This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yamagata, H. Articles by Okazaki, K. Search for Related Content PubMed PubMed Citation Articles by Yamagata, H. Articles by Okazaki, K.

Acceleration of Smad2 and Smad3 Phosphorylation via c-Jun NH₂-Terminal Kinase during Human Colorectal Carcinogenesis

¹ Third Department of Internal Medicine, ² Department of Surgical Pathology, and ³ Department of Surgery, Kansai Medical University, Osaka, Japan

Requests for reprints: Koichi Matsuzaki, Third Department of Internal Medicine, Kansai Medical University, 10-15 Fumizonocho, Moriguchi, Osaka 570-8506, Japan. Phone: 81-6-6992-1000, ext. 45207; Fax: 81-6-6996-4874; E-mail: matsuzak{at}takii.kmu.ac.jp

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Conversion of normal epithelial cells to tumors is associated with a shift in transforming growth factor-ß (TGF-ß) function: reduction of tumor suppressor activity and increase of oncogenic activity. However, specific mechanisms of this functional alteration during human colorectal carcinogenesis remain to be elucidated. TGF-ß signaling involves Smad2/3 phosphorylated at linker regions (pSmad2/3L) and COOH-terminal regions (pSmad2/3C). Using antibodies specific to each phosphorylation site, we herein showed that Smad2 and Smad3 were phosphorylated at COOH-terminal regions but not at linker regions in normal colorectal epithelial cells and that pSmad2/3C were located predominantly in their nuclei. However, the linker regions of Smad2 and Smad3 were phosphorylated in 31 sporadic colorectal adenocarcinomas. In particular, late-stage invasive and metastatic cancers typically showed a high degree of phosphorylation of Smad2/3L. Their extent of phosphorylation in 11 adenomas was intermediate between those in normal epithelial cells and adenocarcinomas. Whereas pSmad2L remained in the cytoplasm, pSmad3L was located exclusively in the nuclei of Ki-67-immunoreactive adenocarcinomas. In contrast, pSmad3C gradually decreased as the tumor stage progressed. Activated c-Jun NH₂-terminal kinase in cancers could directly phosphorylate Smad2/3L. Although Mad homology 2 region sequencing in the Smad4 gene revealed a G/A substitution at codon 361 in one adenocarcinoma, the mutation did not correlate with phosphorylation. No mutations in the type II TGF-ß receptor and Smad2 genes were observed in the tumors. In conclusion, pSmad3C, which favors tumor suppressor activity of TGF-ß, was found to decrease, whereas c-Jun NH₂-terminal kinase tended to induce the phosphorylation of Smad2/3L in human colorectal adenoma-carcinoma sequence.

Key Words: colorectal carcinogenesis • TGF-ß • Smad • JNK • TGF-ß receptor

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The concept of adenoma-carcinoma sequence is widely accepted concerning sporadic colorectal carcinogenesis (1). On the other hand, emerging evidence indicates that transforming growth factor-ß (TGF-ß) signaling participates not only in tumor suppressor activities such as growth inhibition and apoptosis but also in oncogenic processes such as growth stimulation, increases in motility, invasion, and metastasis (2). Conversion of nontumorigenic phenotype of human colonic adenoma cell lines to a tumorigenic phenotype is associated with reduced tumor suppressor activities and increased oncogenic activities of TGF-ß (3). Modulation of growth factor effects can be achieved by various mechanisms, including changes in ligand concentration, activation of latent forms of the ligand, modulation of number and affinity of receptors, and alterations in postreceptor pathways.

Progress over the past several years has disclosed some details of how TGF-ß elicits its responses. TGF-ß signaling is initiated when this ligand induces formation of a heteromeric complex composed of TGF-ß receptor type I (TßRI) and type II (TßRII; ref. 4). This allows TßRII to phosphorylate TßRI, which then transmits the signal through phosphorylation of receptor-regulated Smads such as Smad2 and Smad3 (5). Both of those Smads are directly phosphorylated at COOH-terminal SXS regions by TßRI and then undergo formation of heteromeric complexes with Smad4 (6). Activated Smad complexes then are translocated into the nucleus, where they regulate expression of target genes both by direct DNA binding and through interaction with other transcription factors, coactivators, and corepressors (7). Smads contain two highly conserved domains, the Mad homology 1 (MH1) and 2 (MH2) domains, which are connected by interposed linker regions (8). Although their MH1 domains can interact with DNA, the MH2 domains are endowed with transcriptional activation properties (6).

The TßRI/Smad pathway is widely represented in most cell types and tissues studied to date, and additional pathways are activated following cell stimulation by TGF-ß in specific contexts. The most prominent pathways are mediated by the mitogen-activated protein kinase (MAPK) family, which consists of the extracellular signal-regulated protein kinase pathway and two stress-activated protein kinase pathways: c-Jun NH₂-terminal kinase (JNK) and p38 pathways (9). TGF-ß induces activation of MAPK pathways through the upstream mediators Ras, RhoA, PP2A, and TGF-ß-activated kinase 1 (10). To investigate the roles of Smad2 and Smad3 phosphorylation in TGF-ß signal transduction, we developed four types of polyclonal antibodies (Abs) in our laboratory that specifically recognized the phosphorylated linker regions and the phosphorylated COOH-terminal SXS regions in Smad2 and Smad3 (11).⁴ Studies using the Abs showed that TGF-ß signal phosphorylated Smad2 and Smad3 not only at COOH-terminal SXS regions but also at linker regions. Smad2 or Smad3 phosphorylated at linker regions or COOH-terminal regions existed as separate molecules with different functions and transmitted distinct signals. In particular, JNK and/or p38 MAPK activated on TGF-ß treatment could directly phosphorylate Smad2 and Smad3 at linker regions. Moreover, TGF-ß treatment led to an increase in plasminogen activator inhibitor type 1 transcriptional activity through pSmad3L (11).

Whereas JNK signals can modify TßRI-mediated signaling in vitro, proof has been lacking that this event occurs in vivo. Additionally, the cellular distribution of phosphorylated Smad2 and Smad3 has not been studied in human tissues. Without Abs to selectively distinguish phosphorylation sites in Smad2 and Smad3, determination of phosphorylation sites and investigation of their distinct phosphorylated domain-mediated signals in vivo has been difficult. Using domain-specific phospho-Smad2/3 Abs, we carried out the present study to elucidate how pSmad2/3-mediated signals changed during human colorectal carcinogenesis. Our results indicated that JNK activation occurred during progression to malignancy, accompanied by apparent Smad2/3 phosphorylation at linker regions in situ. Moreover, JNK in cancerous tissues could directly phosphorylate Smad2 and Smad3 at linker regions. In particular, Smad3 phosphorylated at the linker region was localized predominantly to cell nuclei in actively growing Ki-67-immunoreactive adenocarcinoma with distant metastasis. Collectively, Smad2 and Smad3 phosphorylation at linker regions could play an important role in transmitting JNK-mediated signals in human sporadic colorectal cancer.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Tumor Specimens. According to criteria of the Japanese Classification of Colorectal Carcinoma (12), the principal histologic distinction is between benign epithelial tumors including adenoma and malignant epithelial tumors including adenocarcinoma. Pathology records and histologic slides involving diagnoses of either adenoma or adenocarcinoma made from 2000 to 2003 in the Department of Surgical Pathology at Kansai Medical University Hospital were reviewed independently by two pathologists with an interest in gastrointestinal neoplasia (Y.U. and N.S.).

This study was approved by the Ethics Committee of Kansai Medical University. Seventeen men and 14 women, ages 47 to 88 years at diagnosis, were included in the advanced cancer group, whereas 9 men and 2 women, ages 52 to 78 years at diagnosis, constituted the adenoma group. Colorectal adenomas included tumors with mild (n = 2), moderate (n = 7), and severe (n = 2) atypia. Colorectal cancers included tumors classified histologically as well-differentiated adenocarcinoma (n = 20), moderately differentiated adenocarcinoma (n = 10), and mucinous adenocarcinoma (n = 1). Sites of cancers included ascending colon (n = 2), transverse colon (n = 3), descending colon (n = 2), sigmoid colon (n = 10), and rectum (n = 14). Cancers were staged at the time of surgery by standard criteria for tumor-nodes-metastasis–based staging using the unified international colorectal cancer staging classification (13).

Reverse Transcription-PCR. Isolation of RNA, RT, and PCR for TßRII, Smad2, and Smad4 was done in accordance with the manufacturer's protocol (Toyobo, Osaka, Japan). Two sets of primers were used based on the TßRII mRNA sequence (Genbank accession no. M85079). We designed one pair of primers (TßRII1) to contain the (A)₁₀ microsatellite sequence -between nucleotides 709 and 718 (14). The other primer (TßRII2) was designed to contain the (GT)₃ microsatellite sequence between nucleotides 1,931 and 1,936 (15). To detect any mutations of Smad2 and Smad4 genes in their MH2 domains (16, 17), other primers were designed based on their mRNA sequences (Genbank accession nos. AF027964 and NM005359). The primer sequences were as follows: TßRII1 forward (405/422) 5'-CGGAATTCGCATGAAGGACAACGTGTTG-3' and TßRII1 reverse (795/812) 5'-ATAGAATTCGAGCTATTTGGTAGTGTTTAGGGA-3', TßRII2 forward (1,792/1,811) 5'-AAGGATCCATCCATCCCACCGCACGTTCAG-3' and TßRII2 reverse (2,019/2,042) 5'-ATACTCGAGGTCAGGATTGCTGGTGTT-3', Smad2 forward (961/979), 5'-GAATTCGGTTGGAGAAACCTTCCAT-3' and Smad2 reverse (1,495/1,510), 5'-GAACTCGAGTTATGACATGCTTGAG-3', and Smad4 forward (1,149/1,167) 5'-GGATCCAATGAGCTTGCATTCCAGC-3' and Smad4 reverse (1,924/1,943) 5'-TTTTGTAGTCCACCATCCTG-3'.

Domain-Specific Abs against the Phosphorylated Smad2 and Smad3. Polyclonal anti-phospho-Smad2 and anti-phospho-Smad3 Abs [ pSmad2L (Ser^249/254), pSmad2C (Ser^465/467), pSmad3L (Ser^207/212), and pSmad3C (Ser^423/425)] were raised against the phosphorylated linker regions and COOH-terminal regions of Smad2 and Smad3 by immunization of rabbits with synthetic peptides.⁴ The relevant antisera were affinity purified with the phosphorylated peptides.

Immunohistochemical Analyses. After fixation in 3% formalin for 2 to 3 days, human colorectal tissues were dehydrated through graded alcohol series, embedded in paraffin, and sectioned at a thickness of 4 µm. Paraffin sections then were deparaffinized in xylene and rehydrated. Nonenzymatic antigen retrieval was done by heating sections to 121°C in 0.01 mol/L sodium citrate buffer (pH 6.0) for 10 minutes. After cooling, sections were rinsed in TBS containing 0.1% Tween 20 (TBST) and incubated in methanol-3% H₂₂ for 30 minutes to quench endogenous peroxidase activity. After rinsing with TBST, sections were incubated with primary Abs for 1 hour at room temperature in a humid chamber. Primary Abs used in this study included mouse monoclonal Ab anti-Ki-67 (0.8 µg/mL, DAKO, Glostrup, Denmark), anti-pSmad2L (0.5 µg/mL), anti-pSmad2C (0.5 µg/mL), anti-pSmad3L (1 µg/mL), and anti-pSmad3C (1 µg/mL). Anti-pSmad3C Abcross-reacted weakly with COOH-terminally phosphorylated Smad2. To block binding of anti-pSmad3C Ab to phosphorylated domains in Smad2, anti-pSmad3C Ab was adsorbed with 1 µg/mL COOH-terminally phosphorylated Smad2 peptide.⁴ After sections were rinsed in TBST, they were incubated with peroxidase-labeled polymer conjugated to goat anti-mouse or anti-rabbit immunoglobulin (DAKO) for 1 hour at room temperature. Finally, the sections were developed with 3,3'-diaminobenzidine (Vector Laboratories, Burlingame, CA), counterstained with Mayer's hematoxylin (Merck, Darmstadt, Germany), and mounted under coverslips.

Immunoprecipitation and Immunoblotting. Frozen tissues were extracted with TNE buffer [10 mmol/L Tris-HCl (pH 7.5), 150 mmol/L NaCl, 1 mmol/L EDTA, 1% NP40, 100 mmol/L phenylmethylsulfonyl fluoride]. Cell extracts were subjected to immunoprecipitation with monoclonal anti-Smad2/3 Ab (BD Transduction Laboratories, Lexington, KY) followed by adsorption to protein G-Sepharose (Pharmacia, Peapack, NJ) for 1 hour. After washing thrice with TNE buffer, the immunoprecipitates were separated by 7.5% SDS-PAGE and transferred to a nitrocellulose membrane (Pharmacia). After the membranes were blocked overnight in TBST containing 5% bovine serum albumin (Nakarai, Kyoto, Japan), they were incubated with the indicated primary Abs for 1 hour at room temperature. Extent of phosphorylation of Smad2 and Smad3 was determined using each anti-pSmad2/3 Ab (11). Immunoblotting of total cell extracts was carried out using rabbit polyclonal anti-pJNK1/2 Ab (Cell Signaling Technology, Beverly, MA), anti-pp38 MAPK Ab (Promega, Madison, WI), anti-JNK1/2 Ab (Cell Signaling Technology), and anti-p38 MAPK Ab (Cell Signaling Technology). After the membranes were rinsed in TBST, they were incubated with horseradish peroxidase–conjugated anti-rabbit polyclonal Ab for 1 hour at room temperature. Proteins were detected by enhanced chemiluminescence (Pharmacia) and autoradiography. Densities of immunoreactive bands were measured using a densitometer (LKB, Bromma, Sweden).

In vitro Kinase Assay. Bacterial expression and purification of GST-Smad2 and GST-Smad3 were carried out according to the manufacturer's instructions (Amersham Biosciences, Piscataway, NJ). Endogenous kinases were isolated from the protein extracts using anti-pJNK1/2 Ab (Promega). Immune complexes collected with protein G-Sepharose were washed with kinase assay buffer [25 mmol/L Tris-HCl (pH 7.5), 5 mmol/L ß-glycerophosphate, 2 mmol/L DTT, 0.1 mmol/L Na₃VO₄, 10 mmol/L MgCl₂]. Pellets were resuspended in kinase assay buffer supplemented with100 µmol/L ATP and 2 µg of bacterially expressed GST-Smad2 or GST-Smad3. Assays were carried out at 30°C for 30 minutes and then were stopped by addition of Laemmli sample buffer. Phosphorylation sites in Smad2 and Smad3 were determined by immunoblotting using each anti-pSmad2/3 Ab.

Statistical Analysis. Statistical evaluation was done using the nonparametric Wilcoxon and Mann-Whitney U ranking tests. All values were based on two-tailed statistical analysis. The first test was used to evaluate significant differences in Smad2/3 phosphorylation between 31 paired primary adenocarcinomas and uninvolved normal colorectal mucosa from the same patients. The Mann-Whitney U test was used to test significant differences in Smad2/3 phosphorylation between 11 adenomas and uninvolved normal colorectal mucosa in the patients with colorectal cancer. P < 0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
DNA Sequencing of RT-PCR Products from 31 Human Colorectal Adenocarcinomas and 11 Adenomas Detects One Mutation of the Smad4 Gene in One Case. Mutations of TßRII, Smad2, and Smad4 genes have been detected in human colorectal tumors (14–17). To clarify the contribution of TGF-ß signaling in colorectal carcinogenesis, we initially analyzed mutations of these genes in 31 adenocarcinoma and 11 adenoma samples. The TßRII gene was amplified by RT-PCR using primers that included extracellular and kinase domains (18). RT-PCR primers for Smad2 and Smad4 genes included the MH2 domains, which are involved in homo- and hetero-oligomerization (6, 19, 20). When the PCR products from the tumor samples were sequenced, we detected a G/A substitution in one sample at codon 361, resulting in a missense mutation (Arg-to-His; patient 15 in Table 1). We detected no mutations in TßRII or Smad2 genes from any colorectal tumor samples. This represented a relatively low mutation frequency, which is consistent with recent reports (21).

View larger version (95K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Differences in localization of pSmad2/3L and pSmad2/3C among human colonic normal mucosa, adenoma, and adenocarcinoma. A, Smad2 and Smad3 proteins as well as domain-specific Abs. Anti-pSmad2L (Ser249/254) and anti-pSmad3L (Ser207/212) Abs recognize stress-activated protein kinase (SAPK)–dependent phosphorylation sites in Smad2/3, whereas anti-pSmad2C (Ser465/467) and anti-pSmad3C (Ser423/425) Abs recognize phosphorylated COOH-terminal SXS sites in Smad2/3 activated by TßRI. B, differential localization of pSmad2L and pSmad3L in human colonic adenocarcinoma. Formalin-fixed, paraffin-embedded sections of colonic cancerous tissue and uninvolved mucosa from patient 6 in Table 1 were stained with anti-pSmad2L ( pSmad2L), anti-pSmad2C ( pSmad2C), anti-pSmad3L ( pSmad3L), or anti-pSmad3C ( pSmad3C) Ab. These Abs in turn were bound by goat anti-rabbit immunoglobulins conjugated to peroxidase-labeled polymer. Peroxidase activity was detected with 3,3'-diaminobenzidine. C, higher magnification in the same tissue from the same patient. All sections were counterstained with hematoxylin (blue). Brown staining, specific Ab reactivity. Bar, 100 µm. In normal colonic epithelial cells (right), Smad2L and Smad3L are only minimally phosphorylated. In contrast, Smad2C and Smad3C show considerable phosphorylation; pSmad2C and pSmad3C can be seen in nuclei of the normal epithelial cells. In adenocarcinoma (left), Smad2L and Smad3L are more highly phosphorylated than in normal epithelial cells; pSmad2L is localized mainly to the cancer cell cytoplasm, whereas pSmad3L is seen exclusively in cancer cell nuclei. Whereas pSmad2C is distributed evenly in nuclei of the adenocarcinoma, nuclear pSmad3C shows a scattered distribution throughout the adenocarcinoma specimen. D, Smad2L and Smad3L are moderately phosphorylated in human colonic adenoma. Formalin-fixed, paraffin-embedded sections of colonic adenoma from patient 1 were stained with each anti-pSmad2/3 Ab as described above. Bar, 100 µm. Whereas pSmad2L is localized mainly in the cytoplasm of adenoma cells, pSmad3L shows accumulation in the nuclei of these cells. In contrast, both pSmad2C and pSmad3C are located predominantly in the nuclei.

Fig. 1B and C, right, pSmad2C and pSmad3C

Fig. 1B and C, left compared with right, pSmad2C

Fig. 1B and C, left, pSmad3C

Fig. 1B andC,right, pSmad2L and pSmad3L

Fig. 1B and C, left, pSmad2L and pSmad3L

Fig.1C, left, pSmad2L and pSmad3L

Figure 1D illustrates the distribution of phosphorylated Smad2/3 molecules in a human colonic adenoma. As in normal epithelial cells, both Smad2 and Smad3 were phosphorylated at COOH-terminal regions and pSmad2C and pSmad3C were located predominantly in the nuclei of adenoma cells (Fig. 1D, pSmad2C and pSmad3C). Similarly, Smad2 and Smad3 were phosphorylated at linker regions in the adenoma (Fig. 1D, pSmad2L and pSmad3L). Their extent of phosphorylation in the adenoma was intermediate between those in normal epithelial cells and adenocarcinomas. Whereas pSmad2L was found mainly in the cytoplasm of the adenoma cells, pSmad3L accumulated in the nuclei of these cells.

To evaluate extent of phosphorylation of each domain in Smad2 and Smad3, we subjected the extracts from colorectal tumors and noninvolved normal mucosa to immunoblotting with domain-specific Abs against the phosphorylated Smads. The linker regions of Smad2 and Smad3 showed very little phosphorylation in normal mucosa, where the COOH-terminal regions of Smad2 and Smad3 were moderately phosphorylated (Fig. 2A). In contrast, Smad2 and Smad3 in adenocarcinomas were highly phosphorylated at linker regions. Although the extent of phosphorylation of Smad2C in adenocarcinomas was almost the same as that in normal mucosa, Smad3C showed somewhat less phosphorylation in primary invasive adenocarcinoma than in normal mucosa. Considering immunoblotting findings together with the results obtained from immunohistochemical analyses (Fig. 1B and C), Smad2 and Smad3 were constitutively phosphorylated at linker regions in adenocarcinomas.

View larger version (29K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Phosphorylation of Smad2/3 is up-regulated at linker regions in human colorectal adenocarcinoma. A, cell lysates obtained from colorectal cancerous tissues and uninvolved mucosa from patients 6, 19, and 29 in Table 1 and adenoma from patient 1 were subjected to anti-Smad2/3 immunoprecipitation (IP) and were immunoblotted with each anti-pSmad2/3 Ab (top). Relative amounts of endogenous Smad2/3 were determined by immunoblotting (IB) using anti-Smad2/3 Ab (bottom). Graphic analyses of immunoblots show the ratio of pSmad2L to Smad2 (B), pSmad2C to Smad2 (C), pSmad3L to Smad3 (D), or pSmad3C to Smad3 (E) in human colorectal tumor tissues. Intensities of pSmad2L, pSmad2C, pSmad3L, or pSmad3C bands were normalized to those of Smad2 or Smad3 in corresponding groups. The ratio of the phosphorylated Smad2/3 to Smad2/3 in uninvolved mucosa was assigned a value of 1. Points, mean for each group. *, P < 0.05; **, P < 0.01; NS, not significant.

JNK in Human Colorectal Adenocarcinoma Directly Phosphorylates Smad2/3 at Linker Regions. We reported previously that Ser^249/254 in Smad2 and Ser^207/212 in Smad3, which anti-pSmad2L Ab and anti-pSmad3L Ab recognized, respectively, servedas substrates for JNK and/or p38 MAPK in vitro after TGF-ß treatment (11).⁴ Accordingly, we further investigated the phosphorylation states of JNK1/2 and p38 MAPK in the process of human colorectal carcinogenesis. We subjected the extracts from colorectal tumors and uninvolved normal mucosa to immunoblotting with Abs specific for phosphorylated JNK1/2 and p38 MAPK. The phosphorylation of JNK1, but not JNK2, was increased 2.4-fold in late-stage invasive and metastatic cancers compared with normal mucosa (Fig. 3A and B). In contrast, the degree of p38 MAPK phosphorylation in cancerous tissues was almost the same as that in uninvolved mucosa (Fig. 3A and C). Because the profiles of Smad2 and Smad3 phosphorylation in specimens representing stages in human colorectal carcinogenesis resembled those of JNK1 activation, these events may be causally linked.

View larger version (19K): [in this window] [in a new window] [Download PPT slide]   Figure 3 JNK1, but not p38 MAPK, is phosphorylated in human colorectal adenocarcinoma. A, cell lysates obtained from colorectal cancerous tissues and uninvolved mucosa from patients 6, 19, and 29 in Table 1 and adenoma from patient 1 were immunoblotted with anti-phospho-JNK1/2 ( pJNK1/2) or anti-phospho-p38 MAPK ( pp38) Ab. Relative amounts of endogenous JNK1/2 and p38 MAPK were determined by immunoblotting using anti-JNK1/2 ( JNK1/2) or anti-p38 MAPK ( p38) Ab. Graphic analyses of immunoblots show the ratio of pJNK1/2 to JNK1/2 (B) or pp38 MAPK to p38 MAPK (C) in human colorectal tumor tissues. Intensities of pJNK1/2 or pp38 MAPK bands were normalized to those of JNK1/2 or p38 MAPK in corresponding groups. The ratio of the phosphorylated stress-activated protein kinase to stress-activated protein kinase in uninvolved mucosa was assigned a value of 1. Points, mean for each group. *, P < 0.05; **, P < 0.01; NS, not significant.

View larger version (18K): [in this window] [in a new window] [Download PPT slide]   Figure 4 JNK in human colorectal adenocarcinoma directly phosphorylates Smad2/3 at linker regions. A, cell lysates obtained from colorectal cancerous tissues and uninvolved mucosa from patients 6, 19, and 29 in Table 1 and adenoma from patient 1 were subjected to anti-phospho-JNK1/2 immunoprecipitation and were mixed with bacterially expressed GST-Smad2 and GST-Smad3. Phosphorylation of Smad2/3L was analyzed by immunoblotting using each anti-pSmad2/3L Ab (top). Total Smad2 and Smad3 were determined by immunoblotting using anti-Smad2/3 Ab (bottom). Graphic analyses of immunoblots show the ratio of pSmad2L to Smad2 (B) or pSmad3L to Smad3 (C) in human colorectal tumor tissues. Intensities of pSmad2L or pSmad3L bands were normalized to those of Smad2 or Smad3 in corresponding groups. The ratio of the phosphorylated Smad2/3L to Smad2/3 in uninvolved mucosa was assigned a value of 1. Points, mean for each group. **, P < 0.01; NS, not significant.

Growing Ki-67-Immunoreactive Adenocarcinoma with Distant Metastasis Shows pSmad3L Localization Exclusively in Cell Nuclei. We finally investigated the relationship between pSmad3L and cellular proliferation in primary invasive adenocarcinoma with distant metastasis. To assess the proliferative states of adenocarcinoma, nuclear expression of proliferation-associated antigen Ki-67 was examined immunohistochemically. Figure 5 shows pSmad3L distribution and Ki-67 expression in primary invasive adenocarcinoma and uninvolved colonic mucosa from patient 29 in Table 1. Smad3L was only minimally phosphorylated in normal colonic epithelial cells (Fig. 5A); scattered pSmad3L-immunoreactive cells were essentially confined to the proliferative compartments in the basal third of the mucosal crypts (Fig. 5B). Throughout the glands formed by the adenocarcinoma, pSmad3L was significantly more abundant than in normal glands. Smad3L occasionally showed phosphorylation in inflammatory cells within mucosal connective tissue. Glands showing highly phosphorylated Smad3L were evenly distributed in the adenocarcinoma. Expression of Ki-67 was low in the normal mucosa. In particular, the superficial epithelium failed to show any Ki-67 expression. In parallel with the distribution of pSmad3L-immunoreactive cells, scattered Ki-67-immunoreactive cells were confined to the basal third of crypts. Colonic adenocarcinoma showed a higher frequency of expression of Ki-67, but this still was patchy. A significant positive relationship was evident between pSmad3L distribution and Ki-67 expression, although the constitutive distribution of pSmad3L was greater than that of Ki-67.

View larger version (87K): [in this window] [in a new window] [Download PPT slide]   Figure 5. pSmad3L is localized exclusively in cell nuclei of growing Ki-67-immunoreactive adenocarcinoma with distant metastasis. A, sections of formalin-fixed, paraffin-embedded colonic cancerous tissue and uninvolved mucosa from patient 29 in Table 1 were exposed to anti-pSmad3L or anti-Ki-67 ( Ki-67) Ab. The Abs in turn were bound by goat anti-rabbit or anti-mouse immunoglobulins conjugated with peroxidase-labeled polymer. Peroxidase activity was detected with 3,3'-diaminobenzidine. B, higher magnification of the boxed areas in A. All sections were counterstained with hematoxylin (blue). Brown staining, specific Ab reactivity. Bar, 100 µm. In normal colonic mucosa, scattered cells immunoreactive for pSmad3L are seen in the proliferative compartment represented by the basal third of epithelial crypts (closed arrow). Similarly, in an adjacent serial section, Ki-67 expression is confined to the basal third of the crypts, with no superficial expression. High-power views of a cancerous area show strong signal for pSmad3L in cell nuclei of the adenocarcinoma. In adenocarcinomas, pSmad3L often was associated closely with Ki-67-expressing cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

TGF-ß inhibited growth of a rat intestinal epithelial cell line (22) and was found to directly induce apoptotic cell death (23). In an investigation of ligand-receptor interactions in normal colorectal epithelial cells, we reported previously that both TßRI and TßRII were strongly expressed in human normal colorectal epithelial cells (24). In situ hybridization and immunohistochemistry to detect the ligand suggested that TGF-ß is produced by and is present within human colorectal epithelial cells (24, 25). Taken together with our current findings that both pSmad2C and pSmad3C were located predominantly in the nuclei of normal colorectal epithelial cells, TGF-ß in normal colorectal epithelial cells mainly transmits autocrine signals through pSmad2/3C mediated by TßRI. In most normal epithelial cell types, TGF-ß arrests cell cycle progression in the G₁ phase by up-regulating expression of cyclin-dependent kinase inhibitors p21^WAF1/CIP1 and/or p15^INK4B (26, 27). TGF-ß also induces expression of death-associated protein kinase as an immediate-early response by the cells in which it induces apoptosis (28). The death-associated protein kinase promoter is activated by TGF-ß through the action of Smad2, Smad3, and Smad4. Induction of both cyclin-dependent kinase inhibitors and death-associated protein kinase expression requires COOH-terminal phosphorylation of Smad2 and Smad3 (28–30). Thus, pSmad2/3C-mediated signaling seems to take part in growth inhibition and apoptosis in normal colorectal epithelial cells.

Our current findings highlighted a significant increase in JNK-dependent pSmad2/3L as the process of neoplasia progressed from normal colorectal epithelial cells to invasive adenocarcinoma with distant metastasis. In addition, nuclear pSmad3L showed a positive correlation with Ki-67 expression, a proliferative marker. These results confirm and extend previous findings that JNK activation is involved in proliferation of rat intestinal neoplasia (31). Because JNK-dependent Smad3L phosphorylation promotes nuclear accumulation of Smad3, reduction of pSmad3C could be explained by reduced accessibility of Smad3 to TßRI (11). Accordingly, an increase in pSmad3L during colorectal carcinogenesis (Fig. 2D) could lead to the observed decrease in pSmad3C (Fig. 2E), resulting in desensitization of the cell to TGF-ß tumor suppressor activity (11). Moreover, TGF-ß treatment caused to an increase in plasminogen activator inhibitor type 1 transcriptional activity through pSmad3L (11). Because plasminogen activator inhibitor type 1 conducts the cells to migration and invasion by blocking cellular adhesion and by promoting basement membrane degradation, stimulation of plasminogen activator inhibitor type 1 production might lead to an increase in invasive capacity of cancer cells (32). In support of this role of pSmad3L, late-stage cancer with deep invasion and/or metastasis typically was characterized by high phosphorylation of Smad3L. Collectively, the change of phosphorylation sites in Smad2/3 from COOH-terminal regions to linker regions is one of the major mechanisms for the complex transition of TGF-ß signaling during human colorectal carcinogenesis.

Few pSmad2/3L-positive epithelial cells could be seen in normal epithelial crypts. Adenomas showed degree of linker phosphorylation that was a few times greater than in normal mucosa (i.e., a pSmad2/3L signal intensity intermediate between those typical of normal epithelial cells and adenocarcinomas). Taken together with the high degree of pSmad2/3L in primary invasive adenocarcinoma, our results suggest that the linker phosphorylation develops relatively late in the evolution of colorectal tumors accompanied by persistent JNK activity.

Because K-ras mutations have been detected in both adenoma and carcinoma (1) and JNK can be activated through ras (33), increased JNK-dependent phosphorylation of Smad2/3L observed in colorectal tumors may be a direct consequence of the K-ras mutation. However, the degree of the linker phosphorylation in individual tumors did not reflect the presence of K-ras mutation. Thus, only 47% of tumors studied harbored mutations of K-ras (data not shown), whereas all adenomas and adenocarcinomas examined displayed elevated JNK-dependent phosphorylation of Smad2/3L. Several pathways including Src are responsible for activation of JNK during colorectal carcinogenesis (34, 35).

Because TGF-ß signaling participates in the physiology of normal epithelial cells, unraveling the molecular mechanisms of TGF-ß signal leading to pathogenesis of human cancer is critical to development of new therapies (36, 37). In addition to continued molecular studies in vitro, investigations need to be extended to more complex in vivo models (2). The present work represents such application of studies on TGF-ß signaling in vitro to one in vivo situation, resected human tissues. In the process of human colorectal carcinogenesis, pSmad3C, which favors tumor suppressor activity of TGF-ß, was found to decrease, whereas JNK tended to induce the phosphorylation of Smad2/3L. From the view point of TGF-ß signaling, a key therapeutic challenge in cancer would be restoration of the lost tumor suppressor function observed in normal colorectal epithelial cells at the expense of oncogenic effects that would lead to more aggressive adenocarcinoma (36). Specific inhibitors of the JNK-mediated Smad2/3L pathway might prove useful in this respect. Investigation concerning the distribution of pSmad2/3L in a given specimen would be needed to predict potential response to molecularly targeted therapy for human sporadic colorectal cancer.

	Acknowledgments

 
Grant support: Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan.

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. R. Derynck (University of California at San Francisco, San Francisco, CA) for providing the cDNAs encoding human Smad2 and Smad3.

	Footnotes

 
⁴ K. Matsuzaki et al. Counterbalanced Smad2/3 phosphorylation at linker and C-terminal regions determines the response to Smad7 transcription, submitted for publication.

Received 6/22/04. Revised 10/ 4/04. Accepted 10/27/04.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic alterations during colorectal-tumor development. N Engl J Med 1988;88:525–32.
2. Wakefield LM, Roberts AB. TGF-ß signaling: positive and negative effects on tumorigenesis. Curr Opin Genet Dev 2002;12:22–9.[CrossRef][Medline]
3. Manning AM, Williams AC, Game SM, Paraskeva C. Differential sensitivity of human colonic adenoma and carcinoma cells to transforming growth factor ß (TGF-ß): conversion of an adenoma cell line to a tumorigenic phenotype is accompanied by a reduced response to the inhibitory effects of TGF-ß. Oncogene 1991;6:1471–6.[Medline]
4. Heldin CH, Miyazono K, ten Dijke P. TGF-ß signalling from cell membrane to nucleus through SMAD proteins. Nature 1997;390:465–71.[CrossRef][Medline]
5. Wrana JL. Regulation of Smad activity. Cell 2000;100:189–92.[CrossRef][Medline]
6. Shi Y, Massagu E, J. Mechanisms of TGF-ß signaling from cell membrane to the nucleus. Cell 2003;113:685–700.[CrossRef][Medline]
7. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-ß family signalling. Nature 2003;425:577–84.[CrossRef][Medline]
8. Kretzschmar M, Doody J, Timokhina I, Massagu E, J. A mechanism of repression of TGFß/Smad signaling by oncogenic Ras. Genes Dev 1999;13:804–16.[Abstract/Free Full Text]
9. Robinson MJ, Cobb MH. Mitogen-activated protein kinase pathways.Curr Opin Cell Biol 1997;9:180–6.[CrossRef][Medline]
10. Mulder KM. Role of Ras and Mapks in TGFß signaling. Cytokine Growth Factor Rev 2000;11:23–35.[CrossRef][Medline]
11. Mori S, Matsuzaki K, Yoshida K, et al. TGF-ß and HGF transmit the signals through JNK-dependent Smad2/3 phosphorylation at the linker regions. Oncogene 2004;23:7416–29.[CrossRef][Medline]
12. Japanese Society for Cancer of the Colon and Rectum. Japanese classification of colorectal carcinoma. 1st ed. Tokyo: Japanese Society for Cancer of the Colon and Rectum; 1997.
13. Hutter RVP, Sobin LH. A universal staining system for cancer of the colon and rectum. Arch Pathol Lab Med 1986;110:367–8.[Medline]
14. Markowitz S, Wang J, Myeroff L, et al. Inactivation of the type II TGF-ß receptor in colon cancer cells with microsatellite instability. Science 1995;268:1336–8.[Abstract/Free Full Text]
15. Grady WM, Myeroff LL, Swinler SE, et al. Mutational inactivation of transforming growth factor ß receptor type II in microsatellite stable colon cancers. Cancer Res 1999;59:320–4.[Abstract/Free Full Text]
16. Riggins GJ, Thiagalingam S, Rozenblum E, et al. Mad-related genes in the human. Nat Genet 1996;13:347–9.[CrossRef][Medline]
17. Takagi Y, Kohmura H, Futamura M, et al. Somatic alterations of the DPC4 gene in human colorectal cancers in vivo. Gastroenterology 1996;111:1369–72.[CrossRef][Medline]
18. Lin HY, Wang X-F, Ng-Eaton E, Weinberg RA, Lodish HF. Expression cloning of the TGF-ß type II receptor, a functional transmembrane serine/threonine kinase. Cell 1992;68:775–85.[CrossRef][Medline]
19. Eppert K, Scherer SW, Ozcelik H, et al. MADR2 maps to 18q21 and encodes a TGF-ß-regulated MAD-related protein that is functionally mutated in colorectal carcinoma. Cell 1996;86:543–52.[CrossRef][Medline]
20. Hahn SA, Schutte M, Shamsul Hoque ATM, et al. DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1. Science 1996;271:350–3.[Abstract]
21. Duff EK, Clarke AR. Smad4 (DPC4)—a potent tumour suppressor? Br J Cancer 1998;78:1615–9.[Medline]
22. Barnard JA, Beauchamp RD, Coffey RJ, Moses HL. Regulation of intestinal epithelial cell growth by transforming growth factor type ß. Proc Natl Acad Sci U S A 1989;86:1578–82.[Abstract/Free Full Text]
23. Wang CY, Eshleman JR, Willson JKV, Markowitz S. Both transforming growth factor-ß and substrate release are inducers of apoptosis in a human colon adenoma cell line. Cancer Res 1995;55:5101–5.[Abstract/Free Full Text]
24. Matsushita M, Matsuzaki K, Date M, et al. Down-regulation of TGF-ß receptors in human colorectal cancer: implications for cancer development. Br J Cancer 1999;80:194–205.[CrossRef][Medline]
25. Avery A, Paraskeva C, Hall P, Flanders KC, Sporn M, Moorghen M. TGF-ß expression in the human colon: differential immunostaining along crypt epithelium. Br J Cancer 1993;68:137–9.[Medline]
26. Koff A, Ohtsuki M, Polyak K, Roberts JM, Massagu E, J. Negative regulation of G₁ in mammalian cells: inhibition of cyclin E-dependent kinase by TGF-ß. Science 1993;260:536–9.[Abstract/Free Full Text]
27. Hannon GJ, Beach D. p15^INK4B is a potential effector of TGF-ß-induced cell cycle arrest. Nature 1994;371:257–60.[CrossRef][Medline]
28. Jang CW, Chen CH, Chen CC, Chen JY, Su YH, Chen RH. TGF-ß induces apoptosis through Smad-mediated expression of DAP-kinase. Nat Cell Biol 2002;4:51–8.[CrossRef][Medline]
29. Feng XH, Lin X, Derynck R. Smad2, Smad3 and Smad4 cooperate with Sp1 to induce p15^Ink4B transcription in response to TGF-ß. EMBO J 2000;19:5178–93.[CrossRef][Medline]
30. Moustakas A, Kardassis D. Regulation of the human p21/WAF1/Cip1 promoter in hepatic cells by functional interactions between Sp1 and Smad family members. Proc Natl Acad Sci U S A 1998;95:6733–8.[Abstract/Free Full Text]
31. Licato LL, Keku TO, Wurzelmann JI, et al. In vivo activation of mitogen-activated protein kinase in rat intestinal neoplasia. Gastroenterology 1997;113:1589–98.[CrossRef][Medline]
32. Hirashima Y, Kobayashi H, Suzuki M, Tanaka Y, Kanayama N, Terao T. Transforming growth factor-ß1 produced by ovarian cancer cell line HRA stimulates attachment and invasion through an up-regulation of plasminogen activator inhibitor type-1 in human peritoneal mesothelial cells. J Biol Chem 2003;278:26793–802.[Abstract/Free Full Text]
33. Derijard B, Hibi M, Wu I, et al. JNK1: a protein kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun activation domain. Cell 1994;76:1025–37.[CrossRef][Medline]
34. Talamonti MS, Rob MS, Curley SA, Gallik GE. Increase inactivity and level of pp60^c-src in progressive stages of human colorectal cancer. J Clin Invest 1993;91:53–60.
35. Xie W, Herschman HR. v-src Induces prostaglandin synthase 2 gene expression by activation of the c-Jun N-terminal kinase and the c-Jun transcription factor. J Biol Chem 1995;270:27622–8.[Abstract/Free Full Text]
36. de Caestecker MP, Piek E, Roberts AB. Role of transforming growth factor-ß signaling in cancer. J Natl Cancer Inst2000;92:1388–402.[Abstract/Free Full Text]
37. Moses HL, Yang EY, Pietenpol JA. TGF-ß stimulation and inhibition of cell proliferation: new mechanistic insights. Cell 1990;63:245–7.[CrossRef][Medline]

This article has been cited by other articles:

				A. Saadeddin, R. Babaei-Jadidi, B. Spencer-Dene, and A. S. Nateri The Links between Transcription, {beta}-catenin/JNK Signaling, and Carcinogenesis Mol. Cancer Res., August 1, 2009; 7(8): 1189 - 1196. [Abstract] [Full Text] [PDF]

				K. Matsuzaki, C. Kitano, M. Murata, G. Sekimoto, K. Yoshida, Y. Uemura, T. Seki, S. Taketani, J.-i. Fujisawa, and K. Okazaki Smad2 and Smad3 Phosphorylated at Both Linker and COOH-Terminal Regions Transmit Malignant TGF-{beta} Signal in Later Stages of Human Colorectal Cancer Cancer Res., July 1, 2009; 69(13): 5321 - 5330. [Abstract] [Full Text] [PDF]

				G. Wang, I. Matsuura, D. He, and F. Liu Transforming Growth Factor-{beta}-inducible Phosphorylation of Smad3 J. Biol. Chem., April 10, 2009; 284(15): 9663 - 9673. [Abstract] [Full Text] [PDF]

				Q. Zeng, S. Phukan, Y. Xu, M. Sadim, D. S. Rosman, M. Pennison, J. Liao, G.-Y. Yang, C.-C. Huang, L. Valle, et al. Tgfbr1 Haploinsufficiency Is a Potent Modifier of Colorectal Cancer Development Cancer Res., January 15, 2009; 69(2): 678 - 686. [Abstract] [Full Text] [PDF]

				H. J. Lee, Y. Ji, S. Paul, H. Maehr, M. Uskokovic, and N. Suh Activation of Bone Morphogenetic Protein Signaling by a Gemini Vitamin D3 Analogue Is Mediated by Ras/Protein Kinase C{alpha} Cancer Res., December 15, 2007; 67(24): 11840 - 11847. [Abstract] [Full Text] [PDF]

				T. Shirakihara, M. Saitoh, and K. Miyazono Differential Regulation of Epithelial and Mesenchymal Markers by {delta}EF1 Proteins in Epithelial Mesenchymal Transition Induced by TGF-beta Mol. Biol. Cell, September 1, 2007; 18(9): 3533 - 3544. [Abstract] [Full Text] [PDF]

				G. Sekimoto, K. Matsuzaki, K. Yoshida, S. Mori, M. Murata, T. Seki, H. Matsui, J.-i. Fujisawa, and K. Okazaki Reversible Smad-Dependent Signaling between Tumor Suppression and Oncogenesis Cancer Res., June 1, 2007; 67(11): 5090 - 5096. [Abstract] [Full Text] [PDF]

				H.-A Seong, H. Jung, K.-T. Kim, and H. Ha 3-Phosphoinositide-dependent PDK1 Negatively Regulates Transforming Growth Factor-beta-induced Signaling in a Kinase-dependent Manner through Physical Interaction with Smad Proteins J. Biol. Chem., April 20, 2007; 282(16): 12272 - 12289. [Abstract] [Full Text] [PDF]

				Y. Xu and B. Pasche TGF-{beta} signaling alterations and susceptibility to colorectal cancer Hum. Mol. Genet., April 15, 2007; 16(R1): R14 - R20. [Abstract] [Full Text] [PDF]

				K. H. Wrighton, D. Willis, J. Long, F. Liu, X. Lin, and X.-H. Feng Small C-terminal Domain Phosphatases Dephosphorylate the Regulatory Linker Regions of Smad2 and Smad3 to Enhance Transforming Growth Factor-beta Signaling J. Biol. Chem., December 15, 2006; 281(50): 38365 - 38375. [Abstract] [Full Text] [PDF]

				Y. Sun, L. Ding, H. Zhang, J. Han, X. Yang, J. Yan, Y. Zhu, J. Li, H. Song, and Q. Ye Potentiation of Smad-mediated transcriptional activation by the RNA-binding protein RBPMS Nucleic Acids Res., December 4, 2006; 34(21): 6314 - 6326. [Abstract] [Full Text] [PDF]

				J. Araya, S. Cambier, A. Morris, W. Finkbeiner, and S. L. Nishimura Integrin-Mediated Transforming Growth Factor-{beta} Activation Regulates Homeostasis of the Pulmonary Epithelial-Mesenchymal Trophic Unit Am. J. Pathol., August 1, 2006; 169(2): 405 - 415. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yamagata, H. Articles by Okazaki, K. Search for Related Content PubMed PubMed Citation Articles by Yamagata, H. Articles by Okazaki, K.