Smad7 Modulates Epidermal Growth Factor Receptor Turnover through Sequestration of c-Cbl

Abstract

Epidermal growth factor (EGF) regulates various cellular events, including proliferation, differentiation, migration, and tumorigenesis. For the maintenance of homeostasis, EGF signaling should be tightly regulated to prevent the aberrant activation. Smad7 has been known as inhibitory Smad that blocks the signal transduction of transforming growth factor β. In the process of cell proliferation or transformation, Smad7 has been shown the opposite activities as a promoter or suppressor depending on cell types or microenvironments. We found that the overexpression of Smad7 in human HaCaT keratinocyte cells and mouse skin tissues elevated EGF receptor (EGFR) activity by impairing ligand-induced ubiquitination and degradation of activated receptor, which is induced by the E3 ubiquitin ligase c-Cbl. The C-terminal MH2 region but not MH1 region of Smad7 is critical for interaction with c-Cbl to inhibit the ubiquitination of EGFR. Interestingly, wild-type Smad7, but not Smad6 or mutant Smad7, destabilized the EGF-induced complex formation of c-Cbl and EGFR. These data suggest a novel role for Smad7 as a promoter for prolonging the EGFR signal in keratinocyte and skin tissue by reducing its ligand-induced ubiquitination and degradation.

INTRODUCTION

Transforming growth factor β (TGF-β) family cytokines have been found to play diverse roles in regulating growth, differentiation, and the immune response, as well as in development in multiorgan systems. In the classical TGF-β signaling pathway, TGF-β activates Smad signaling via its two receptors, leading to Smad-mediated transcriptional regulation (1, 2). Smad7 negatively regulates TGF-β signaling in both the cytoplasm and the nucleus through the various mechanisms. For example, Smad7 interferes the phosphorylation of R-Smad by blocking the catalytic domain of TGF-β receptor I (3). On the other hand, Smad7 also recruits some of the E3 ubiquitin ligases such as Smurf1/2 or Nedd4-2 to TGF-β receptor to promotes the degradation of receptor proteins via the proteasome pathway (4). In addition to the activities of Smad7 in cytoplasm, Smad7 can also inhibit the TGF-β signaling through disrupting the association of functional R-Smad/Smad4 complexes, as well as binding of the R-Smad complex to DNA in nucleus (5). In addition to its role in the negative regulation of TGF-β signaling, Smad7 modulated other intracellular pathways in both a TGF-β-dependent and a TGF-β-independent manner (6). Due to these abilities, deregulated Smad7 protein expression is able to play a pathogenic role in a variety of human disorders.

Increasing evidence indicates that Smad7 is differently expressed in human cancers, and it could either sustain or restrain cancer cell growth. The Smad7 expression levels were increased in gastric cancer patients who had a poor prognosis (7). Conversely, Smad7 was observed reducing in human hepatocellular carcinomas compared to the adjacent normal tissues (8). In skin cancer, Smad7 levels were also elevated compared to normal epidermis. Smad7 in H-ras-transduced keratinocytes promoted the conversion of benign to malignant epithelial cells and a rapid progression to squamous cell carcinoma in chemically induced skin carcinogenesis model (9). In contrast, another study reported that stable overexpression of Smad7 in human melanoma cells impairs bone metastasis in both in vitro and in vivo (6). Altogether, these data indicated that Smad7 can have both pro- and antitumor actions depending on cancer types.

Epidermal growth factor receptor (EGFR) tyrosine kinase engaged a vast array of signaling pathways to regulate tissue development and homeostasis (10). EGFR signaling was normally induced by ligand binding, leading to receptor dimerization, autophosphorylation, activation of downstream signaling molecules, and cellular events such as proliferation, migration, and differentiation. Ligand binding also triggered the internalization and finally degradation of the activated EGFR, which are the intrinsic mechanisms by which cells attenuate the mitogenic signals (11). To prevent the sustained activation, c-Cbl-mediated ubiquitination has been shown to be essential for regulating these events and ensuring proper degradation of EGFR (12, 13). The c-Cbl protein is an E3 ubiquitin ligase, which has a tyrosine kinase-binding domain to bind with phosphotyrosine residue of activated receptor, as well as nonreceptor tyrosine kinases, a RING finger domain to interact with the ubiquitin-conjugating enzymes, and a highly proline-rich region to make a complex with SH3 domain-containing proteins (14, 15). Upon EGF induction, c-Cbl binds directly to the EGFR via Tyr-1045 and indirectly through the SH3 domain of Grb2 (16, 17). Interaction of c-Cbl and its consequential phosphorylation resulted in the activation of the E3 ligase activity of c-Cbl, recruitment of the ubiquitin-conjugating enzyme Ubc-H7 (18), and mediation of EGFR ubiquitination. Since ErbB family signaling pathways are central to regulating epithelial cell growth, it is not surprising that they are dysregulated during carcinogenesis.

In the present study, we show that Smad7 enhances EGFR signaling in keratinocyte and skin tissue. Mechanistically, Smad7 elevated EGFR activity by sequestering c-Cbl to inhibit the ubiquitination and degradation of EGFR. In this process, the C-terminal region of Smad7 interacted with the proline-rich domain of c-Cbl to block the inhibitory activity of c-Cbl. These data were also validated by using a Smad7 transgenic mouse, which expresses Smad7 protein on skin-specific manner. In conclusion, Smad7 functions as a scaffold protein in various signaling pathways to regulate the many biological processes.

MATERIALS AND METHODS

Cell culture and reagents.

A human embryonic kidney cell line (HEK293T) and keratinocytes (HaCaT) were maintained in Dulbecco modified Eagle medium (DMEM; Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS; Welgene, Daegu, South Korea) and 1% streptomycin-penicillin in a 5% CO₂ atmosphere at 37°C. Human recombinant EGF was purchased from R&D Systems (Minneapolis, MN). Nontargeting small interfering RNA (siRNA) and human Smad7 siRNA (sc-36508) were purchased from Santa Cruz (Santa Cruz, CA). 12-O-Tetradecanoylphorbol-13-acetate (TPA) was purchased from LC Laboratories (Woburn, MA).

Immunoblotting and immunoprecipitation.

Cultured cells or skin tissues were harvested and lysed in a lysis buffer containing 25 mM HEPES (pH 7.5), 150 mM NaCl, 1% Triton X-100, 10% glycerol, 5 mM EDTA, phosphatase inhibitors, and a protease cocktail (Roche, Indianapolis, IN). After 30 min on ice, the debris was removed by centrifugation at 10,000 × g for 10 min. The protein concentrations were determined by the BCA method (Pierce, Rockford, IL). Next, proteins were separated by SDS-PAGE and then transferred onto polyvinylidene difluoride membranes, which were treated with the appropriate primary antibodies. Antibody-bound proteins were detected by a chemiluminescence method according to the manufacturer's protocol (Pierce). For immunoprecipitation, the cell lysates were incubated with the appropriated antibody for overnight at 4°C, followed by incubation with protein A-Excellose-binding bead (Bioprogen, Daejeon, South Korea) for 1.5 h at 4°C. Beads were washed four times with the buffer used for cell solubilization. Immune complexes were then eluted by boiling for 5 min in 2× Tris-glycine-SDS sample buffer, and then extracts were analyzed by immunoblotting as described above.

Quantification of RNA by qRT-PCR.

Total RNA was prepared from cultured cells using TRIzol reagent (Invitrogen) and 2 μg of RNA was reverse transcribed with random hexamers using SuperScript II (Invitrogen). Gene-specific primers were selected by using Primer Express software (Perkin-Elmer Life Sciences, Boston, MA), and primer sequences are provided in Table S1 in the supplemental material. Quantitative real-time PCR was performed using an Applied Biosystems Prism 7900HT sequence detection system and SYBR green Premix Ex-Tag II (TaKaRa, Madison, WI). The results of quantitative reverse transcription-PCR (qRT-PCR) were analyzed by the comparative threshold cycle (C_T) method using cyclophilin genes as housekeeping control genes. All samples were analyzed triplicate and are expressed as means ± the standard deviations.

Lentiviral infection and generation of stable cell lines.

To generate Smad7 wild-type and Smad7 mutant-expressing lentiviruses, Lenti HEK293T packaging cells were plated at 5 × 10⁶ cells/100-mm tissue culture dishes and transfected with pCAG lentiviral vector (green fluorescent protein [GFP], 3×Flag-tagged Smad7-WT or Smad7-MT) by Lipofectamine 2000 reagent (Invitrogen). To knock down c-Cbl in cells, MISSION short hairpin RNA (shRNA) was transfected with packaging DNAs according to the manufacturer's instructions (Sigma-Aldrich, St. Louis, MO). Transfected cells were grown in DMEM containing 10% FBS for 48 h. The conditional medium containing recombinant lentiviruses was collected and filtered through 0.45-μm-pore-size filters. Polybrene (Sigma-Aldrich) was added to a final concentration of 8 μg/ml, and the supernatants were incubated with cells for 12 h. The infection of lentiviral supernatant was repeated three times every 12 h. After infection, the cells were placed in fresh growth medium and cultured as usual. The expression of Smad7 or c-Cbl was confirmed by immunoblotting with specific antibodies.

Cell proliferation assay.

A total of 2 × 10⁴ cells per well were seeded onto 96-well plates in DMEM in triplicate and then stimulated with or without different concentrations of EGF. Cell proliferation was determined for 2 days after EGF treatment using CellTiter-Glo luminescent cell viability assay reagent (Promega, Madison, WI) in accordance with the manufacturer's protocol. The absorbance at 470 nm was measured using SpectraMax L luminescence microplate reader (Molecular Devices, Sunnyvale, CA).

Colony formation assay.

Smad7-WT, Smad7-MT, and control HaCaT cells were seeded at 10⁴ cells per well onto six-well plate in complete DMEM in triplicate. The cells were placed in a humidified CO₂ incubator at 37°C overnight to adhere and then stimulated with or without EGF (20 ng/ml). Subsequently, the cells were allowed to form the colonies for 2 weeks and then fixed and stained with crystal violet staining solution (0.5% crystal violet in 25% methanol) before the images were captured and the colonies were counted.

Silencing of human Smad7.

To knock down the human Smad7 gene, we followed the Thermo Scientific DharmaFECT transfection reagent siRNA transfection protocol (Wilmington, DE). HEK293T cells were transfected with human siRNA Smad7 or nontargeting siRNA. At 48 h after transfection, knockdown of Smad7 was confirmed by qRT-PCR and Western blotting.

Ubiquitination assay.

Ubiquitination assays were performed as previously described (19). Briefly, HEK293T cells were transfected with relevant plasmids. At 36 h after transfection, the cells were preincubated with MG132 for 4 h, stimulated with EGF (20 ng/ml) at different times, and harvested in 1 ml of phosphate-buffered saline (PBS) containing 5 mM N-ethylmaleimide. Noncovalent protein interactions were dissociated using 1% SDS and boiling for 10 min. Samples were diluted 10-fold with lysis buffer (PBS containing 0.5% Triton X-100, 20 mM HEPES [pH 7.4], 150 mM NaCl, 12.5 mM β-glycerol phosphate, 1.5 mM MgCl₂, 10 mM NaF, 2 mM dithiothreitol, 1 mM sodium orthovanadate, 2 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, and protease inhibitor cocktail) and subsequently suspended using a 1-ml syringe. The samples were cleared by centrifugation at 16,000 × g for 10 min. For immunoprecipitation of ubiquitinated proteins, lysates were incubated with protein A/G-agarose beads and with anti-Myc antibody at 4°C for 12 to 16 h. The beads were washed four times with lysis buffer, and immunoprecipitates were fractioned by SDS-PAGE and detected with ubiquitin antibody.

In vitro binding assay.

To make recombinant glutathione S-transferase (GST)-tagged Smad7 or His-tagged c-Cbl proteins, Smad7 and c-Cbl genes were cloned into pGEX vector (Amersham Biosciences, Uppsala, Sweden) and pET28a vector (Novagen, Darmstadt, Germany), respectively. Recombinant GST-Smad7 or His–c-Cbl proteins were induced in E. coli BL21 with IPTG (isopropyl-β-d-thiogalactopyranoside) and purified using glutathione-Sepharose 4B beads (Amersham Biosciences) or Ni-NTA affinity resin (Qiagen, Valencia, CA). Equal amounts (500 μg) of purified His–c-Cbl were incubated with affinity-purified GST or GST-Smad7 in ice-cold binding buffer (50 mM Tris-Cl [pH 7.4], 150 mM NaCl, 0.1% NP-40, and protease inhibitors) at 4°C for overnight with gentle rocking. The beads were washed five times with 1 ml of binding buffer, and bound proteins were eluted in 2× SDS loading buffer and examined by immunoblotting.

Generation of mutant Smad7 transgenic mice.

The generation of mutant Smad7 transgenic mice was performed by a method similar to the method used for wild-type Smad7 transgenic mice (20). Briefly, full-length mouse mutant Smad7 cDNA that contained substitutions of Asn-Pro-Asp-Ser-Arg-Thr to Ala-Pro-Gly-Ala-Arg-Ala at positions 349 to 354 was inserted into the K5 vector. The K5.Smad7 mutant transgene was microinjected into mouse embryos obtained from mating between ICR females and B6D2 males. After birth, transgenic mice were identified by Southern blotting and PCR analysis of tail DNA using Flag-Smad7 primers. The positive PCR product for the Smad7 mutant transgene was ∼320 bp. The primer sequences are shown in Table S1 in the supplemental material. The positive strains were backcrossed with C57BL/6 strain for at least five generations. Mice were housed under pathogen-free conditions with food and water supplied ad libitum. All experiments involving animals were approved in advance by the Institutional Animal Care and Use Committee at Gachon University and were carried out in accordance with Australian code of practice for the Care and Use of Animals for Scientific Purposes.

RESULTS

Promotion of EGFR signaling by Smad7 in keratinocytes.

It has been shown that Smad7 overexpression accelerates skin tumor progression by inhibiting TGF-β signaling and the phosphorylation of EGFR (9). To better understand the role of Smad7 in modulating EGFR signaling, we generated Smad7 wild-type (Smad7-WT) and Smad7 mutant (Smad7-MT) strains in human HaCaT keratinocyte cells which express very low levels of endogenous Smad7. In a previous study, our group reported that Smad7-MT contains point mutations at the TAB2-binding domain, which are responsible for the inhibition of the tumor necrosis factor alpha-induced NF-κB signaling pathway but do not affect the inhibition of TGF-β signaling (21). As shown in Fig. S1 in the supplemental material, both Smad7-WT and Smad7-MT cells significantly suppressed TGF-β signaling compared to control cells.

In order to study whether overexpression of Smad7 would enhance the EGF-induced cell proliferation, we assessed the growth of Smad7-WT, Smad7-MT, and control keratinocytes in the presence or absence of EGF. As shown in Fig. 1A, overexpression of Smad7 enhanced EGF-induced cell growth compared to control cells. Moreover, Smad7-MT blocked the ability of Smad7 to enhance EGF-induced cell proliferation. To further confirm the involvement of Smad7 in activation of EGFR signaling, we next analyzed the mRNA expression of two downstream transcriptional target genes of EGFR signaling, c-Myc and CCND1, in the absence or presence of cetuximab (5 μg/ml), an anti-EGFR antibody (22). Upon stimulation of EGF, Smad7-WT increased the expression level of both c-Myc and CCND1 compared to Smad7-MT and control cells (Fig. 1B). However, the effects of Smad7 on EGFR signaling were alleviated by treatment of cetuximab, suggesting that gene regulation is mainly mediated through EGFR pathway. Because Smad7 could block the senescence of keratinocytes and increase the frequency of transformation in vitro (9), we hypothesized that Smad7-mediated activation of EGFR signaling may have a role in keratinocyte transformation. To test this hypothesis, we performed an anchorage-independent cell transformation assay using Smad7-WT, Smad7-MT, and HaCaT-GFP control cells. As shown in Fig. 1C, Smad7 induced significantly more abundant and larger colonies either with or without EGF stimulation. Moreover, the treatment of cetuximab blocked EGF-induced colony formation in Smad7-WT cells (Fig. 1C and D). Taken together, these data indicated that Smad7 positively regulates EGFR signaling in cell proliferation and transformation.

FIG 1.

Enhanced activity of EGF signaling by Smad7. (A) HaCaT control, Smad7 wild-type (WT), or Smad7 mutant (MT) cells were treated with various concentrations of EGF. After 3 days, cell numbers were measured as described in Materials and Methods. The results are representative of at least three independent experiments. (B) HaCaT control, Smad7-WT, or Smad7-MT cells were incubated with or without cetuximab (Cetu; 5 μg/ml) for 2 h after treatment with EGF (20 ng/ml) for 4 h. The mRNA levels of c-Myc and CCND1 were examined by qRT-PCR. The expression level of each mRNA was normalized based on cyclophilin mRNA expression. The results are representative of at least three independent experiments. *, P < 0.05 (Student t test). (C) HaCaT control, Smad7-WT, or Smad7-MT cells were plated and stimulated with EGF and cextuximab as described in Materials and Methods. After 2 weeks, the colonies that formed were counted and compared between samples. Representative phase-contrast images are shown. *, P < 0.05 (Student t test).

Increased stability of EGFR by Smad7.

Because Smad7 could enhance the activity of EGFR signaling pathway, we first tested the transcriptional levels of EGFR in HaCaT-control, Smad7-WT, or Smad7-MT cells with or without treatment of EGF. However, we did not find any statistically significant differences between these cell lines (Fig. 2A). This means that Smad7 does not regulate the EGF pathway at the transcriptional level of its receptor.

FIG 2.

Increased stability of EGFR protein by Smad7. (A) To check the effect of Smad7 on the expression level of EGFR mRNA, HaCaT control, Smad7-WT, or Smad7-MT cells were treated with EGF (20 ng/ml) for 4 h. EGFR mRNA expression levels were detected by qRT-PCR. The mRNA expression levels of EGFR were normalized based on cyclophilin mRNA expression. The results are representative of at least three independent experiments. (B) To examine the protective activity of Smad7 on stability of the EGFR protein, control and Smad7-WT- or Smad7-MT-overexpressing HaCaT cells were starved for 24 h to allow maximum expression of EGFR. The cells were pretreated with cycloheximide (CHX) for 1 h after stimulation with EGF (20 ng/ml) for various times or left untreated. Immunoblot analyses were performed with the relevant antibodies. The relative band intensity of EGFR was measured with densitometry. (C) To see the effect of Smad7 on EGFR signaling, control and Smad7-WT- or Smad7-MT-overexpressing HaCaT cells were starved for 24 h and stimulated with EGF (20 ng/ml) for 15, 30, and 60 min or left untreated. Immunoblot analyses were performed with the relevant antibodies. The β-actin was detected as loading control. Representative gels of at least three different experiments are shown. The relative band intensity of phosphorylated or total EGFR was measured using densitometry.

We next examined the effect of Smad7 on ligand-induced EGFR activation and subsequent signal termination. Cells were starved serum overnight and pretreated with cycloheximide (10 μg/ml), an inhibitor of new protein synthesis; the cells were then stimulated with EGF (20 ng/ml) at various times. As shown in Fig. 2B, EGFR protein was rapidly degraded from 2 h and almost disappeared after 4 h after treatment of EGF in control and Smad7-MT cells. Interestingly, the overexpression of Smad7-WT increased the half-life of EGFR after 4 h. Consistent with the increased stability of the receptor protein, the phosphorylation of EGFR and activation of downstream kinases were also enhanced in Smad7-WT cells (Fig. 2C). Interestingly, we also observed an increased basal level of total and phosphorylated EGFR in Smad7-WT cells compared to control and Smad7-MT cells. Overall, these data suggested that wild-type Smad7 significantly suppressed the EGF-induced EGFR degradation, promoted the expression of EGFR at the posttranscriptional level, and resulted in the activation and prolongation of its downstream signaling.

Inhibitory activity of Smad7 on ubiquitination of EGFR.

Binding of EGF to its cognate receptor stimulates the intrinsic tyrosine kinase activity and culminates in cell fate decisions. Importantly, these positively acting processes are coupled to a variety of negatively acting feedback loops (23). One of the negative regulations is governed through ubiquitin-dependent receptor degradation. To examine the effect of Smad7 on ligand-induced ubiquitination and degradation of EGFR, we cotransfected Flag-Smad7-WT or Smad7-MT with Myc-tagged EGFR and hemagglutinin (HA)-tagged ubiquitin constructs into HEK293T cells. To exclude nonspecific binding to ubiquitin, cell lysates were prepared in lysis buffer containing 1% SDS, and membranes containing SDS-PAGE-separated immunoprecipitates were treated with denaturation buffer containing 6 M guanidine chloride (19). As shown in Fig. 3A, ubiquitination and degradation of EGFR were increased after EGF treatment in control cells. In contrast, the overexpression of wild-type Smad7 markedly suppressed the EGF-induced EGFR ubiquitination and degradation. As expected, Smad7-MT rescued the EGFR ubiquitination and degradation by stimulation of EGF. Consistent with the modulation of EGFR degradation, downstream signaling such as phosphorylated ERK and Akt was significantly increased in Smad7-WT-overexpressing cells compared to control or Smad7-MT-expressing cells.

FIG 3.

Inhibitory activity of Smad7 on the ubiquitination of EGFR. (A) To check the differences in ubiquitination pattern of EGFR, HEK293T cells were transiently transfected with control vector, Flag-Smad7-WT, or Flag-Smad7-MT in combination with Myc-EGFR and HA-ubiquitin (Ub). At 36 h posttransfection, the cells were starved and stimulated with EGF (20 ng/ml) for 30, 60, and 90 min. Cell lysates were immunoprecipitated with antibodies specific for Myc (Myc-EGFR) and monitored by immunoblotting with anti-HA (ubiquitin) or anti-Myc antibodies. Total cell lysates (TCL) were immunoblotted with the relative antibodies. (B) To confirm the involvement of Smad7 on ubiquitination of EGFR, Smad7-overexpressing HEK293T cells were transfected with siRNA targeting Smad7 or scramble oligonucleotide. After 8 h, the cell lines were cotransfected with Myc-EGFR and HA-ubiquitin. The cells were then starved and stimulated with EGF (20 ng/ml) for 15 and 30 min. The cell lysates were immunoprecipitated with antibodies specific for Myc-EGFR and immunoblotted with anti-HA (ubiquitin) or antiphosphotyrosine (p-tyrosine) antibodies. The total cell lysates were immunoblotted with the relative antibodies. Representative gels from at least three different experiments are shown.

Conversely, we examined the ubiquitination and degradation of EGFR using siRNA against Smad7. In agreement with previous results, expression of Smad7 blocked the ubiquitination of EGFR and enhanced the phosphorylation of receptor in response to EGF treatment (Fig. 3B). However, knockdown of Smad7 by siRNA restored the EGFR ubiquitination and degradation and inhibited downstream signaling after treatment of EGF. These results suggested that Smad7 inhibits the ligand-induced EGFR degradation through reducing its ubiquitination process.

Blocking of EGFR/c-Cbl complex formation by Smad7.

Ubiquitination of EGFR is mediated by c-Cbl protein, which acts as an E3 ubiquitin ligase after stimulation with ligand. C-Cbl can bind directly to pY¹⁰⁴⁵ or indirectly to pY¹⁰⁶⁸ sites of EGFR and then ubiquitinates the activated receptor (24). Because Smad7 reduced the EGF-induced EGFR ubiquitination, we hypothesized that Smad7 might have a negative effect on EGFR association with c-Cbl protein. To determine the effect of Smad7 on the formation of c-Cbl/EGFR complex, we transfected Smad7 construct with EGFR, c-Cbl and ubiquitin plasmids into HEK293T cells and then stimulated with EGF (20 ng/ml) for 10 min. Equal amounts of proteins were immunoprecipitated with anti-EGFR antibody and interaction between c-Cbl and EGFR was analyzed with cognate antibodies. In control cells, treatment of EGF showed rapid recruitment of c-Cbl to EGFR (Fig. 4A). Large amounts of c-Cbl were interacted with EGFR in control cells upon treatment of EGF, whereas c-Cbl/EGFR interaction was significantly reduced in Smad7-WT transfected cells. Interestingly, the expression of Smad7 mutant did not block the formation of c-Cbl/EGFR complex. In consistent with the disassembly of c-Cbl/EGFR complex, EGFR ubiquitination was almost inhibited in Smad7-WT-expressing cells.

FIG 4.

Blocking of EGFR/c-Cbl complex formation by Smad7. (A) To identify the molecular mechanism of EGFR stability, HEK293T cells were transiently cotransfected with control vector, Flag-Smad7-WT, or Flag-Smad7-MT and with Myc-EGFR, Flag-c-Cbl, or HA-ubiquitin. At 36 h posttransfection, the cells were starved and pretreated with MG132 for 4 h and stimulated with EGF (20 ng/ml) for 10 min. The cell lysates were immunoprecipitated with antibody specific for Myc-EGFR, followed by immunoblotting with anti-Flag (c-Cbl), anti-HA (ubiquitin) and anti-Myc antibodies. The total cell lysates were immunoblotted with anti-Flag and anti-Myc antibodies to validate the expression of each protein. (B) Stably Smad7-expressing HEK293T cells were infected with lentiviral shRNA against c-Cbl and transfected with siRNA targeting Smad7 or scramble oligonucleotide. At 8 h posttransfection, the cells were cotransfected with Myc-EGFR and HA-ubiquitin and stimulated with EGF (20 ng/ml) for 30 min. The cell lysates were immunoprecipitated with antibodies specific for Myc-EGFR, followed by immunoblotting with anti-HA (ubiquitin) or anti-p-tyrosine antibodies. The total cell lysates were immunoblotted with relative antibodies. (C) To confirm the effect of endogenous Smad7, Smad7^+/+ and Smad7^−/− MEFs were starved and incubated for 10 min with EGF. The total EGFR was immunoprecipitated with anti-EGFR antibody, and ubiquitinated protein was detected with antiubiquitin antibody. The total cell lysates were checked with the indicated antibodies to confirm the expression of each protein.

To validate the involvement of c-Cbl in stability of EGFR, c-Cbl was suppressed with shRNA to restore the level of EGFR, which was inhibited by siRNA against Smad7. As shown in Fig. 4B, knockdown of c-Cbl significantly recovered the level of EGFR which is reduced by siSmad7. These results strongly suggest that c-Cbl is involved in the degradation of EGFR and regulated by Smad7.

To confirm the importance of endogenous Smad7 on regulation of EGFR, we analyzed the EGF-induced EGFR ubiquitination and association of c-Cbl and EGFR in Smad7 knockout mouse embryonic fibroblasts (MEFs) upon stimulation of EGF. As shown in Fig. 4C, treatment with EGF for 10 min rapidly induced the interaction of endogenous c-Cbl with EGFR and the ubiquitination of EGFR in Smad7-WT MEFs compared to the nontreated control. However, the deficiency of endogenous Smad7 increased the EGF-induced EGFR ubiquitination and interaction of c-Cbl with EGFR compared to those in Smad7-WT MEFs. Together, these results support the hypothesis that Smad7 plays a role in negative regulation of EGFR ubiquitination through the modulation of c-Cbl/EGFR complex formation.

Binding domain mapping for Smad7 and c-Cbl.

To study the inhibitory mechanism of Smad7 for c-Cbl/EGFR complex formation, we tested the interaction of Smad7 with c-Cbl. To investigate the interaction between Smad7 and c-Cbl, we performed an immunoprecipitation experiment using GST-fused Smad7 constructs. GST-fused Smad7-WT, Smad7-MT, and N- and C-terminal constructs were cotransfected with c-Cbl into HEK293T cells. As shown in Fig. 5A, c-Cbl was immunoprecipitated with Smad7-WT, but not with Smad7-MT. Interestingly, the C-terminal MH2 region of Smad7 (Smad7 C; amino acids 259 to 426) was able to interact with c-Cbl but not the MH1 region of Smad7 (Smad7 N; amino acids 1 to 258). To confirm the direct interaction of Smad7 with c-Cbl, an in vitro binding assay was performed using recombinant Smad7 and c-Cbl proteins. As shown in Fig. 5B, GST-Smad7 but not GST alone interacted with His-tagged c-Cbl. These results suggest that Smad7 directly interacts with c-Cbl.

FIG 5.

Binding domain mapping of Smad7 and c-Cbl. (A) To map the binding domain of Smad7, plasmids encoding GST-fused Smad7-WT, Smad7-MT, N-terminal (N), or C-terminal (C) fragments were transfected with Myc-tagged c-Cbl in to HEK293T cells. The cell lysates were immunoprecipitated with glutathione (GSH) beads, and interacted proteins were detected with anti-Myc and GST antibodies. (B) Recombinant GST or GST-Smad7 was incubated with equal amounts of His–c-Cbl fusion protein. After pulldown with GSH-beads, the bound proteins were immunoblotted with anti-His antibody. GST and GST-Smad7 were detected by immunoblotting with anti-GST antibody. (C) To confirm the specificity of Smad7 on interaction with c-Cbl, plasmids encoding GST-fused Smad7 or Smad6 were transfected with Myc-tagged c-Cbl into HEK293T cells. Cell lysates were immunoprecipitated with GSH beads, and interacted proteins were detected with anti-Myc and GST antibodies. (D) To map the binding site of c-Cbl, various deletion constructs were cotransfected into HEK293T cells with Smad7. Coimmunoprecipitation assay was performed with anti-Myc (c-Cbl) antibody, and immunoprecipitated proteins were detected with anti-Flag antibody. TKB, tyrosine kinase-binding domain; L, linker domain; RF, RING finger type E3 ligase domain; PRO, proline-rich domain; UBA, ubiquitin-associated domain.

On the other hand, we also observed that Smad7, but not Smad6, interacted with c-Cbl (Fig. 5C). This result is correlated with the findings of a previous study which indicated that the overexpression of Smad7 but not of Smad6 enhanced the malignant conversion in ras-initiated squamous cell carcinoma (9). In order to exclude the possibility of direct interaction of Smad7 with EGFR, we also performed an immunoprecipitation experiment with EGFR and Smad7. As shown in Fig. S2 in the supplemental material, the interaction of EGFR and Smad7 was not detected, but c-Cbl immunoprecipitated with EGFR. These data indicated that the inhibitory activity of Smad7 on EGF-induced EGFR ubiquitination is mediated through direct interaction of the Smad7 C-terminal domain with c-Cbl by reducing c-Cbl/EGFR complex formation.

We next determined which region of c-Cbl is responsible for the interaction with Smad7. We tested the interaction of several deletion mutants of c-Cbl with Flag-Smad7. As shown in Fig. 5D, the proline-rich domain contained an important site for the interaction between Smad7 and c-Cbl. The c-Cbl protein also contains a RING finger, which is important for the transfer of ubiquitin moieties from E2 enzymes to the target molecules, and another domain which is often deleted or mutated in oncogenic c-Cbl variants (25, 26). E3 ligase inactive mutations on linker region of c-Cbl such as Q367P, Y371S, or TKB domain 70Z mutation have been reported in carcinogenesis (27, 28). In a binding assay, these mutant c-Cbl proteins showed a similar interaction pattern with Smad7 compared to the wild-type protein (see Fig. S3 in the supplemental material). These results demonstrated that the C-terminal region of Smad7 directly interacted with the proline-rich domain of the c-Cbl protein.

In vivo validation for Smad7 activity on EGF signaling.

To confirm the positive role of Smad7 in EGFR signaling in vivo, we tested the activation of EGFR signaling in TPA-treated mouse skin. It has been well known that TPA could activate EGFR signaling and downstream signaling pathways (10, 29). For this purpose, we generated a mouse strain that expresses Smad7-MT in a skin-specific manner based on X.-J. Wang's method (20) (see Fig. S4A in the supplemental material). We confirmed the skin-specific expression of Smad7 using RT-PCR and Western blotting (see Fig. S4B and C in the supplemental material). Smad7-WT mice, Smad7-MT transgenic mice, and littermate control C57BL/6 mice were treated with TPA for the indicated times. As shown in Fig. 6A, the phosphorylation of EGFR was significantly increased after 2 h of TPA treatment in mouse skin. After 6 and 24 h of TPA treatment, the activation of EGFR signaling was sustained in only Smad7-WT mouse skin and not in control and Smad7-MT mice. Consistently, the half-life of the total EGFR protein was extended from 6 h in TPA-treated nontransgenic skin to up to 24 h post-TPA treatment in Smad7-WT mice. The activation of downstream kinases such as AKT was markedly higher in TPA-treated Smad7-WT mouse skin compared to control and Smad7-MT mice.

FIG 6.

In vivo validation of Smad7 activity on EGF signaling. (A) Groups of mice expressing Smad7-WT or MT on the skin were treated with TPA for the indicated times. The protein lysates were prepared from skin tissues taken from the backs of mice. Immunodetection of the total EGFR, the p-EGFR, the downstream effectors, and the p-AKT was performed using specific antibodies. β-Actin was detected as a loading control. (B) To check the effect of Smad7 on stability or ubiquitination of EGFR, skin lysates after TPA treatment were immunoprecipitated with antibodies specific for EGFR, followed by immunoblotting with antiubiquitin, antiphosphotyrosine, and anti-EGFR antibodies. The total cell lysates were immunoblotted with relative antibodies. (C) To confirm the in vivo interaction of Smad7 and c-Cbl, protein lysates after TPA treatment were immunoprecipitated with antibody specific for Smad7, followed by immunoblotting with anti-c-Cbl or anti-Smad7 antibodies. The total cell lysates were immunoblotted with the relative antibodies. Representative gels from at least three different experiments are shown.

To further validate the in vivo role of Smad7 in regulating of EGFR ubiquitination and interaction between Smad7 and c-Cbl, we performed binding assays using protein lysates of mice skins after treatment of TPA. As shown in Fig. 6B, transgenic expression of Smad7-WT significantly suppressed TPA-induced EGFR ubiquitination compared to Smad7-MT and control mice. We also observed the interaction between Smad7 and endogenous c-Cbl only in Smad7-WT mouse skin (Fig. 6C). These results provided strong evidence for critical role of Smad7 in the positive regulation of EGFR signaling by modulating c-Cbl activity in skin.

DISCUSSION

We found a novel function of Smad7 in promoting the EGF signaling through the inhibition of ligand-induced EGFR degradation by reducing its ubiquitination process (Fig. 7). Ubiquitination of EGFR has been shown to be a critical step in EGFR degradation, where c-Cbl plays a key role as E3 ubiquitin ligase (30). During ligand-mediated EGFR activation, c-Cbl is phosphorylated and translocated to the cell membrane, where it binds the activated form of EGFR, acts as ubiquitin ligase, and mediates EGFR degradation to stop the EGF-induced signal. We have shown here that Smad7 directly interacts with activated c-Cbl, reducing the c-Cbl/EGFR interaction complex. As a result, the protein level of EGFR remained at a sufficiently high level to transduce the signal even at late times. Interestingly, Smad7 did not affect the basal level of c-Cbl (Fig. 6C) and did not modulate EGF signaling through interaction with receptor (see Fig. S2 in the supplemental material). Consequently, enhanced EGFR signaling induces the promotion of its downstream signals, which contributes to cell proliferation and transformation.

FIG 7.

Schematic diagram for the promoting activity of Smad7 on EGF signaling. In low levels or the absence of Smad7, c-Cbl upon EGF stimulation binds directly or indirectly to activated EGFR. The association of c-Cbl with EGFR results in activation of the E3 ligase activity of c-Cbl, which then targets EGFR for ubiquitination and degradation. In high levels or the presence of Smad7, c-Cbl proteins are sequestered far away from EGFR by the interaction between Smad7 and c-Cbl. The disruption of c-Cbl/EGFR association could enhance the activation of EGFR signaling, leading to the activation of its downstream signals, which contributes to cell proliferation and tumor formation.

Consistent with the impairment of the c-Cbl/EGFR interaction complex, Smad7 was able to reduce EGFR ubiquitination and degradation. During EGFR activation, Tyr1045 is autophosphorylated (31), and this creates a second EGFR docking site to which Cbl binds through its N-terminal TKB domain (32). This enables several Cbl molecules to be recruited to active EGFR complexes and to mediate ubiquitination and subsequent internalization of the oligomeric receptors (15). In Fig. 5D, we suggest that the proline-rich domain, not the N-terminal TKB domain of c-Cbl, is responsible for the interaction with Smad7. It is possible that Smad7 may not directly compete with EGFR for interaction with c-Cbl at the N-terminal TKB domain but indirectly sequesters c-Cbl far away from the EGFR. On the other hand, the proline-rich region is essential for the interaction of c-Cbl with the adaptor proteins (such as Grb2 and FRs2α) required to maintain a stable attachment between c-Cbl and its substrate, with proteins involved in the endocytosis of target receptors (33, 34). We hypothesized that the binding of Smad7 to the proline-rich domain of c-Cbl could impact on the interaction of c-Cbl with its substrate involved in the endocytosis pathway, reducing the adaptor function of c-Cbl. We have also shown that the wild-type MH2 (C-terminal) domain of Smad7 is important for the interaction with c-Cbl (Fig. 5A). A mutation at C-terminal of Smad7 significantly blocked the interaction of Smad7 and c-Cbl and thus had no effect on EGFR signaling or c-Cbl-mediated ubiquitination (Fig. 4A and 5A). Previously, we reported that C-terminal residues are important for the interaction of Smad7 with both TAB2 and TAB3 (21). Based on these results, we hypothesized that amino acids of C-terminal region are very important for the interaction of Smad7 with partner proteins. Moreover, the mutation of amino acids induced structural change of Smad7 to interrupt the interaction with other proteins. However, mutant Smad7 was still functional in terms of inhibiting activity for TGF-β signaling (see Fig. S1 in the supplemental material). These results indicated that the residues between 325 and 355 of Smad7 are very dynamic and important for the regulatory function of Smad7 in many signaling pathways.

Because EGF signaling is an important pathway for cell proliferation and transformation, understanding in detail the regulatory mechanism is very useful for artificial intervention for dysregulated signaling in human diseases. Based on our study, Smad7 and c-Cbl can be used as targets to control the human diseases that are mediated by elevated EGFR signaling. Also, mouse strains expressing Smad7-WT and Smad7-MT can be used as model systems for screening therapeutic candidates used to treat EGF-related human diseases.