The cytokines of transforming growth factor β (TGF-β) and its superfamily members are potent regulators of tumorigenesis and multiple cellular events. Myostatin is a member of TGF-β superfamily and plays a negative role in the control of cell proliferation and differentiation. We now show that myostatin rapidly activated the extracellular signal–regulated kinase 1/2 (Erk1/2) cascade in C2C12 myoblasts. A more remarkable Erk1/2 activation stimulated by myostatin was observed in differentiating cells than proliferating cells. The results also showed that Ras was the upstream regulator and participated in myostatin-induced Erk1/2 activation because the expression of a dominant-negative Ras prevented myostatin-mediated inhibition of Erk1/2 activation and proliferation. Importantly, the myostatin-suppressed myotube fusion and differentiation marker gene expression were attenuated by blockade of Erk1/2 mitogen-activated protein kinase (MAPK) pathway through pretreatment with MAPK/Erk kinase 1 (MEK1) inhibitor PD98059, indicating that myostatin-stimulated activation of Erk1/2 negatively regulates myogenic differentiation. Activin receptor type IIb (ActRIIb) was previously suggested as the only type II membrane receptor triggering myostatin signaling. In this study, by using synthesized small interfering RNAs and dominant-negative ActRIIb, we show that myostatin failed to stimulate Erk1/2 phosphorylation and could not inhibit myoblast differentiation in ActRIIb-knockdown C2C12 cells, indicating that ActRIIb was required for myostatin-stimulated differentiation suppression. Altogether, our findings in this report provide the first evidence to reveal functional role of the Erk1/2 MAPK pathway in myostatin action as a negative regulator of muscle cell growth. (Cancer Res 2006; 66(3): 1320-6)
- Activin receptor IIb
Transforming growth factor β (TGF-β) and the signaling transducers in TGF-β pathway are potent suppressors during tumorigenesis for their regulatory roles on cell proliferation, differentiation, and apoptosis. Myostatin, also named growth and differentiation factor 8, is a member of the TGF-β superfamily ( 1). Myostatin knockout mice and cattle with genetic mutations in myostatin gene were characterized by a widespread increase in skeletal muscle mass with hyperplasia and hypertrophy ( 1, 2). Similarly, the blockade of endogenous myostatin in mdx mice (a Duchenne muscular dystrophy model) resulted in increases of muscle mass, size, and strength along with a significant decrease in muscle degeneration ( 3, 4). Recently, the reduced level of myostatin protein due to missplicing was identified to be associated with enlarged muscles mass and the development of unusual strength in humans ( 5). On the other hand, systematical administration of exogenous myostatin to adult mice induced severe muscle and fat loss analogous to human cachexia syndromes ( 6). Importantly, the increased myostatin expression was associated with skeletal muscle degeneration–related diseases, such as chronic illnesses, HIV infection, and the aging process ( 7– 9). These genetic and clinical analyses showed that myostatin is a negative regulator for skeletal muscle growth. To date, a number of human and animal disorders are associated with loss of or functionally impaired muscle tissue. Therefore, therapies based on manipulating the biological activities of myostatin would significantly improve the quality of life for these patients ( 10).
Myostatin was able to arrest the cell cycle at the G1 phase of cultured myoblasts and then induced inhibition of proliferation and differentiation. Further investigation showed that the myostatin-stimulated cell cycle arrest was probably due to an increased expression of cyclin-dependent kinase (Cdk) inhibitor p21Waf1,Cip1 and a decreased cyclin E/Cdk2 activity, which further caused the hypophosphorylation of Rb protein ( 11). It was also reported that myostatin inhibited rhabdomyosarcoma cell proliferation through an Rb-independent pathway ( 12). MyoD and myogenin were implicated to participate in myostatin-induced differentiation suppression ( 10). Although the biological roles of myostatin on skeletal muscle development were well established ( 10), the intracellular signaling pathways involved in myostatin function are poorly defined. Like the other members of the TGF-β superfamily, myostatin was shown to elicit its function by regulating a TGF-β-like signaling pathway through activin type IIb (ActRIIb), activin type Ib (ActRIb), or TGF-β type I receptors ( 13, 14), followed by activation of the R-Smads. Recently, p38 mitogen-activated protein kinase (MAPK) was reported to participate in myostatin-regulated signal transduction ( 15). Philip et al. have provided evidence that myostatin activated p38 MAPK through the TGF-β-activated kinase 1 and this activation was independent of Smads. Together, these findings suggest that the complexity of the myostatin signal transduction may be due to the multiple signaling cascades and precise integration of these signaling modules plays an important role in regulating cell proliferation and differentiation.
Myogenesis is a complex and highly organized process that involves a two-step mechanism. First, precursor cells become committed to the myogenic lineage; second, after proliferating, myoblasts begin to withdraw from the cell cycle and undergo terminal differentiation by fusing into multinucleated myotubes ( 16, 17). The myogenic regulatory factors (including MyoD, MRF4, Myf5, and myogenin) and members of the MEF2 family play a key role during embryonic myogenesis and postnatal skeletal muscle growth and repair ( 18, 19). It has been well documented that by negatively regulating myogenic regulatory factors and their coregulators during myogenesis, both basic fibroblast growth factor (bFGF) and TGF-β are potent inhibitors for myoblast differentiation ( 20, 21). Although the Ras/MAPK-extracellular signal–regulated kinase (Erk) kinase 1 (MEK1)/Erk signaling pathway has been proposed to be involved in myogenic suppression induced by bFGF and TGF-β, the molecular basis by which those growth factors elicit their myogenic inhibition remains largely unknown. The MAPK-mediated signal transduction is crucial for muscle development and maintenance of striated muscle homeostasis ( 22, 23). The core component of MAPK module is a set of three acting protein kinases that include Erk, c-Jun NH2-terminal kinases/stress-activated protein kinases, and p38 MAPKs. bFGF and TGF-β suppress myogenesis through a common mechanism of the activation of Ras-MEK-Erk module ( 23).
Myostatin is a member of the TGF-β superfamily whereas TGF-β inhibits myogenic differentiation by activating multiple signaling pathways including Ras/MEK/Erk and p38. Therefore, we ask whether the Ras/MEK/Erk pathway is involved in myostatin signaling and the possible role of this pathway in myostatin-regulated myogenesis. In this report, we provide biochemical evidence to show that myostatin rapidly induced a sustained Erk1/2 activation in C2C12 cells through ActRIIb. Furthermore, ActRIIb was required for inhibition of Erk1/2 activation and differentiation in C2C12 cells in response to myostatin. Our results reveal a new mechanism of myostatin-regulated myogenesis inhibition from the involvement of the Ras/MEK/Erk pathway mediated by ActRIIb.
Materials and Methods
Cell culture and animals. Mouse C2C12 myoblasts were maintained in DMEM supplemented with 4.5 g/L glucose, 4 mmol/L l-glutamine, 10% fetal bovine serum (FBS; Hyclone, Logan, UT), and penicillin/streptomycin at 37°C in a 5% CO2 atmosphere. The myogenic differentiation in vitro was induced by the change of 10% FBS to 2% horse serum (Hyclone) in medium. The exponential proliferating cells or differentiating cells were treated with purified recombinant myostatin as described ( 24, 25) at the final concentration of 500 ng/mL for the indicated time, and the cells treated with the same amount of PBS or with 1 nmol/L bFGF were used as controls. For Erk1/2 MAPK pathway inhibition, cells were pretreated from 30 minutes to 1 hour with 10 μmol/L of MEK1 inhibitor PD98059 (Cell Signaling, Beverly, MA). For animal experiments, 4-week-old C57 female mice were given i.m. injections of 0.1 mg/kg recombinant myostatin or equivalent volume of PBS supplemented with 5% bovine serum albumin (BSA) every 2 days for 2 weeks. The animals were maintained on standard diet in accordance with the Animal Care and Use Committee policy of the school and killed by cervical dislocation. The gastrocnemius was collected separately and stored at −70°C for protein isolation.
Constructs. The retroviral construct carrying dominant-negative Ras17N was a gift from Dr. D.M. Tang (McMaster University, Hamilton, Ontario, Canada). The dominant-negative ActRIIb (truncated form lacking intracellular kinase domain) was PCR amplified with the forward primer (5′-CCGCTCGAGATGACGGCGCCCTGGG-3′) containing an XhoI site and the reverse primer (5′-CCATCGATTCACAGATCCTCTTCTGAGATGAGTTTTTGTTCGTACATCCAAAAGGCCAG-3′) containing a myc tag and a ClaI site, and then cloned into retroviral vector pLNCX according to standard procedure.
Retrovirus production and infection. Retrovirus was produced by transient transfection of retroviral constructs into the Phoenix helper-free retrovirus producer cell line (gift from Dr. Gary Nolan, Stanford University, Stanford, CA) with calcium phosphate method according to the standard protocol (http://www.stanford.edu/group/nolan/). For infection of C2C12 myoblasts, the retroviral supernatant (48 hours posttransfection) was filtered and added into each C2C12 plate with addition of 3 μg/mL polybrene and the cells were placed at 37°C with an overnight incubation for infection.
Antibodies and Western blot. Antibodies against phospho-MEK1, phospho-Erk1/2, phospho-p90RSK, phospho-Elk1, Erk1/2, ActRIIb, and Ras were from Cell Signaling; antitubulin, β-actin, myosin heavy chain (MyHC), MyoD, myogenin, c-myc, and horseradish peroxidase (HRP)–conjugated second antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). For cell lysate preparation, monolayer cells on a 100-mm plate were lysed with 1.2 mL of lysis buffer [50 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 0.5% NP40, 50 mmol/L NaF, 1 mmol/L Na3VO4, 5 mmol/L β-glycerophosphate, 1 mmol/L DTT, 1 mmol/L phenylmethylsulfonyl fluoride]. The lysate was clarified by centrifugation at 14,000 × g for 20 minutes. For protein extraction from tissues, 0.2 g of tissue was rapidly homogenized in 0.5 mL of homogenization buffer [50 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 0.5% NP40, 50 mmol/L NaF, 1 mmol/L Na3VO4, 5 mmol/L β-glycerophosphate, 1 mmol/L DTT, 1 mmol/L phenylmethylsulfonyl fluoride] and the lysate was clarified by centrifugation at 14,000 × g for 20 minutes. Boiled samples with 2× SDS loading buffer were loaded onto 12% polyacrylamide gel and, after electrophoresis, the proteins were transferred onto polyvinylidene difluoride membrane (PALL, East Hills, NY). The resulting blots were blocked with 1% BSA for phospho-protein antibodies and with 5% milk for non-phospho-protein antibodies for 1 hour, and then incubated for another 1 hour at room temperature with primary antibody. The secondary antibody used in the immunoblot was a 1:5,000 dilution of HRP-linked anti–immunoglobulin G (IgG). The enhanced chemiluminescence reagent (Amersham Biosciences, Piscataway, NJ) was used as the substrate for detection and the membrane was exposed to an X-ray film for visualization.
Immunofluorescence. The cells were fixed with 4% (w/v) paraformaldehyde in PBS and then washed thrice with permeabilization buffer (0.3% Triton X-100 in PBS) for 10 minutes each time and blocked with 3% (w/v) BSA (Calbiochem, San Diego, CA) in PBS. The samples were incubated for 2 hours at room temperature with anti-MyHC antibody (1:100) and for 1 hour with FITC-conjugated secondary antibodies against mouse IgG (Santa Cruz Biotechnology). Finally, the cells were washed with PBS and mounted onto microscope glass slides. 4′,6-Diamidino-2-phenylindole (DAPI) staining was done simultaneously to show the position of the nuclei.
Small interfering RNA. The target sequences of double-stranded nucleotides used for small interfering RNA (siRNA) knockdown are GGCTCAGCTCATGAACGAC for ActRIIb (Ambion) and ATCCAATGGCACCGTCAAG for glyceraldehyde-3-phosphate dehydrogenase (GAPDH; RIBOBIO, Guangzhou, Guangdong, China). Cells cultured in a six-well plate (2 × 105 per well) were transfected with 70 μmol/L of siRNAs with Lipofectimin 2000 (Invitrogen, Carlsbad, CA). Forty-eight hours posttransfection, cells were treated with myostatin, bFGF, or PBS according to experimental design.
Proliferation assay. About 2 × 105 cells were seeded in six-well plates and maintained in DMEM culture medium for retrovirus infection and myostatin stimulation. During the last 12 hours of each treatment, 5 μCi of [3H]thymidine were added to the culture medium. The cells were washed thrice with PBS and trypsinized for radioactivity measurement in scintillation vials. The assay was done in three replicates and repeated thrice for statistical analysis.
Myostatin activates Erk1/2 MAPK signaling pathway. It has been recently reported that various MAPK cascades can be activated by members of the TGF-β superfamily in several cell types, indicating that activation of these pathways by TGF-β exerts critical roles in mediating growth, differentiation, and apoptosis ( 26– 28). To determine whether the Erk1/2 MAPK pathway is involved in myostatin signaling, the proliferating and differentiating C2C12 cells were treated with recombinant myostatin for the time periods of 5 minutes to 24 hours, and bFGF was used as a positive reference for activation of Erk1/2 MAPK in myoblasts. As shown in Figs. 1A and 2A , myostatin activated Erk1/2 within 5 minutes and this activation persisted for up to 24 hours of treatment in both proliferating and differentiating C2C12 cells. A more obvious and rapid myostatin-induced Erk1/2 activation was more evident in differentiating cells than in proliferating cells ( Figs. 1A and 2A), which suggests that in low-serum condition, myostatin had more potency to activate Erk1/2. At least 10-fold induction of Erk1/2 phosphorylation by myostatin was detected and the total levels of Erk1/2 proteins remained unchanged. As expected, myostatin-induced Erk1/2 activation was further supported by comparably increased phosphorylation levels of MEK1 and p90RSK, the upstream and downstream molecules in this pathway, in a manner similar with Erk1/2 ( Figs. 1A and 2A).
Myostatin plays a very important role in regulating postnatal muscle growth by controlling the satellite cell quiescence ( 29). We then tested the activation of Erk1/2 by myostatin in adult mouse skeletal muscle tissue in vivo. Four-week-old C57 mice were systematically administrated with myostatin as described in Materials and Methods. The immunoblots showed that myostatin induced an ∼3-fold activation of Erk1 and the upstream kinase MEK1 was also activated by myostatin ( Fig. 2C). Interestingly, instead of p90RSK activation by myostatin observed in the cultured cells, another Erk1/2-regulated downstream transcription factor, Elk1, was activated by myostatin in skeletal muscle tissue ( Fig. 2C). In addition, to understand the specificity of the Erk1/2 MAPK activation stimulated by myostatin, we treated the mouse NIH 3T3 cells with myostatin for 12 hours; the result shown in Fig. 2D indicates that myostatin has no potency to activate Erk1/2 in NIH 3T3 fibroblasts. These data suggest a clear and strong activation of Erk1/2 MAPK pathway by myostatin in both cultured myogenic cells and adult skeletal muscle tissues.
Inhibition of Ras blocks myostatin-induced Erk1/2 activation and proliferation inhibition. To determine whether Ras was required for the ability of myostatin to activate Raf/MEK/Erk cascade, we used a retrovirus-mediated overexpression of the dominant-negative Ras (Ras17N) in C2C12 cells. Compared with the green fluorescent protein (GFP) control virus infection, overexpression of Ras17N dramatically reduced the basal level of Erk1/2 phosphorylation ( Fig. 3A ), indicating that the activity of Ras was functionally blocked by Ras17N in C2C12 cells. Therefore, these cells were used for assaying the requirement of Ras in Erk1/2 activation in response to myostatin. The C2C12 cells overexpressing Ras17N were treated with myostatin or the same volume of PBS as control for 12 hours. Western blot analyses showed that Erk1/2 MAPK cannot be activated by myostatin in the absence of a functional upstream Ras ( Fig. 3B), suggesting that myostatin-mediated Erk1/2 activation was through the Ras/MEK1 signaling pathway. Then, the next question was whether Ras activity was required for myostatin-mediated proliferation inhibition. To answer this question, we measured the relative proliferation rate by using a [3H]thymidine incorporation assay in both C2C12- and Ras17N-overexpressing C2C12 cells in the presence or absence of myostatin, respectively. As shown in Fig. 3C, an almost 40% reduction of [3H]thymidine incorporation was observed in myostatin-treated C2C12 cells than control cells. However, we found that levels of [3H]thymidine incorporation were indistinguishable between myostatin-treated and untreated C2C12 cells, which functionally expressed Ras17N. These results indicated that inactivation of Ras by dominant-negative Ras in C2C12 cells essentially abolished the growth inhibitory effect of myostatin. Taken together, our findings suggest that not only is Ras, as a relative upstream intracellular regulator of Erk1/2, potentially required for Erk1/2 activation by myostatin in C2C12 cells but also that the Ras/MEK1/Erk1/2 cascade plays a critical signaling transduction role in myostatin-mediated proliferation suppression.
Myostatin inhibits myogenic differentiation by down-regulating differentiation related genes via MEK1/Erk1/2 pathway. It has been well established that the Ras/MEK/Erk1/2 activation participates in myogenic differentiation ( 30). Both sustained activation and overactivation of Erk1/2 MAPK are sufficient to repress myogenesis. Myostatin has been shown with inhibitory function on myogenic differentiation. And in this report, we have shown that myostatin activated the Ras/MEK/Erk1/2 pathway in C2C12 cells. Therefore, these two lines of evidence suggest that myostatin-induced Erk1/2 activation may contribute to inhibition of myoblast differentiation. To test this possibility, subconfluent C2C12 cells were cultured in differentiation medium for 24 hours and then treated with myostatin or the same volume of PBS as control for another 3 days in the presence or absence of 10 μmol/L PD98059. Subsequently, total cell lysates were prepared and analyzed by Western blots for total and phosphorylated Erk1/2 MAPK expression as well as myogenic differentiation markers, such as MyHC, MyoD, and myogenin. The results from Fig. 4A indicated that the levels of total Erk1/2 proteins were unchanged in all of the experimental groups. However, we found a significant correlation between Erk1/2 phosphorylation and reduced levels of proteins for MyHC, myogenin, and MyoD in response to myostatin. The results showed that the myostatin-induced reduction of MyHC, MyoD, and myogenin proteins was rescued in the C2C12 cells pretreated with PD98059 ( Fig. 4A), which was also morphologically evidenced by an immunofluorescence staining of MyHC with anti-MyHC antibody ( Fig. 4B). Altogether, our findings indicate that myostatin may inhibit myoblast differentiation by down-regulating MyoD and myogenin gene expression via MEK/Erk1/2 pathway and that MEK/Erk1/2 MAPK may play a very important role in myostatin-mediated myogenic differentiation suppression.
ActRIIb mediates myostatin-stimulated Erk1/2 activation and differentiation suppression. TGF-β receptor type I and ActRIIb are suggested receptors for myostatin action. As the type II receptor, ActRIIb is important and specific for myostatin binding and signaling transduction ( 14). To determine whether myostatin-stimulated Erk1/2 activation was through ActRIIb, the synthesized siRNA duplexes against mouse ActRIIb mRNA were generated in vitro and transfected into C2C12 cells to knock down ActRIIb expression. A siRNA duplex for mouse GAPDH mRNA was used as nonspecific control in this experiment. As shown in Fig. 5A , the protein level of ActRIIb was knocked down dramatically by transfection of the ActRIIb siRNA compared with the GAPDH control. Thus, the Erk1/2 activation was assayed in the C2C12 cells with a loss of ActRIIb protein expression in response to the myostatin and bFGF treatment. We found that Erk1/2 could not be activated by myostatin when the ActRIIb expression was blocked; however, the knockdown of GAPDH expression had no effect on myostatin-stimulated Erk1/2 activation ( Fig. 5A) and Erk1/2 activation induced by bFGF was not affected in the cells. As shown in Fig. 5B, the requirement of ActRIIb for myostatin-stimulated Erk1/2 activation was further confirmed by overexpression of a dominant-negative form of ActRIIb (a truncated mutant of ActRIIb lacking intracellular kinase domain). In addition, the effect of ActRIIb knockdown on myogenic differentiation in response to myostatin was also examined. Both morphologic and biochemical evidence show that the negative regulation of myostatin on myogenic differentiation was significantly attenuated in C2C12 cells lacking ActRIIb ( Fig. 6A and B ). Our findings provide molecular and cellular evidence to indicate that ActRIIb was specific and crucial for myostatin-mediated inhibition of Erk1/2 activation and differentiation in C2C12 cells.
In this study, we determined the involvement of Erk1/2 MAPK cascade in myostatin signaling. Myostatin has a unique effect on skeletal myogenesis: an initial inhibitory role on cell proliferation associated with a cell cycle arrest at G1 phase, followed by myogenic differentiation deficiency associated with a decreased expression of differentiation necessary genes ( 11, 13). All of these effects were thought to be mediated by the membrane receptors ( 14). However, the mechanisms responsible for myostatin intracellular signaling, from early inhibition of proliferation to late repression of differentiation, remain largely unknown. The MAPK signaling pathway is well known as a mediator of a variety of extracellular stimuli and growth signals to the nucleus, and thus, to influence cell proliferation, differentiation, and apoptosis. In light of the diversity of biological responses that can be mediated by the Ras/Raf/MEK1/Erk1/2 cascade, it is not surprising that a growth inhibitor, such as myostatin, might also be able to activate this pathway for its function.
In this report, we showed that myostatin dramatically activated Erk1/2 MAPK in a time-dependent manner both in proliferating and differentiating C2C12 cells. Furthermore, the phosphorylation levels of upstream kinases and downstream effectors in this pathway (such as MEK1, Elk-1, and p90RSK) were also increased following myostatin stimulation. The Erk1/2 MAPK module has been regarded as a key regulator for cell proliferation ( 31). However, previous studies have also suggested that an elevated degree of Erk1/2 activity was sufficient to induce the differentiation inhibition of various cell types, including myogenic and erythroid differentiation ( 30, 32). Although it has been known that activated Ras or Raf inhibits multinuclear myotube fusion through Erk1/2 MAPK signaling, the upstream molecules and downstream targets in muscle cells affected by this pathway have remained unclear for a long time. Recently, MEF2 and myogenin have been considered to contribute to inhibition of differentiation through the MAPK pathway ( 20, 33, 34). In addition to activation of Erk1/2 by myostatin in cultured muscle cells, we also analyzed the influence of myostatin on Erk1/2 pathway in mature skeletal muscle tissue. The same activation of Erk1/2 pathway by myostatin was observed in vivo. Thus, in this study, we present experimental evidence to reveal the activation of Erk1/2 MAPK pathway by myostatin in vitro and in vivo.
To test the functional roles of Erk1/2 activation induced by myostatin in muscle cell proliferation and differentiation, the chemical inhibitor of MEK1 and the dominant-negative form of Ras were used to block the myostatin-activated Erk1/2 in the C2C12 cells. Pretreatment with PD98059 completely inhibited myostatin-stimulated phosphorylation of Erk1/2. In addition, PD98059 was able to rescue the differentiation suppression phenotype induced by myostatin. These data indicate that the inhibitory effects of myostatin on differentiation were mediated through the MEK1/Erk1/2 MAPK signaling pathway. Furthermore, using dominant-negative Ras allowed us to show that the myostatin-induced inhibition of Erk1/2 activation and proliferation was via upstream Ras. It has been recently reported that high level of Raf activity, downstream target of Ras, could inhibit myogenic program through both MEK-dependent and MEK-independent signaling pathways ( 35). Very interestingly, TGF-β expression was up-regulated in differentiation-defective myoblasts caused by high level of Raf activation ( 35). Together with our data from this study, it is most likely that myostatin functions through an autocrine loop to synergize Raf signaling and repress myogenesis in a more effective way.
It has been well established that TGF-β family members can use MAPK signaling pathways to elicit their biological effects and the significance of cross talk between those pathways is becoming a central scheme for understanding the specificity and multiplicity of cellular events induced by a given elicitor ( 36). Recently, Philip et al. ( 15) reported that p38 MAPK was activated by myostatin and Smad was not required for myostatin-activated p38 MAPK pathway in HepG2, A204, and C2C12 cells. Those findings suggest that multiple signaling cascades are activated by myostatin and the cross talk between those pathways plays a significant role in myostatin signaling. Therefore, further investigation of the complexity of the myostatin signal transduction and precise integrations of these signaling modules will shed light on understanding the molecular mechanisms of myostatin function as a negative regulator of muscle growth.
Grant support: National Basic Research Program of China grant 2005CB522400 and the National Natural Science Foundation of China grants 30025027, 30400231, 30429002, and 30330430.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Chengyu Jiang for the suggestions during manuscript preparation and Dr. Damu Tang for the dominant-negative Ras17N plasmid.
Note: W. Yang and Y. Chen contributed equally to this work.
- Received August 25, 2005.
- Revision received October 19, 2005.
- Accepted November 11, 2005.
- ©2006 American Association for Cancer Research.