells.


Figure 3
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. A role for Rac1 in HNSCC cell migration and invasion. Serum-starved HNSCC cells were assessed either in the presence or absence of EGF for their ability to migrate using a Boyden chamber. Serum-free DMEM containing 0.1% BSA (control) or EGF (100 ng/ml) (EGF) was placed in the lower part of the chamber, whereas cells in DMEM/0.1% BSA were added to the top part of the Boyden chamber, with both components separated by a membrane. After incubation (9–12 h), the chamber was dismantled and cell migration was assessed by counting the stained cells attached to the membrane. Data are represented as the number of cells in 20 random high-power fields ± SEM in triplicate membranes from a representative experiment (A). Next, HaCaT cells infected with Adeno-GFP and Adeno-Rac1N17 were assessed by western blot for efficiency of gene delivery (B, insert). Uninfected cells (5 x 104) (control) or cells infected with Adeno-GFP or Adeno-Rac1N17 were assessed for their migratory activity toward wells containing medium without (C) or with EGF (EGF), as described above. Data are represented as the number of cells in 20 random high-power fields ± SEM in triplicate membranes from a representative experiment (B). For invasion, similar control and treated cells in DMEM/0.1% BSA were placed into Transwell inserts coated with Matrigel. Either DMEM/BSA containing EGF (100 ng/ml) (EGF) or DMEM/BSA alone as control (C) was added to the lower chamber and incubated at 37°C for 24 h, after which the cells were fixed and stained, and invasive cells were microscopically counted in 20 random high-power fields per filter. Data are represented as the number of invading cells ± SEM from a typical experiment run in triplicate (B). Analysis of the variance and Tukey's post-test statistical analysis was performed to compare among cell lines and groups, as indicated (*P < 0.05, **P < 0.01 and ***P < 0.001).

 
We next investigated whether Rac1N17 could influence the capacity of these HNSCC cells to invade in vitro using Matrigel layers in the presence or absence of EGF. As shown in Figure 3B, mock-transfected HNSCC, but not HaCaT cells, invaded the matrix in basal culturing conditions. As expected, similar to HaCaT cells, most HNSCC exhibiting limited Rac1 activity displayed poor basal invasive potential (data not shown). EGF enhanced such behavior in all tumoral and non-tumoral cells with the single exception of the HN30 cell line (Figure 3C). Likewise, the over-expression of Rac1N17 abrogated the invasiveness of all cell lines but that of HN30 (Figure 3B). Taken together, these findings suggest that Rac1 does play an essential role in the chemoattractant-induced migration and invasion of most HNSCC cells. However, there is a small population of HNSCC cell lines, reflected by HN30, whose migration/invasiveness is Rac1 independent.

Persistent tyrosine phosphorylation of Vav2 in HNSCC cells
We next sought to investigate the mechanism underlying the elevated activity of Rac1 in HNSCC cells. In this regard, considering that a large fraction of the HNSCC cells display elevated activity of the EGFRs (24), we focused our attention on the Vav family of Rho/Rac guanine nucleotide exchange factors, as they act downstream from both receptor and non-receptor tyrosine kinases (25). Among the Vav family members, Vav1 and Vav3 are primarily expressed in cells of the immune system, whereas Vav2 is widely expressed, including in cells of epithelial origin (25). Thus, Vav2 represented the best candidate to mediate the activation of Rac1 in HNSCC cells. The catalytic activity of this GEF family is regulated by direct phosphorylation on tyrosine residues, a property that allows monitoring its level of activation in vivo by performing anti-phosphotyrosine immunoprecipitations or immunoblots (19,25,26). To assess the activation levels of Vav2, cell lysates from HNSCC cells were immunoprecipitated with anti-phosphotyrosine antibodies and the presence of this GEF was determined by anti-Vav2 immunoblots. As shown in Figure 4 (top panel), high levels of phospho-Vav2 were observed in six out of the nine cell lines tested. Two of the cell lines negative for Vav2 phosphorylation expressed reduced or non-detectable levels of Vav2 protein (Figure 4, compare the first and second panels). Anti-ERK2 immunoblots confirmed that all samples contained similar amounts of loaded material (Figure 4, bottom panel). As observed previously in other cell types, non-tumoral cells showed low levels of phospho-Vav2 unless they were stimulated with mitogens such as EGF (Figure 4, top panel on the right). In the case of control HaCaT cells, this phosphorylation was mediated by the EGFR as judged by the dramatic drop of Vav2 phosphorylation in cells that had been treated with AG1478, an specific EGFR inhibitor (Figure 4, top panel on the right). These data indicate that the constitutive tyrosine phosphorylation of the Rac1 GEF Vav2 is a common feature in most HNSCC cells, suggesting that the Vav2/Rac1 axis is a common point of signal convergence in these cells.


Figure 4
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Phosphorylation of Vav2 in HNSCC. Total cell lyates from exponentially growing HNSCC cells were used to immunoprecipiate tyrosine-phosphorylated proteins with a cocktail of appropriate antibodies (PY20, PY99, PY100 and 4G10). The protein complexes were captured (GammaBind), washed and subsequently analyzed by western blot for the presence of phosphorylated Vav2. Total expression levels of Vav2 were monitored in parallel using 50 µg of the same total cell lysates. Loading control was assessed by ERK2 expression. As additional controls, lysates from HaCaT cells, either untreated or pre-treated with AG1478 followed by stimulation with EGF, were analyzed by western blot for the status of phosphorylation of Vav2, together with that of total Vav2 and ERK. The data shown are representative of those obtained from three independent experiments.

 
Of interest, one HNSCC cell line, HN8, exhibited a persistently elevated level of tyrosine-phosphorylated Vav2 (Figure 4), but not enhanced Rac1 activity (see above, Figure 2), thus raising the possibility that Vav2 phosphorylation may not be sufficient to activate Rac1. Alternatively, considering that certain Rho and Rac GAPs, such as p190 RhoGAP, can be stimulated by tyrosine phosphorylation (27), it is still conceivable that in this cell line EGF may trigger two opposing and thus counteracting effects on Rac1, by regulating its GEFs and GAPs simultaneously. On the other hand, a possibility still exists that the overall tyrosine phosphorylation of Vav2 may not reflect fully its activation state, as in HN8 cells, Vav2 may not be phosphorylated in certain specific tyrosine residues that are required to elevate its Rac–GEF activity (25). These and additional possibilities that may help explain why this particular cell line exhibiting detectable levels of tyrosine-phosphorylated Vav2 does not display increased Rac1 activity are under current investigation.

Rac1 activation in HNSCC occurs through EGFR/Vav2-dependent and -independent routes
Given that Vav2 is a target of the EGFR and that this tyrosine kinase receptor is frequently activated in HNSCC cells, we next investigated whether the constitutive high levels of activated Rac1 in those cells were dependent on an EGFR-initiated signaling event. This hypothesis was evaluated by measuring the levels of phosphorylated EGFR by immunoblot analysis. We selected for these studies the HN13 and HN30 cell lines because they show high levels of Rac1 activity but different patterns of Vav2 phosphorylation (see above). These experiments indicated that quiescent, non-stimulated HN13 cells have detectable levels of EGFR activity (Figure 5A, top panel, lane 1). This activity was significantly enhanced upon stimulation of cells for 5 min by EGF (Figure 5A, top panel, lane 3). The phosphorylation levels of the EGFR in both conditions could be eliminated upon treatment of cells with the AG1478 inhibitor (Figure 5A, top panel, lanes 2 and 4, respectively), indicating that this assay was a bonafide reporter of the kinase activity of this membrane receptor. The same cell lysates were then used to estimate the stimulation status of Rac1 and its upstream regulator Vav2. The activation of both Rac1 and Vav2 followed an identical profile to that observed before for the EGFR (Figure 5A, third and fifth panels from top), suggesting that this receptor does promote the stimulation of the Vav2/Rac1 signaling pathway. Immunoblot analysis confirmed the presence of equal amounts of each of these proteins in the lysates under study (Figure 5A, second, fourth, sixth and seventh panels from top). Interestingly, HN30 cells showed a different behavior under identical experimental conditions, since no variations in the levels of either Rac1 or Vav2 were observed upon EGF stimulation or AG1478 treatment (Figure 5B, third and fifth panels from top). This was not due to alterations in the signaling of the EGFR, since this molecule underwent similar responses to EGF and AG1478 when compared with the HN13 cell line (Figure 5B, top panel). These results indicate that activation of Rac1 in HNSCC cells can occur through two different signaling mechanisms.


Figure 5
View larger version (40K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Role of EGFR in Vav2 phosphorylation and Rac1 activity in HNSCC cells. Serum-starved HN13 (A) and HN30 (B) HNSCC cells were either untreated, pre-treated with AG1478, stimulated with EGF, or pre-treated with AG1478 followed by EGF, as indicated, and total cell lysates (TCLs) were analyzed by western blot for activated and total levels of EGFR proteins. In parallel, lysates were immunoprecipitated (IP) with anti-phosphotyrosine antibodies or affinity precipitated (AP) with glutathione-S-transferase–PAK, and analyzed for the presence of Vav2 or Rac1, respectively. Total levels of Vav2 and Rac1 were also determined in cell lysates. Immunoblotting for ß-tubulin served as a loading control. All experiments were performed in triplicates.

 
To characterize the former route that promotes Rac1 activation in HNSCC cells, we used two siRNAs to knock down the human vav2 mRNA. As shown in Figure 6A (third panel from top, lanes 3–6), these siRNAs (referred to as #1 and #2) effectively reduced the levels of the endogenous Vav2 protein. This effect was specific, since a control siRNA did not have any effect on the expression levels of this GEF (Figure 6A, third panel from top, lanes 1 and 2). The levels of other unrelated proteins were not affected either (Figure 6A, second and fourth panels from top). Interestingly, the reduction in the levels of Vav2 protein led to the abrogation of the EGF-dependent activation of Rac1 in those cells (Figure 6A, top panel, compare lanes 4 and 6 with lane 2), further indicating that Vav2 acts as the crucial signaling molecule that links EGFR stimulation with the activation of Rac1 in HN13 cells. Similar results were observed in HaCaT cells (data not shown). However, in good agreement with our previous data (see Figure 5), the elimination of Vav2 protein did not have any effect in the constitutive high levels of Rac1 in HN30 cells (Figure 6B, top panel). Likewise, the depletion of Vav2 protein decreased the basal and EGF-dependent invasive activity of HN12 and HN13 cells (Figure 6) and in HaCaT cells when used as controls (data not shown). In contrast, it had no negative influence on the invasion behavior of HN30 cell line (Figure 7). Taken together, these results indicate that there are two alterative pathways to promote Rac1 activation in HSNCC cells: one of them relies on the activation of the EGFR/Vav2 signaling route whereas the other one is independent on these two molecules.


Figure 6
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. Vav2 knock down inhibits the activation of Rac1 in HN13 cells. HNSCC cells (A, HN13; B, HN30) were transfected with control (C) and Vav2 siRNA (#1 and #2) sequences. Following incubation for 48 h, cells were serum starved overnight and where indicated stimulated with EGF (100 ng/ml) for 10 min. All cells were harvested to assess for the presence of the active form of Rac1. For loading controls, levels of total Rac1 and ß-tubulin were monitored, and Vav2 levels were determined for knock-down efficiency. Because of the very limited levels of Vav2 in HN30 cells, lysates from HEK293T cells transfected with Vav2 expression vector were used as a control.

 


Figure 7
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7. Vav2 knock-down results in reduced basal and EGF-stimulated cell invasion. Exponentially growing HNSCC cells (HN12, HN13 and HN30) were transfected with control (C) and Vav2 (#1) siRNA, and incubated for 48 h prior to serum starvation for an additional 12 h. After detaching and washing, cells (5 x 104) were placed into 12 µ pore Transwell inserts coated with Matrigel. Either DMEM/BSA containing EGF (100 ng/ml) or DMEM/BSA alone as control was added to the lower chamber and incubated at 37°C for 24 h, after which the cells were fixed and stained. After removing the contents of the upper membrane surface, invasive cells were microscopically counted in 20 random high-power fields per filter. Data represent the number of invading cells ± SEM per field from a typical experiment conducted in triplicate.

 
The Ras pathway mediates Rac1 activation in HN30 cells
To identify the nature of the second pathway that mediates Rac1 activation in certain HSNCC cells, we decided to look for the presence of oncogenic events in HN30 cells. We hypothesized that if the constitutive activity of Rac1 in HN30 cells were due to the presence of an oncogenic mutation, that oncogene could be identified by focus formation assays using expression cDNA libraries derived from HN30 transcripts. To this end, we first cloned HN30-derived cDNAs in an EF-1-driven mammalian expression vector and the resulting cDNA library was then used to transfect NIH 3T3 cells and the presence of potential oncogenic transcripts visualized by the generation of foci of transformed cells (see Materials and Methods). This expression cloning strategy led to the identification of multiple plasmids encoding a missense mutation (G12 -> D) in the H-ras proto-oncogene (Figure 8A). This mutation was further corroborated by the detection of this mutation in the exon 1 of H-ras locus of HN30 cells. The wild-type allele was also detected in these assays, indicating that these cells were heterozygous for the mutation (data not shown). This mutation was also found in the HN31 cell line but not in the rest of cells used in our study. This is probably to the fact that the HN30 and HN31 cell lines were both isolated from the same cancer patient (14). However, HN31 cells exhibited lower levels of activated H-Ras, which may explain their reduced Rac1 activity as compared with HN30 (see above, Figure 2A, top and middle panels).


Figure 8
View larger version (37K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8. Mutational and functional roles of H-RasD12 in Rac1 activation in HN30 cells. mRNA extracted from HN30 cells was used as template to construct cDNA expression libraries, and clones underwent sequential selection by virtue of their focus-forming activity in NIH 3T3 cells. (A) Transforming sequences were isolated and sequence data identified multiple plasmids encoding H-ras with a G -> A mutation in codon 12, which changes the coding sequence from glycine to aspartic acid (H-RasD12). (B) To test the ability of this mutation to activate Rac1, AU5-tagged H-Ras (WT, V12 and D12) were transfected in HEK293T cells and after serum starvation, cell lysates were assessed for the presence of active form of Rac1, using the expression of Vav2 as a positive control. Expression of AU5-tagged Ras proteins was confirmed, and total levels of Rac1 and total Ras proteins (pan–Ras) were used as loading controls. (C) Knock down of H-Ras decreases Rac1 activation in HN30 but not in HN13 cells. Levels of activated Rac1 were assessed in cells untransfected (C) and transfected with control or H-Ras siRNA after affinity precipitation (AP) of GTP–Rac1 with glutathione-S-transferase–PAK, as described above. Total cell lysates (TCLs) from these samples were further analyzed by western blotting to detect the total levels of Rac1 and H-Ras, using ß-tubulin as a loading control. Notice the reduction in H-Ras expression levels in cells transfected with its targeting siRNA.

 
To test that the H-RasG12D mutant was sufficient to induce the activation of Rac1 in vivo, we performed pull-down experiments to detect the activation levels of Rac1 in the presence or absence of this Ras mutant. As comparative controls, we also included in these experiments the wild-type version of H-Ras, the standard oncogenic version of H-Ras (G12V mutant) and the oncogenic version of Vav2. As shown in Figure 8B (upper panel), mock-transfected and H-Ras-expressing cells do not have detectable levels of GTP-bound Rac1. In contrast, the expression of either H-RasG12V or H-RasG12D was sufficient to induce levels of activation of Rac1 comparable with those found in Vav2 oncoprotein-expressing cells. These data supported that the EGFR/Vav2-independent pathway involved in the activation of Rac1 in HN30 cells is likely initiated downstream of the H-Ras oncoprotein.

To validate the contribution of H-RasG12D in the activation Rac1 in HN30 cells, we utilized siRNA technology targeting H-Ras in these cells to examine if this would result in a decreased level of active Rac1. HN13 cells were also assessed as a control. As seen in Figure 8C, the efficient knock down of H-Ras in HN13 cells by the use of siRNA to H-Ras did not result in decreased Rac1 activity (top left panel, third lane from left). In contrast, Rac1 activity was clearly diminished in HN30 cells transfected with siRNA targeting H-Ras when compared with the corresponding controls (top right panel, third lane from left). The effectiveness of the H-Ras siRNA was confirmed by assessing the levels of H-Ras, which was reduced only in those cells transfected with the appropriate siRNA. Total Rac1 and ß-tubulin levels were also monitored as loading controls. These findings supported the role of mutated H-ras in promoting the constitutive elevated levels of active Rac1 in HN30 cells.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
Previous publications have shown that GTPases of the Rho family are frequently over-expressed in different cancer types, an event that usually correlates with a poor outcome (28,29) (reviewed in ref. 30). However, most studies have been based primarily on the analysis of mRNA expression that may not necessarily reflect actual changes in protein expression and/or function. In this context, the detection of GTP-bound Rho/Rac proteins is likely to be more informative from a functional point of view. Based on this body of information, we have analyzed the activation status of a number of Ras and Rho/Rac GTPases in HNSCC cells. Our results revealed that HNSCC cells display remarkably high levels of GTP-bound Rac1, a molecular event that correlates with the enhanced migration and invasion of these cells. In contrast, the activation of Ras and RhoA shows a much more restricted profile, and we did not observe major differences in the levels of active Cdc42 among the different HNSCC cell lines. While searching for the underlying mechanism that promotes the activation of this GTPase, we found that HNSCC cells can promote Rac1 activation at least by two independent and mutually exclusive routes. The most prevalent pathway is based on the over-activity of EGFR, and the downstream activity of the Rac1 GEF Vav2.

Several lines of evidence support that Rac1 is activated downstream from EGFR through Vav2 in most HNSCC cells. Indeed, decreasing the activity of EGFR with EGFR inhibitors reduced the basal and EGF-stimulated accumulation of Rac1–GTP. In addition, activation of Rac1 correlated well with the recovery of Vav2 in anti-phosphotyrosine immunoprecipitates, and Vav2 phosphorylation was sensitive to EGFR blockade. Furthermore, reducing Vav2 expression levels by RNA interference techniques decreased both the basal levels of active Rac1 as well as the activation of this GTPase by EGF. Taken these data together, we can postulate that the over-activity of an EGFR/Vav2/Rac1 signaling route is a frequent event in HNSCC. In addition, the observations that the basal and EGF-stimulated invasiveness of HNSCC that have elevated activity of this pathway is totally blocked using vav2-specific siRNA and the Rac1N17 dominant-negative mutant further support the emerging concept that that the EGFR/Vav2/Rac1 route is essential for the enhanced migratory and invasive activity of most HNSCC cells.

A less prevalent pathway that participates in the activation of Rac1 in HNSCC cells is the constitutive activation of the Ras route due to a missense mutation in the H-ras locus. The discovery of H-ras mutations in a small subset of HNSCC cell lines is somewhat unexpected. Indeed, previous studies have shown that these mutations are quite rare in the Western population, although they are detected at rather high levels (35%) in a betel quid-chewing patient cohort from Southeast Asian countries (3,31–33). In addition to the effects that the expression of the mutant H-RasG12D protein may have in the well-known Ras downstream elements, this protein promotes the effective activation of Rac1 independently of extracellular stimuli. The mechanism by which H-RasG12D promotes Rac1 activation is hitherto unknown. Tiam1, a Rac1-specific GEF that is regulated by direct binding to Ras proteins (34), is expressed in HNSCC cells. However, in contrast to previous results, we could not detect any specific interaction between Tiam1 and H-RasG12D in our cells. It is also plausible to hypothesize that H-RasG12D may mediate activation of Rac1 via a different downstream target such as phosphatidylinositol 3-kinase or other Rac–GEFs. Further work in this area may help identify the specific signaling pathway linking H-RasG12D with Rac1 in these cells.

Of interest, the effects of H-RasG12D on Rac1 are probably independent of the specific ras mutation because its best-described mutant, H-RasG12V (35), induces similar levels of Rac1 activation, at least upon its ectopic expression (Figure 8B). HN30 also exhibited very high levels of active RhoA, thus raising the possibility that in HNSCC cells harboring ras mutations their protein products may stimulate the activity of a yet to be identified GEF acting on both RhoA and Rac1 GTPases. On the other hand, inhibition of Rac1 by the use of its dominant-negative mutant did not decrease the basal migratory and invasive capacity of HN30 cells. These cells are characterized for the presence of a wild-type p53 (15), and nonetheless are quite resistant to pro-apoptotic stimuli such as {gamma}-radiation and DNA-damaging agents (36). Thus, we cannot rule out that in cells harboring ras mutations, Rac1 may not contribute to their migratory activity, but instead Rac1 may play additional roles, including the regulation of cell proliferation and survival (37), which warrants further investigation.

In summary, our data indicate that most HNSCC cells exhibit a persistently active EGFR/Vav2/Rac1 signaling axis that contributes to tumor invasion and metastasis. In this regard, Rac1 can coordinate the activation of molecules involved in the remodeling of the actin cytoskeleton, such as the serine–threonine kinase PAK1, with the stimulation of intracellular kinase cascades regulating c-jun N-terminal kinase, a member of the mitogen-activated protein kinase superfamily of serine–threonine kinases, which in turn may regulate the expression of proteases involved in the active remodeling of the extracellular matrix and tissue invasion (9–11,38,39). Thus, in this scenario, our present findings suggest that inhibiting Rac1 GEFs, Rac1 itself or specific Rac1 downstream targets, such as PAK1 and c-jun N-terminal kinase, may represent an effective strategy to prevent HNSCC tumor dissemination and metastasis, especially in patients exhibiting EGFR hyperactivity.


   Acknowledgments
 
Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Jemal A, et al. Cancer statistics, 2005. CA Cancer J. Clin. (2005) 55:10–30.[Abstract/Free Full Text]
  2. Forastiere A, et al. Head and neck cancer. N. Engl. J. Med. (2001) 345:1890–1900.[Free Full Text]
  3. Mao L, et al. Focus on head and neck cancer. Cancer Cell (2004) 5:311–316.[CrossRef][ISI][Medline]
  4. Landis SH, et al. Cancer statistics, 1999. CA Cancer J. Clin. (1999) 49:8–31.[Abstract/Free Full Text]
  5. Parkin DM, et al. Global cancer statistics. CA Cancer J. Clin. (1999) 49:33–64.[Abstract/Free Full Text]
  6. Hanahan D, et al. The hallmarks of cancer. Cell (2000) 100:57–70.[CrossRef][ISI][Medline]
  7. Hunter K. Host genetics influence tumour metastasis. Nat. Rev. Cancer (2006) 6:141–146.[CrossRef][ISI][Medline]
  8. Condeelis J, et al. Macrophages: obligate partners for tumor cell migration, invasion, and metastasis. Cell (2006) 124:263–266.[CrossRef][ISI][Medline]
  9. Sahai E, et al. RHO-GTPases and cancer. Nat. Rev. Cancer (2002) 2:133–142.[CrossRef][ISI][Medline]
  10. Etienne-Manneville S, et al. Rho GTPases in cell biology. Nature (2002) 420:629–635.[CrossRef][Medline]
  11. Raftopoulou M, et al. Rho GTPases lead the way. Dev. Biol. (2004) 265:23–32.[CrossRef][ISI][Medline]
  12. Jeon GA, et al. Global gene expression profiles of human head and neck squamous carcinoma cell lines. Int. J. Cancer (2004) 112:249–258.[CrossRef][ISI][Medline]
  13. Patel V, et al. Altered expression and activity of G1/S cyclins and cyclin-dependent kinases characterize squamous cell carcinomas of the head and neck. Int. J. Cancer (1997) 73:551–555.[CrossRef][ISI][Medline]
  14. Yeudall WA, et al. MTS1/CDK4I is altered in cell lines derived from primary and metastatic oral squamous cell carcinoma. Carcinogenesis (1994) 15:2683–2686.[Abstract/Free Full Text]
  15. Yeudall WA, et al. Functional characterization of p53 molecules expressed in human squamous cell carcinomas of the head and neck. Mol. Carcinogen. (1997) 18:89–96.[CrossRef][ISI][Medline]
  16. Amornphimoltham P, et al. Mammalian target of rapamycin, a molecular target in squamous cell carcinomas of the head and neck. Cancer Res. (2005) 65:9953–9961.[Abstract/Free Full Text]
  17. Amornphimoltham P, et al. Persistent activation of the Akt pathway in head and neck squamous cell carcinoma: a potential target for UCN-01. Clin. Cancer. Res. (2004) 10:4029–4037.[Abstract/Free Full Text]
  18. Chikumi H, et al. Potent activation of RhoA by Galpha q and Gq-coupled receptors. J. Biol. Chem. (2002) 277:27130–27134.[Abstract/Free Full Text]
  19. Servitja JM, et al. Rac1 function is required for Src-induced transformation. Evidence of a role for Tiam1 and Vav2 in Rac activation by Src. J. Biol. Chem. (2003) 278:34339–34346.[Abstract/Free Full Text]
  20. Gutkind JS, et al. Muscarinic acetylcholine receptor subtypes as agonist-dependent oncogenes. Proc. Natl. Acad. Sci. USA (1991) 88:4703–4707.[Abstract/Free Full Text]
  21. He TC, et al. A simplified system for generating recombinant adenoviruses. Proc. Natl. Acad. Sci. USA (1998) 95:2509–2514.[Abstract/Free Full Text]
  22. Pracyk JB, et al. A requirement for the rac1 GTPase in the signal transduction pathway leading to cardiac myocyte hypertrophy. J. Clin. Invest. (1998) 102:929–937.[ISI][Medline]
  23. Boukamp P, et al. Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line. J. Cell Biol. (1988) 106:761–771.[Abstract/Free Full Text]
  24. Sriuranpong V, et al. Epidermal growth factor receptor-independent constitutive activation of STAT3 in head and neck squamous cell carcinoma is mediated by the autocrine/paracrine stimulation of the interleukin 6/gp130 cytokine system. Cancer Res. (2003) 63:2948–2956.[Abstract/Free Full Text]
  25. Bustelo XR. Regulatory and signaling properties of the Vav family. Mol. Cell. Biol. (2000) 20:1461–1477.[Free Full Text]
  26. Crespo P, et al. Phosphotyrosine-dependent activation of Rac-1 GDP/GTP exchange by the vav proto-oncogene product. Nature (1997) 385:169–172.[CrossRef][Medline]
  27. Chang JH, et al. c-Src regulates the simultaneous rearrangement of actin cytoskeleton, p190RhoGAP, and p120RasGAP following epidermal growth factor stimulation. J. Cell Biol. (1995) 130:355–368.[Abstract/Free Full Text]
  28. Liu SY, et al. Overexpression of Rac-1 small GTPase binding protein in oral squamous cell carcinoma. J. Oral Maxillofac. Surg. (2004) 62:702–707.[CrossRef][ISI][Medline]
  29. Kamai T, et al. Overexpression of RhoA, Rac1, and Cdc42 GTPases is associated with progression in testicular cancer. Clin. Cancer Res. (2004) 10:4799–4805.[Abstract/Free Full Text]
  30. Gomez del Pulgar T, et al. Rho GTPase expression in tumourigenesis: evidence for a significant link. Bioessays (2005) 27:602–613.[CrossRef][ISI][Medline]
  31. Das N, et al. ras Gene mutations in oral cancer in eastern India. Oral Oncol. (2000) 36:76–80.[CrossRef][ISI][Medline]
  32. Munirajan AK, et al. Detection of a rare point mutation at codon 59 and relatively high incidence of H-ras mutation in Indian oral cancer. Int. J. Oncol. (1998) 13:971–974.[ISI][Medline]
  33. Xu J, et al. Alterations of p53, cyclin D1, Rb, and H-ras in human oral carcinomas related to tobacco use. Cancer (1998) 83:204–212.[CrossRef][ISI][Medline]
  34. Lambert JM, et al. Tiam1 mediates Ras activation of Rac by a PI(3)K-independent mechanism. Nat. Cell Biol. (2002) 4:621–625.[ISI][Medline]
  35. Bos JL. ras Oncogenes in human cancer: a review. Cancer Res. (1989) 49:4682–4689.[Abstract/Free Full Text]
  36. Patel V, et al. Induction of apoptosis in head-and-neck squamous carcinoma cells by gamma-irradiation and bleomycin is p53-independent. Int. J. Cancer (2000) 88:737–743.[CrossRef][ISI][Medline]
  37. Joneson T, et al. Suppression of Ras-induced apoptosis by the Rac GTPase. Mol. Cell. Biol. (1999) 19:5892–5901.[Abstract/Free Full Text]
  38. Martin GS. Cell signaling and cancer. Cancer Cell (2003) 4:167–174.[CrossRef][Medline]
  39. Jaffe AB, et al. RHO GTPases: biochemistry and biology. Annu. Rev. Cell Dev. Biol. (2005).
Received April 7, 2006; revised December 22, 2006; accepted December 27, 2006.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/6/1145    most recent
bgm008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Patel, V.
Right arrow Articles by Gutkind, J.S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Patel, V.
Right arrow Articles by Gutkind, J.S.
Social Bookmarking
Add to CiteULike