Cancer Research Cancer Research Funding Available Genetics and Biology of Brain Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ding, W.
Right arrow Articles by Mulder, K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ding, W.
Right arrow Articles by Mulder, K. M.
[Cancer Research 65, 6526-6533, August 1, 2005]
© 2005 American Association for Cancer Research

Molecular Biology, Pathobiology and Genetics

A Transforming Growth Factor-ß Receptor–Interacting Protein Frequently Mutated in Human Ovarian Cancer

Wei Ding1, Qian Tang1, Virginia Espina3, Lance A. Liotta3, David T. Mauger2 and Kathleen M. Mulder1

Departments of 1 Pharmacology and 2 Health Evaluation Sciences, Pennsylvania State University College of Medicine, Hershey, Pennsylvania and 3 National Cancer Institute Clinical Proteomics Program, Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland

Requests for reprints: Kathleen M. Mulder, Department of Pharmacology-MC H078, Pennsylvania State College of Medicine, 500 University Drive, Hershey, PA 17033. Phone: 717-531-6789; Fax: 717-531-5013; E-mail: kmm15{at}psu.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Ovarian carcinomas, particularly recurrent forms, are frequently resistant to transforming growth factor-ß (TGF-ß)–mediated growth inhibition. However, mutations in the TGF-ß receptor I and receptor II (TßR-I and TßR-II) genes have only been reported in a minority of ovarian carcinomas, suggesting that alterations in TGF-ß–signaling components may play an important role in the loss of TGF-ß responsiveness. Using laser-capture microdissection and nested reverse-transcription-PCR, we found that km23, which interacts with the TGF-ß receptor complex, is altered at a high frequency in human ovarian cancer patients. A novel form of km23, missing exon 3 ({Delta}exon3-km23), was found in 2 of 19 tumor tissues from patients with ovarian cancer. In addition to this alteration, a stop codon mutation (TAA -> CAC) was detected in two patients. This alteration results in an elongated protein, encoding 107-amino-acid residues ({Delta}107km23), instead of the wild-type 96-amino-acid form of km23. Furthermore, five missense mutations (T38I, S55G, T56S, I89V, and V90A) were detected in four patients, providing a total alteration rate of 42.1% (8 of 19 cases) in ovarian cancer. No km23 alterations were detected in 15 normal tissues. Such a high alteration rate in ovarian cancer suggests that km23 may play an important role in either TGF-ß resistance or tumor progression in this disease. In keeping with these findings, the functional studies described herein indicate that both the {Delta}exon3-km23 and S55G/I89V-km23 mutants displayed a disruption in binding to the dynein intermediate chain in vivo, suggesting a defect in cargo recruitment to the dynein motor complex. In addition, the {Delta}exon3-km23 resulted in an inhibition of TGF-ß–dependent transcriptional activation of both the p3TP-lux and activin responsive element reporters. Collectively, our results suggest that km23 alterations found in ovarian cancer patients result in altered dynein motor complex formation and/or aberrant transcriptional regulation by TGF-ß.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Epithelial ovarian cancer is often diagnosed at an advanced stage and is the leading cause of death from gynecologic neoplasia, accounting for >14,300 deaths per year (1). Despite advances in surgical approaches and chemotherapeutic agents, the overall survival rates for women with this cancer have not improved significantly because the majority of women are diagnosed with ovarian cancer at an advanced stage, when it has already metastasized to remote organs (1, 2).

Overall, the molecular changes that underlie the initiation and development of this tumor are poorly understood. It has been reported that >75% of ovarian carcinomas are resistant to transforming growth factor-ß (TGF-ß), particularly recurrent ones (3, 4). The loss of TGF-ß responsiveness may play an important role in the pathogenesis and/or progression of ovarian cancer. It has been shown that TGF-ß1, the TGF-ß receptors (TßR-II and TßR-I), and the TGF-ß-signaling component Smad2 are altered in ovarian cancer (5–8). Alterations in TßR-II have been identified in 25% of ovarian carcinomas (5), whereas mutations in TßR-I were reported in 33% of such cancers (6). Loss of function mutations of TGF-ß1, TßR-I, and TßR-II can lead to disruption of TGF-ß-signaling pathways and subsequent loss of cell cycle control (5–7). However, these alterations only account for a minority of TGF-ß-resistant ovarian carcinomas, suggesting that other alterations in TGF-ß-signaling components may be involved in the pathogenesis of this type of cancer.

Recently, we have identified a novel TGF-ß-signaling component, termed km23, which is also a light chain of the motor protein dynein (9). Our previous studies have shown that is a TßRII-interacting protein that can be phosphorylated on serine residues after TGF-ß binding (9). Forced expression of km23 induced specific TGF-ß responses, including c-jun NH2-terminal kinase (JNK) activation, c-Jun phosphorylation, and cell growth inhibition (9). Furthermore, TGF-ß induced the recruitment of km23 to the dynein intermediate chain (DIC). A kinase-deficient form of TGF-ß RII prevented both km23 phosphorylation and interaction with DIC (9). From these studies, we have proposed a model for km23 function, whereby the binding of km23 to the DIC after TGF-ß receptor activation is important for specifying the nature of the cargo (i.e., TGF-ß-signaling components) that will be transported along the microtubules (9, 10). In keeping with this model, any disruption of the binding of km23 to the DIC should disrupt cargo binding and subsequent TGF-ß signaling. In this way, km23 may not only be an important component in TGF-ß signaling, but any alterations in km23 sequence may contribute to dysregulated TGF-ß signaling.

Here, we report the analysis of sequence alterations in human km23 in epithelial ovarian cancer tissues using laser-capture microdissection (LCM) and nested reverse-transcription-PCR (RT-PCR) approaches. We found a high incidence (42.1%) of km23 mutations in human ovarian carcinomas, whereas normal tissues did not contain such alterations (P < 0.005). This is the first report of alterations in a cytoplasmic dynein light chain (DLC) in epithelial ovarian cancer. Furthermore, we found that both the {Delta}exon3-km23 and S55G/I89V-km23 mutants resulted in a disrupted interaction with the DIC. The {Delta}exon3-km23 mutant form of km23 also inhibited TGF-ß-dependent activation of both p3TP-lux and ARE-lux transcription. In addition, two phosphorylation site mutants, S32A-km23 and S73A-km23, disrupted the interaction with DIC. Overall, our results show that altered forms of this novel TGF-ß-signaling component found in ovarian cancer patients can result in altered dynein motor interactions and/or altered regulation of TGF-ß-dependent gene transcription.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell cultures. 293T cells and Mink lung epithelial cells CCL-64 (Mv1Lu) were purchased from American Type Culture Collection (Rockville, MD). 293T and Mv1Lu cells were cultured in DMEM with glutamine. All medium was supplemented with 10% heat-inactivated fetal bovine serum. Cells were maintained in 5% CO2 at 37°C. Cultures were routinely screened for Mycoplasma using Hoechst staining.

Ovarian tissues. Nineteen ovarian carcinoma tissues and 12 normal ovarian tissues were provided by Cooperative Human Tissue Network (Eastern division, Philadelphia, PA and Midwestern division, Columbus, OH). Tissues were obtained within 1 hour of surgery, snap frozen, sent on dry ice, and stored at –80°C before use. The Institutional Review Board approval for this study was received from the Penn State College of Medicine (Hershey, PA).

H&E staining. Sections of frozen tissues (10 µm) were mounted on slides (SL Microtest, Jena, Germany). H&E staining was done as described by Goldsworthy et al. (11). All reagents for H&E staining were prepared with diethyl-pyrocarbonate–treated distilled water.

Laser-capture microdissection. LCM was done using a µCut Laser Microdissection system (SL Microtest) according to the manufacturer's instructions. After microdissection of each specimen, the thermoplastic film-coated cap containing the captured tissue was placed in a 0.5-mL microtube.

RNA isolation. Total RNA was isolated from normal ovarian tissues using TRIzol reagent according to the manufacturer's protocol (Invitrogen, Life technologies, Carlsbad, CA). Total RNA was isolated from tissues subjected to LCM using a Total RNA Microprep Kit (Stratagene, La Jolla, CA) according to the manufacturer's recommendations.

Nested reverse-transcription-PCR and DNA sequencing. cDNA synthesis was done using Sensicript Reverse Transcriptase (Qiagen, Valencia, CA), according to the manufacturer's protocol. Nested PCR was done using two pairs of primers spanning the whole open reading frame of km23. The forward primer for the first round of PCR was 5'-GTTTTGACAGAAACCTTTGCG-3' and the reverse primer was 5'-TTGGTGCACACAGG GGTTC-3'. The conditions used for the first round of PCR were 30 cycles at 94°C for 50 seconds, at 54°C for 50 seconds, and at 72°C for 1 minute. The second round of PCR was done using the first round of PCR products as a reaction template (forward primer, 5'-ACTCGCTAAGTGTTCGCTACG-3'; reverse primer, 5'-TGCCATGTGCTAGTCCACTGA-3'), with the following conditions: 33 cycles at 94°C for 50 seconds, at 62°C for 50 seconds, and at 72°C for 1 minute. All PCR assays were done using recombinant Pfu polymerase with 3' to 5' exonuclease activity (Stratagene). The PCR products were electrophoresed on 2% agarose gels, visualized with ethidium bromide staining, and purified using a QIAquick Gel Extraction Kit (Qiagen). DNA sequencing was done in both directions using the primers for the second round of PCR.

DNA constructs. Flag-tagged wt-km23, {Delta}exon3-km23, and S55G/I89V-km23 constructs were generated by inserting the corresponding PCR products into pFlag-CMV5a vector (Sigma-Aldrich, St. Louis, MO) after digestion with BglII and SalI restriction enzymes. The Flag-tagged S32A-km23, S55A-km23, and S73A-km23 constructs were produced using the QuikChange Site-Directed Mutagenesis Kit (Stratagene) following the manufacturer's protocols.

Immunoprecipitation/immunoblotting. 293T cells were transiently transfected with the indicated Flag-tagged forms of km23. Twenty-four hours after transfection, cells were harvested using radioimmunoprecipitation assay buffer (1x PBS, 1% NP40, 0.5% sodium deoxycholate, and 0.1% SDS) supplemented with protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). Immunoprecipitation/blot assays were done as described previously (9, 12). The anti-DIC antibody for immunoprecipitation analyses was purchased from Chemicon (Temecula, CA), and the anti-Flag antibody for immunoblot analysis was purchased from Sigma-Aldrich.

Luciferase reporter assays. TGF-ß-dependent p3TP-lux and ARE-lux reporter assays were done in Mv1Lu cells. Mv1Lu cells were cotransfected either with p3TP-lux (13) or ARE-Lux (along with forkhead activin signal transducer-1, FAST-1; ref. 14), and the indicated amounts of either wild-type (wt) km23 or {Delta}exon3-km23. The reporter assays were done as described previously (15). All assays were done in triplicate.

Statistics. The statistical difference in alteration rate between the ovarian carcinoma tissues and the tumor-free ovarian tissue groups was calculated using Fisher's exact test.

Phosphorylation site prediction. Phosphorylation site prediction for km23 was done using PROSITE4 and NetPhos 2.0 Server (16).5


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
A truncated form of km23, missing exon 3 ({Delta}exon3-km23), was detected in 2 of 19 ovarian carcinoma tissues. Because alterations in TGF-ß-signaling components and pathways have been reported in several types of human cancers (17, 18), including epithelial ovarian cancer (5–8), it was of interest to investigate whether km23, a novel TGF-ß receptor-interacting protein, was altered in ovarian cancer patients. To examine whether sequence alterations in km23 existed in ovarian cancer patients, we developed a novel approach to detect km23 mutations using two microscale technologies. First, LCM was done on ovarian cancer tissues from patients to separate epithelial tumor cells from adjacent tumor-free normal cells. Nested RT-PCR and DNA sequencing were then done to analyze km23 sequence alterations in the RNA isolated from the tissues subjected to LCM. Using these two microscale technologies, we amplified a PCR fragment of 424 bp in 19 ovarian cancer tissues from patients, using forward and reverse primers located in the 5'-untranslated region (UTR) and 3'-UTR regions, respectively. This 424-bp fragment spans the entire coding region of human km23. As shown in Fig. 1A, a smaller band of 256 bp, in addition to the 424-bp product, was detectable by agarose gel electrophoresis in 2 of 19 tumor tissues (cases 3 and 8). DNA sequencing indicated that this short fragment is a truncated form of km23 ({Delta}exon3-km23). It is missing exon 3 and contains only 123 nucleotides, as opposed to the 291 nucleotides in the coding region of wt-km23. In case 8, we also analyzed the km23 sequence in tumor-free stromal cells and found no alterations in km23.



View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. A, detection of a truncated form of km23 ({Delta}exon3-km23) in 2 of 19 ovarian carcinoma tissues. Agarose electrophoresis (2%) of nested RT-PCR products. In addition to a 424-bp fragment, which corresponds to the wt-km23, a truncated form of km23 ({Delta}exon3-km23, 256 bp) was detected in two patients (cases 3 and 8). This short fragment was not present in the other tumor samples (cases 1, 7, and 15). B, schematic of human km23 sequence depicting exon 3 and potential phosphorylation sites. wt-km23 consists of four exons. Among them, exon 3 is the largest and encodes a region of 56 amino acids (underlined). The predicted size for the {Delta}exon3-km23 mutant is 40 amino acids. Potential phosphorylation sites for PKA, PKC, and/or CKII were predicted using NetPhos 2.0 Server and PROSITE (16). Four serine residues that are conserved among the mammalian km23 forms are boxed. Among them, three are located in exon 3.

 
To illustrate the important sites that would be lost in the {Delta}exon3-km23 mutant, a schematic of the human km23 sequence is depicted in Fig. 1B. km23 consists of four exons encoding 96 amino acids. Among these exons, exon 3 is the largest and encodes a region from amino acids 27 to 82 (shown by underline). From the sequencing results for {Delta}exon3-km23 mutant, the predicted size for {Delta}exon3-km23 is 40 amino acids. Because the 56 amino acids in exon 3 are missing, this large truncation of the km23 protein is likely to cause significant alterations in km23 function.

To determine whether exon 3 of km23 contains any important functional domains, we also did phosphorylation site prediction analysis using NetPhos 2.0 Server (16) and PROSITE. As shown in Fig. 1B, there are several potential phosphorylation sites for protein kinase A (PKA), protein kinase C (PKC), and/or casein kinase II (CKII) in km23. For these sites, the prediction scores for serine and threonine were all over 0.91, indicating that the probability of these sites being true phosphorylation sites is quite high (16). In addition, there are four serine residues that may potentially be phosphorylated and are conserved among the mammalian km23 forms (shown by boxes). Among these, three are located in exon 3. Thus, the {Delta}exon3-km23 mutant may have defective phosphorylation reactions as well.

Six missense mutations in km23 were detected in 31.6% of epithelial ovarian carcinoma tissues. In addition to the abovementioned findings, we obtained six missense mutations in km23 in six patients (Table 1). In case 1, the 163rd nucleotide of the coding region was changed from an adenine to a guanine (A -> G). As shown in Fig. 2A, this alteration corresponds to an amino acid change from a serine to a glycine (S55G). Also in this patient, the 265th nucleotide of the coding region was altered from an adenine to a guanine (A -> G), corresponding to an amino acid change in residue 89 of km23 from an isoleucine to a valine (I89V; Fig. 2A). Furthermore, in two other cases (cases 4 and 14), the stop codon of km23 was altered from TAA to CAC, as shown in Fig. 2B. This alteration results in a larger protein, encoding 107-amino-acid residues, instead of the wt 96-amino-acid form of km23. In addition to these alterations, the three other missense mutations detected in three other patients (cases 7, 15, and 18) were T38I, T56S, and V90A (Table 1). Because these alterations were detected in RNA, the possibility exists that not all these alterations are genomic mutations.


View this table:
[in this window]
[in a new window]

  		Table 1. Clinical characteristics of eight ovarian carcinomas with km23 alterations from 19 patients

 


View larger version (64K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Nested RT-PCR and DNA sequencing results for the mutations found in ovarian cancer patients. A, S55G/I89V alterations found in one patient. The DNA sequence from one normal tissue showing wt-km23 (left top and bottom). In one patient with ovarian cancer (case 1), the 163rd nucleotide of the coding region of km23 was altered from A to G, and the corresponding amino acid was substituted from a serine to a glycine (S55G; top right). In the same patient, another mutation from A -> G occurred at 265th nucleotide, resulting in an isoleucin -> valine substitution (I89V; bottom right). B, a stop codon mutation was detected in two patients. In two patients (cases 4 and 14), the stop codon of km23 was altered from TAA (top) to CAC (bottom). This alteration results in a larger protein, with an additional 11 amino acids at the COOH terminus, encoding a protein of 107-amino-acid residues (bottom), instead of the 96-amino-acid residues wt-km23 (top).

 
Table 1 also compares the clinical characteristics of eight patients with km23 alterations, summarizing the age of onset, histology, and stage and grade of the tumor. In these cases, only one patient was diagnosed at stage I (case 7), and two other patients were diagnosed at stage II (cases 3 and 15). Five other patients were diagnosed at a later stage (stage IIIA-IV). In addition, most of the tumors from these eight patients were moderately or poorly differentiated (Table 1). This sample population is representative of the general population, in that most ovarian cancer patients are diagnosed at a later stage of tumor progression.

No similar alterations in km23 were detectable in normal tissues. It was of interest to determine whether the km23 alterations found in the malignant tissues were also present in normal ovarian tissues. Thus, we analyzed 12 normal ovarian samples, as well as three tumor-free stroma samples (procured from cases 8, 9, and 11 by LCM), using the nested PCR strategy. In all 15 normal samples, the km23 sequence was not altered. Our findings show a statistically significant difference in the km23 alteration rate between cancer and noncancer groups (42.1% versus 0%; P < 0.005, Fisher's exact test, Table 2). Thus, the km23 alterations we have observed are specific to the malignant cells.


View this table:
[in this window]
[in a new window]

  		Table 2. Comparison of km23 alteration rate between ovarian carcinoma and normal ovarian tissues

 
{Delta}exon3-km23 disrupted the interaction with dynein intermediate chain. Because the major function of DLCs such as km23 is to bind to the DIC (9, 19, 20), it was of interest to determine whether km23 mutants could still interact with DIC. Immunoprecipitation/blot analyses were done in 293T cells transfected with Flag-tagged versions of either wt-km23 or {Delta}exon3-km23. As shown in Fig. 3, compared with the wt-km23 (lane 3), the interaction of {Delta}exon3-km23 (lane 4) with the DIC was disrupted. No interactions were detectable in lysates from cells expressing only the empty vector or when using normal mouse IgG as the immunoprecipitation antibody (lanes 1 and 2). The expression and inputs of the Flag-tagged proteins used for the immunoprecipitation reaction were verified by Western blotting as shown in Fig. 3 (bottom). Thus, the {Delta}exon3-km23 mutant we identified in the ovarian cancer patients prevent the association of km23 with the rest of the dynein motor complex.



View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. {Delta}exon3-km23 disrupts binding to the DIC. Total lysates were prepared from 293T cells that had been transfected with the indicated forms of km23 and immunoprecipitation (IP)/immunoblotting assays were done. Flag-tagged wt-km23 (lane 3) and {Delta}exon3-km23 (lane 4) were immunoprecipitated using an anti-DIC antibody. The interacting proteins were analyzed by SDS-PAGE and immunoblotting using an anti-Flag antibody (top). Normal mouse IgG and lysates from empty vector-transfected cells were used as negative controls (lanes 1 and 2, top). Equal expression and inputs of Flag-tagged proteins were verified by Western blotting (bottom). Representative of two experiments.

 
{Delta}exon3-km23 inhibited transforming growth factor-ß–dependent transcriptional activation of p3TP-lux and ARE-lux reporters. As mentioned above, we have identified km23 as a novel TGF-ß receptor–interacting protein (9) and have shown that {Delta}exon3-km23 can disrupt binding to the DIC. Based upon our model for km23 function (9, 10), it was conceivable that alterations in km23 could disrupt TGF-ß signaling. To assess whether TGF-ß signaling was still intact in the presence of the truncated form of km23 ({Delta}exon3-km23), we did TGF-ß-dependent reporter assays in Mv1Lu cells. First, we tested the effect of either wt-km23 or {Delta}exon3-km23 on TGF-ß-inducible transcriptional activity of the p3TP-lux reporter, a TGF-ß-regulated reporter frequently used to evaluate TGF-ß signaling (13, 21, 22). This reporter construct contains three consecutive 12-O-tetradecanoylphorbol-13-acetate response elements and a TGF-ß-inducible 100-bp fragment of the plasminogen activator inhibitor-1 promoter (13). Mv1Lu cells were cotransfected with p3TP-lux and the indicated amounts of either wt-km23 or {Delta}exon3-km23. As shown in Fig. 4A, TGF-ß induced a 13-fold increase in p3TP-lux activity after expression of EV, as previously reported (23). Similarly, after expression of wt-km23, TGF-ß induced a 17- to 18-fold activation of the p3TP-lux reporter. In contrast, the ability of TGF-ß to activate the p3TP-lux reporter was reduced to only 5- to 7-fold after expression of {Delta}exon3-km23. Thus, the {Delta}exon3-km23 mutant results in a repression of ligand-dependent p3TP-lux activation.



View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. {Delta}exon3-km23 inhibits TGF-ß-dependent transcriptional activation in p3TP-lux and ARE-lux reporter assays. Mv1Lu cells were transfected with the indicated amounts of either wt-km23 or {Delta}exon3-km23 along with the p3TP-lux reporter (A) or the ARE-lux reporter and FAST-1 (B). Twenty-four hours after transfection, the medium was replaced with serum-free medium for 1 hour, and the cells were incubated in the absence (open columns) or presence (black columns) of TGF-ß1 (5 ng/mL) for an additional 18 hours. The fold induction of luciferase activity by TGF-ß is indicated (parentheses on top of relevant columns). Bars, SE. Representative of at least two experiments, each done in triplicate.

 
To determine whether {Delta}exon3-km23 displayed similar effects on other TGF-ß-dependent promoters, we also examined induction of the activin-responsive element (ARE) promoter reporter from the Xenopus Mix.2 gene (14). This reporter has been shown to be activated by TGF-ß or activin in a Smad2-dependent manner (24). Mv1Lu cells were transfected with the indicated amounts of either wt-km23 or {Delta}exon3-km23, along with the ARE-lux reporter and FAST-1. As shown in Fig. 4B, compared with wt-km23, {Delta}exon3-km23 resulted in an elevation of the basal level of ARE-lux activity, suggesting aberrant regulation of this promoter in the absence of TGF-ß. Furthermore, the ability of TGF-ß to induce ARE reporter activity was significantly repressed, from 13-fold for wt-km23 to 3-fold for {Delta}exon3-km23. Thus, our results show a loss of ligand-dependent regulation of the ARE promoter as well.

S55G/I89V-km23 but not S55A-km23 disrupts the interaction with dynein intermediate chain. The results from Figs. 3 and 4 indicated that {Delta}exon3-km23 not only disrupted the interaction with DIC but also inhibited TGF-ß-dependent transcriptional activation of the p3TP-lux and ARE-lux promoters. Thus, disruption of the km23-DIC interaction blocks TGF-ß-mediated transcriptional events. To examine whether another mutant identified in human ovarian cancer could also block the DIC interaction, we did immunoprecipitation/blot analyses in 293T cells. As shown in Fig. 5A, compared with wt-km23 (lane 3), the S55G/I89V-km23 mutant (lane 4) completely blocked the interaction with DIC. As expected, no bands were observed in the negative control lanes (lanes 1 and 2).



View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. The double-mutant S55G/I89V-km23 but not the single mutant S55A-km23 disrupts the interaction with DIC, and Ser32 and Ser73 of km23 are required for DIC binding. A, S55G/I89V-km23 blocks the interaction with DIC, whereas S55A-km23 is without effect. Immunoprecipitation (IP)/blot assays were done using total lysates prepared from 293T cells that had been transfected with the indicated forms of km23. Flag-tagged wt-km23, S55G/I89V-km23, and S55A-km23 (top, lanes 3, 4, and 5, respectively) were immunoprecipitated using an anti-DIC antibody. The interacting proteins were analyzed by SDS-PAGE and immunoblotting using an anti-Flag antibody (top). Normal mouse IgG and lysates from EV-transfected cells were used as negative controls (top, lanes 1 and 2). The expression and inputs of Flag-tagged proteins were verified by Western blotting (bottom). B, both S32A-km23 and S73A-km23 disrupt the interaction with DIC. Immunoprecipitation/blot assays were done as in (A) using cell lysates prepared from 293T cells transfected with wt-km23, S32A-km23, and S73A-km23 (top, lanes 3, 4, and 5, respectively). The expression and inputs of Flag-tagged proteins were verified by Western blotting (bottom). Representative of two experiments.

 
Because we have previously shown that km23 phosphorylation is critical for km23's DIC-binding function (9, 10), it is conceivable that the S55G/I89V-km23 mutant might not have interacted with DIC due to the S55G mutation. However, this mutant also contains an I89V alteration. Accordingly, to determine whether the blockade of DIC binding was due to the S55G mutation alone, we did immunoprecipitation/blot assays using an S55A-km23 construct lacking the I89V alteration. In this construct, the serine residue at position 55 was converted to an alanine by site-directed mutagenesis. As shown in Fig. 5A, the S55A-km23 mutant could still bind to DIC (lane 5). Collectively, these data indicate that S55 is not a critical site for km23-DIC interaction, but that I89 is crucial for the DIC-binding function of km23.

Ser32 and Ser73 are both required for km23 binding to the dynein intermediate chain. As mentioned earlier, there are four conserved serine residues in km23 that could potentially be phosphorylated by TGF-ß. Of these four serine residues, three are located in exon 3 (S32, S55, and S73). Our data above have shown that S55 is not required for the km23-DIC interaction, suggesting that this site may not be a critical phosphorylation site. However, we have previously shown that the kinase activity of TßRII was required for the km23-DIC interaction (9, 10), indicating that km23 phosphorylation is a critical event for km23 function. To address whether two other serine residues, S32 and S73, were pivotal for the binding to the DIC, we did immunoprecipitation/blot assays using Flag-tagged S32A-km23 and S73A-km23 constructs, in which the corresponding serine residues were converted to alanine by site-directed mutagenesis. As shown in Fig. 5B, wt-km23 (lane 3) was immunoprecipitated by the anti-DIC antibody, whereas neither S32A-km23 nor S73A-km23 interacted with DIC (lanes 4 and 5). No interactions were detectable when cells were transfected with only the EV or using normal mouse IgG as the immunoprecipitation antibody (Fig. 5B, top, lanes 1, and 2). The expression of the Flag-tagged proteins used for the immunoprecipitation reaction was verified by Western blotting as shown in Fig. 5B (bottom). These results indicate that S32 and S73, both of which are missing in the {Delta}exon3-km23 mutant, are critical phosphorylation sites for the recruitment of km23 to the DIC.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In the present study, we report that km23, a TGF-ß receptor-interacting protein and a dynein motor protein light chain, was altered with high frequency (42.1%) in tissues from ovarian cancer patients. In 2 of 19 patients, a truncated form of km23 missing exon3 ({Delta}exon3-km23) was detected. In addition to this alteration, six other missense mutations have been detected in an additional six patients. None of these alterations were detectable in normal ovarian tissue samples. Furthermore, functional studies showed that the km23 mutant {Delta}exon3-km23 not only disrupted the interaction with DIC but also inhibited transcriptional activation of the TGF-ß-dependent reporters p3TP-lux and ARE-lux. In addition, our results indicate that alterations in km23 disrupt TGF-ß-signaling events, thereby implicating km23 in the TGF-ß resistance associated with ovarian cancer development and/or progression. This report is the first demonstration of a link between cytoplasmic dynein and ovarian cancer.

Our results of the predicted potential phosphorylation sites for PKA, PKC, and/or CKII in km23 (Fig. 1B) suggest that these kinases might be involved in modifications of km23 required for subsequent signaling events. Furthermore, the {Delta}exon3-km23 alteration, which is missing multiple phosphorylation sites that may potentially be modulated by TGF-ß or other kinases, not only resulted in an abrogation of the DIC-km23 interaction, but TGF-ß-dependent gene transcription was also inhibited. In addition, the S32A and S73A phosphorylation mutants both resulted in a blockade of km23 binding to DIC. Thus, the loss of key phosphorylation sites in km23 may be another mechanism underlying the TGF-ß resistance associated with ovarian tumors.

Interestingly, a deletion mutant found in the roadblock (robl) gene, a homologue of km23 in Drosophila, is similar to the {Delta}exon3-km23 mutant we identified in ovarian cancer patients. This Drosophila robl mutant lacks portions of intron 2 and exon 3 (25). The phenotype of this mutant exhibits a variety of defects in intracellular transport, including those of axonal transport, accumulation of cargo in intra-axonal areas, severe axonal degeneration, and an increase in mitotic index (25). Accordingly, we propose that {Delta}exon3-km23 may resemble this Drosophila mutant and result in altered intracellular trafficking and accumulation of cargoes in an aberrant manner or in aberrant compartments. Such an outcome for the km23 mutants is consistent with our results herein, which show that {Delta}exon3-km23 caused defects in both DIC binding and TGF-ß-dependent transcriptional events.

In Drosophila, the km23/robl mutant homozygotes cannot be fully rescued by the genomic or cDNA rescue constructs, showing that this mutation can act in a dominant fashion to inhibit the action of wt-km23/robl (25). In addition, here we have shown that both wt-km23 and {Delta}exon3-km23 were present together in two ovarian cancer patients (Fig. 1A). Thus, we suggest that the {Delta}exon3-km23 may behave like the km23/robl mutant in Drosophila to inhibit the normal function of wt-km23 in a dominant fashion. Future studies will address this possibility in more detail.

To date, three classes of cytoplasmic DLCs have been identified in mammals, including Tctex-1/rp3, DLC8, and km23/roadblock (9, 19, 20, 25–27), which have been shown to directly bind to the DIC at distinct regions (28). Together with the dynein intermediate chains and light-intermediate chains, they form the base of the dynein complex and are important for cargo binding and cargo transport along microtubules (19, 20, 26, 27). It has been shown that a 72-amino-acid domain of DIC, spanning amino acids 243 to 314, is required for its binding to km23/roadblock (28). However, the site on km23 that is necessary for binding to DIC has still not been identified. {Delta}exon3-km23 disrupted the interaction with DIC, suggesting that the third exon of km23 may contain the binding domain required for the DIC-km23 interaction. Alternatively, this large truncation of the km23 protein may completely disrupt the folding of km23, secondarily impairing km23-DIC interactions.

Our previous work showed that TGF-ß receptor activation was required for recruitment of km23 to the dynein motor complex (9). Thus, upon receptor activation, TGF-ß-signaling components might be transported along microtubules through the interaction of km23 with the DIC. If this were the case, the disruption of the DIC-km23 interaction by the km23 alterations we report here would be expected to alter TGF-ß-signaling events. The abrogation of the TGF-ß transcriptional events we have observed is consistent with this function for km23.

{Delta}exon3-km23 inhibited the transcriptional activity of the p3TP-lux reporter in Mv1Lu cells, suggesting that this alteration disrupts TGF-ß signaling through several transcriptional factors. The p3TP-lux reporter, an artificial construct, contains three activator protein-1 (AP-1) sites that are regulated by Ras and both the extracellular signal-regulated kinase (Erk) and JNK/mitogen-activated protein kinase pathways (17, 18, 29–31). It has also been reported, for a wide variety of cell types, that TGF-ß can activate the Ras/Erk pathways, as well as p38 and JNK (17, 18, 30–35). Furthermore, our previous data indicated that forced expression of km23 induced specific TGF-ß responses, including JNK activation and c-Jun phosphorylation. These results suggest that km23 may also be involved in a JNK pathway (9). Therefore, collectively, our results suggest that the {Delta}exon3-km23 mutant might disrupt a JNK/AP-1-dependent pathway activated by TGF-ß.

{Delta}exon3-km23 also inhibited the transcriptional activity of the Smad-dependent ARE promoter. This reporter is a natural promoter from the Xenopus Mix.2 gene (14, 24, 36–39). The activin-responsive factor (ARF), which is composed of Smad2, Smad4, and either FAST-1 or FAST-2, is required for binding to the ARE and activation of the Mix.2 promoter (14, 24, 36–39). Because we show that {Delta}exon3-km23 inhibited TGF-ß stimulation of the ARE-lux reporter, {Delta}exon3-km23 might disrupt transcription factors that are required for binding to the ARE. Future studies will address which ARF component(s) might be motor cargo for recruitment by km23.

In summary, we show that km23 is altered at high frequency in epithelial ovarian cancer. In addition, all of the alterations we have identified were specific to the tumor tissues. More importantly, all of the km23 alterations tested to date disrupted the DIC-binding function of km23. Furthermore, the {Delta}exon3-km23 mutant displayed a significant inhibition of the transactivation of TGF-ß-dependent reporters. Of note, the S55G/I89V-km23 mutant found in the ovarian cancer patients abrogated binding of km23 to the DIC, whereas the single phosphorylation site mutant S55A-km23 did not. In contrast, two other phosphorylation site mutants completely blocked the interaction with DIC. Collectively, our results indicate that km23 is an important mediator of TGF-ß-signaling events, due to its ability to bind to the DIC, and regulate TGF-ß-dependent transcriptional events. Thus, alterations in km23 would be expected to play a role in ovarian cancer formation or progression, through a mechanism involving a loss of TGF-ß signaling. As such, our results suggest that km23 may represent a critical target for the development of ovarian cancer diagnostics, prognostics, and/or therapeutics.


   Acknowledgments
 
Grant support: NIH grants CA-51452, CA-100239, CA-90765, and CA-92889 and Department of Defense award DAMD 17-03-1-0287 (K.M. Mulder). This research was also supported in part by a General Clinical Research Center grant from NIH (M01RR10732) awarded to the Pennsylvania State University College of Medicine.

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 the Cooperative Human Tissue Network for providing the ovarian tissues, W. Stewart (Department of Histology, Penn State College of Medicine) for assistance with tissue sectioning, S. Gestl (Jake Gittlen Cancer Research Institute, Penn State College of Medicine) and D. Shearer (Department of Obstetrics/Gynecology, Penn State College of Medicine) for assistance with the LCM, J. Massague (Memorial Sloan-Kettering Cancer Center, New York, NY) for providing us with p3TP-lux construct, and M. Whitman (Harvard Medical School, Boston, MA) for providing us with the ARE-lux and FAST-1 constructs.


   Footnotes
 
Note: Q. Tang is currently at the Lexicon Genetics, Inc., 8800 Technology Forest Place, The Woodlands, TX 77381. V. Espina and L. A. Liotta are currently at George Mason University, Department of Molecular and Microbiology, Center for Applied Proteomics and Molecular Medicine, Manassas, VA.

4 http://us.expasy.org/prosite Back

5 http://www.cbs.dtu.dk/services/NetPhos/ Back

Received 12/ 8/04. Revised 4/15/05. Accepted 5/22/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5–26.[Abstract/Free Full Text]
  2. Ozols RF, Bookman MA, Connolly DC, et al. Focus on epithelial ovarian cancer. Cancer Cell 2004;5:19–24.[CrossRef][Medline]
  3. Hu W, Wu W, Nash MA, Freedman RS, Kavanagh JJ, Verschraegen CF. Anomalies of the TGF-ß postreceptor signaling pathway in ovarian cancer cell lines. Anticancer Res 2000;20:729–33.[Medline]
  4. Yamada SD, Baldwin RL, Karlan BY. Ovarian carcinoma cell cultures are resistant to TGF-ß1-mediated growth inhibition despite expression of functional receptors. Gynecol Oncol 1999;75:72–7.[CrossRef][Medline]
  5. Lynch MA, Nakashima R, Song H, et al. Mutational analysis of the transforming growth factor ß receptor type II gene in human ovarian carcinoma. Cancer Res 1998;58:4227–32.[Abstract/Free Full Text]
  6. Chen T, Triplett J, Dehner B, et al. Transforming growth factor-ß receptor type I gene is frequently mutated in ovarian carcinomas. Cancer Res 2001;61:4679–82.[Abstract/Free Full Text]
  7. Cardillo MR, Yap E, Castagna G. Molecular genetic analysis of TGF-ß1 in ovarian neoplasia. J Exp Clin Cancer Res 1997;16:49–56.[Medline]
  8. Wang D, Kanuma T, Mizunuma H, et al. Analysis of specific gene mutations in the transforming growth factor-ß signal transduction pathway in human ovarian cancer. Cancer Res 2000;60:4507–12.[Abstract/Free Full Text]
  9. Tang Q, Staub CM, Gao G, et al. A novel TGFß receptor-interacting protein that is also a light chain of the motor protein dynein. Mol Biol Cell 2002;13:4484–96.[Abstract/Free Full Text]
  10. Ding W, Mulder KM. km23: a novel TGFß signaling target altered in ovarian cancer. In: Kumar R, editor. Molecular targeting and signal transduction. Boston/New York/Dordrecht/London: Kluwer Academic Publishers; 2004. p. 315–27 (Cancer Treat Res, vol. 119).
  11. Goldsworthy SM, Stockton PS, Trempus CS, Foley JF, Maronpot RR. Effects of fixation on RNA extraction and amplification from laser capture microdissected tissue. Mol Carcinog 1999;25:86–91.[CrossRef][Medline]
  12. Jin Q, Ding W, Staub CM, Gao G, Tang Q, Mulder KM. Requirement of km23 for TGFß-mediated growth inhibition and induction of fibronectin expression. Cell Signal. In press 2005.
  13. Wrana JL, Attisano L, Carcamo J, et al. TGFß signals through a heteromeric protein kinase receptor complex. Cell 1992;71:1003–14.[CrossRef][Medline]
  14. Chen X, Rubock MJ, Whitman M. A transcriptional partner for MAD proteins in TGF-ß signalling. Nature 1996;383:691–6.[CrossRef][Medline]
  15. Yue J, Mulder KM. Requirement of Ras/MAPK pathway activation by transforming growth factor ß for transforming growth factor ß1 production in a Smad-dependent pathway. J Biol Chem 2000;275:30765–73.[Abstract/Free Full Text]
  16. Blom N, Gammeltoft S, Brunak S. Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol 1999;294:1351–62.[CrossRef][Medline]
  17. Yue J, Mulder KM. Transforming growth factor-ß signaling transduction in epithelial cells. Pharmacol Ther 2001;91:1–34.[CrossRef][Medline]
  18. Massague J, Blain SW, Lo RS. TGFß signaling in growth control, cancer, and heritable disorders. Cell 2000;103:295–309.[CrossRef][Medline]
  19. King SM. The dynein microtubule motor. Biochim Biophys Acta 2000;1496:60–75.[Medline]
  20. Vallee RB, Williams JC, Varma D, Barnhart LE. Dynein: an ancient motor protein involved in multiple modes of transport. J Neurobiol 2004;58:189–200.[CrossRef][Medline]
  21. Lee HJ, Lee JK, Miyake S, Kim SJ. A novel E1A-like inhibitor of differentiation (EID) family member, EID-2, suppresses transforming growth factor (TGF)-ß signaling by blocking TGF-ß-induced formation of Smad3-Smad4 complexes. J Biol Chem 2004;279:2666–72.[Abstract/Free Full Text]
  22. Abecassis L, Rogier E, Vazquez A, Atfi A, Bourgeade MF. Evidence for a role of MSK1 in transforming growth factor-ß-mediated responses through p38{alpha} and Smad signaling pathways. J Biol Chem 2004;279:30474–9.[Abstract/Free Full Text]
  23. Souchelnytskyi S, Tamaki K, Engstrom U, Wernstedt C, ten Dijke P, Heldin CH. Phosphorylation of Ser465 and Ser467 in the C terminus of Smad2 mediates interaction with Smad4 and is required for transforming growth factor-ß signaling. J Biol Chem 1997;272:28107–15.[Abstract/Free Full Text]
  24. Felici A, Wurthner JU, Parks WT, et al. TLP, a novel modulator of TGF-ß signaling, has opposite effects on Smad2- and Smad3-dependent signaling. EMBO J 2003;22:4465–77.[CrossRef][Medline]
  25. Bowman AB, Patel-King RS, Benashski SE, McCaffery JM, Goldstein LS, King SM. Drosophila roadblock and Chlamydomonas LC7: a conserved family of dynein-associated proteins involved in axonal transport, flagellar motility, and mitosis. J Cell Biol 1999;146:165–80.[Abstract/Free Full Text]
  26. Gunawardena S, Goldstein LS. Cargo-carrying motor vehicles on the neuronal highway: transport pathways and neurodegenerative disease. J Neurobiol 2004;58:258–71.[CrossRef][Medline]
  27. Vale RD. The molecular motor toolbox for intracellular transport. Cell 2003;112:467–80.[CrossRef][Medline]
  28. Susalka SJ, Nikulina K, Salata MW, et al. The roadblock light chain binds a novel region of the cytoplasmic dynein intermediate chain. J Biol Chem 2002;277:32939–46.[Abstract/Free Full Text]
  29. Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat Rev Cancer 2003;3:859–68.[CrossRef][Medline]
  30. Mulder KM. Role of Ras and Mapks in TGFß signaling. Cytokine Growth Factor Rev 2000;11:23–35.[CrossRef][Medline]
  31. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-ß family signalling. Nature 2003;425:577–84.[CrossRef][Medline]
  32. Mulder KM, Morris SL. Activation of p21ras by transforming growth factor ß in epithelial cells. J Biol Chem 1992;267:5029–31.[Abstract/Free Full Text]
  33. Hartsough MT, Mulder KM. Transforming growth factor ß activation of p44mapk in proliferating cultures of epithelial cells. J Biol Chem 1995;270:7117–24.[Abstract/Free Full Text]
  34. Hartsough MT, Frey RS, Zipfel PA, et al. Altered transforming growth factor ß signaling in epithelial cells when ras activation is blocked. J Biol Chem 1996;271:22368–75.[Abstract/Free Full Text]
  35. Frey RS, Mulder KM. Involvement of extracellular signal-regulated kinase 2 and stress-activated protein kinase/Jun N-terminal kinase activation by transforming growth factor ß in the negative growth control of breast cancer cells. Cancer Res 1997;57:628–33.[Abstract/Free Full Text]
  36. Chen X, Weisberg E, Fridmacher V, Watanabe M, Naco G, Whitman M. Smad4 and FAST-1 in the assembly of activin-responsive factor. Nature 1997;389:85–9.[CrossRef][Medline]
  37. Yeo CY, Chen X, Whitman M. The role of FAST-1 and Smads in transcriptional regulation by activin during early Xenopus embryogenesis. J Biol Chem 1999;274:26584–90.[Abstract/Free Full Text]
  38. Watanabe M, Whitman M. FAST-1 is a key maternal effector of mesoderm inducers in the early Xenopus embryo. Development 1999;126:5621–34.[Abstract]
  39. Liu B, Dou CL, Prabhu L, Lai E. FAST-2 is a mammalian winged-helix protein which mediates transforming growth factor ß signals. Mol Cell Biol 1999;19:424–30.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
Q. Jin, W. Ding, and K. M. Mulder
Requirement for the Dynein Light Chain km23-1 in a Smad2-dependent Transforming Growth Factor-beta Signaling Pathway
J. Biol. Chem., June 29, 2007; 282(26): 19122 - 19132.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
Q. Meng, A. Lux, A. Holloschi, J. Li, J. M. X. Hughes, T. Foerg, J. E. G. McCarthy, A. M. Heagerty, P. Kioschis, M. Hafner, et al.
Identification of Tctex2beta, a Novel Dynein Light Chain Family Member That Interacts with Different Transforming Growth Factor-beta Receptors
J. Biol. Chem., December 1, 2006; 281(48): 37069 - 37080.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
I. G. Campbell, W. A. Phillips, and D. Y.H. Choong
Genetic and Epigenetic Analysis of the Putative Tumor Suppressor km23 in Primary Ovarian, Breast, and Colorectal Cancers.
Clin. Cancer Res., June 15, 2006; 12(12): 3713 - 3715.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ding, W.
Right arrow Articles by Mulder, K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ding, W.
Right arrow Articles by Mulder, K. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online