Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Translational Medicine Conference in Israel
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
Cancer Research 67, 3135-3144, April 1, 2007. doi: 10.1158/0008-5472.CAN-06-3815
© 2007 American Association for Cancer Research
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 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 Corsino, P.
Right arrow Articles by Law, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Corsino, P.
Right arrow Articles by Law, B.

Cell, Tumor, and Stem Cell Biology

Tumors Initiated by Constitutive Cdk2 Activation Exhibit Transforming Growth Factor ß Resistance and Acquire Paracrine Mitogenic Stimulation during Progression

Patrick Corsino1, Bradley Davis1, Mary Law1, Anna Chytil2, Elizabeth Forrester2, Peter Nørgaard3, Nicole Teoh1 and Brian Law1

1 Department of Pharmacology and Therapeutics and the Shands Cancer Center, University of Florida, Gainesville, Florida; 2 Department of Cancer Biology and the Vanderbilt-Ingram Cancer Center, Vanderbilt University, Nashville, Tennessee; and 3 Department of Pathology, Copenhagen University Hospital Herlev, Herlev, Denmark

Requests for reprints: Brian Law, Cancer/Genetics Research Complex, University of Florida, 1376 Mowry Road, Room 275G, P.O. Box 103633, Gainesville, FL 32610-3633. Phone: 352-273-8180; Fax: 352-273-8285; E-mail: bklaw{at}pharmacology.ufl.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Cyclin D1/cyclin-dependent kinase 2 (Cdk2) complexes are present at high frequency in human breast cancer cell lines, but the significance of this observation is unknown. This report shows that expression of a cyclin D1–Cdk2 fusion protein under the control of the mouse mammary tumor virus (MMTV) promoter results in mammary gland hyperplasia and fibrosis, and mammary tumors. Cell lines isolated from MMTV–cyclin D1–Cdk2 (MMTV-D1K2) tumors exhibit Rb and p130 hyperphosphorylation and up-regulation of the protein products of E2F-dependent genes. These results suggest that cyclin D1/Cdk2 complexes may mediate some of the transforming effects that result from cyclin D1 overexpression in human breast cancers. MMTV-D1K2 cancer cells express the hepatocyte growth factor (HGF) receptor, c-Met. MMTV-D1K2 cancer cells also secrete transforming growth factor ß (TGFß), but are relatively resistant to TGFß antiproliferative effects. Fibroblasts derived from MMTV-D1K2 tumors secrete factors that stimulate the proliferation of MMTV-D1K2 cancer cells, stimulate c-Met tyrosine phosphorylation, and stimulate the phosphorylation of the downstream signaling intermediates p70s6k and Akt on activating sites. Together, these results suggest that deregulation of the Cdk/Rb/E2F axis reprograms mammary epithelial cells to initiate a paracrine loop with tumor-associated fibroblasts involving TGFß and HGF, resulting in desmoplasia. The MMTV-D1K2 mice should provide a useful model system for the development of therapeutic approaches to block the stromal desmoplastic reaction that likely plays an important role in the progression of multiple types of human tumors. [Cancer Res 2007;67(7):3135–44]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cyclin-dependent kinase 2 (Cdk2) becomes activated during mammary tumorigenesis through a number of mechanisms including cyclin E overexpression and proteolytic processing (1), p21 and p27 down-regulation and mislocalization (2–7), and cyclin A1 re-expression (8). Recent studies indicate that Cdk2 is likely to be an important target for anticancer agents (9–11). Therefore, it is important to understand how Cdk2 activation through various mechanisms leads to tumor formation and the biochemical and cellular mechanisms involved. Complexes between cyclin D1 and Cdk2 were shown to be present in mammary carcinoma cells some time ago (12), but the function of these complexes is unclear. Cyclin D1 overexpression occurs in ~50% of human breast cancers (13); thus, cyclin D1/Cdk2 complexes might contribute to the oncogenic effects of cyclin D1 overexpression. Different cyclins and Cdks interact with each other rather promiscuously, making it difficult to ascribe specific functions to particular cyclin/Cdk complexes. To circumvent this problem, we designed a cyclin D1-Cdk2 fusion protein in which the cyclin D1 domain stimulates the phosphorylation and kinase activity of the Cdk2 domain through an intramolecular mechanism (14). We constructed a transgenic mouse model in which mammary expression of the cyclin D1-Cdk2 fusion protein is driven by the mouse mammary tumor virus (MMTV) promoter (MMTV-D1K2; Fig. 1A ). MMTV-D1K2 transgenic mice exhibit mammary fibrosis and hyperplasia and develop mammary tumors associated with significant desmoplasia. The MMTV-D1K2 transgenic mouse model may prove useful for testing Cdk2 inhibitors and for the development and testing of novel therapeutic agents targeting tumor cell/fibroblast paracrine loops.


Figure 1
View larger version (63K):
[in this window]
[in a new window]

  		Figure 1. MMTV-D1K2 transgenic model and mammary phenotype. A, model depicting the MMTV promoter driving the expression of the transgene encoding the NH2-terminal FLAG epitope tag, the cyclin D1 domain, a flexible linker, the Cdk2 domain, and the COOH-terminal His6 affinity tag. B, representative mammary whole mounts from 56-wk-old wild-type female mice of the 34 and 44 lines of the MMTV-D1K2 transgenic mice stained with hematoxylin (top). Inset, branching morphology at higher magnification. Bottom, H&E-stained histologic sections at low and high magnification. LN, mammary lymph nodes. C, H&E- and trichrome-stained mammary gland sections from 102-wk-old MMTV-D1K2 transgenic mice of the 44 line displaying regions of epithelial hyperplasia associated with fibrosis. D, Kaplan-Meier curve showing tumor incidence in wild-type and line 44 mice as a function of age in months. Inset legend, number of virgin female animals in each group.

 

   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Construction of MMTV–cyclin D1–Cdk2 transgenic mice. The cDNA encoding the cyclin D1-Cdk2 fusion protein was excised from the pAdTrack vector described previously (14) with EcoRI and EcoRV and subcloned into the EcoRI and BstXI sites of MMTV-transforming growth factor {alpha} (15). This resulted in the replacement of the transforming growth factor {alpha} (TGF{alpha}) transgene with the cDNA encoding the cyclin D1-Cdk2 fusion, creating the MMTV–cyclin D1–Cdk2 (MMTV-D1K2) vector. The MMTV-D1K2 vector was verified by DNA sequencing. The transgene and MMTV promoter were excised from the MMTV-D1K2 vector using AatII and NruI. The purified AatII/NruI fragment was submitted to the Vanderbilt University Transgenic Mouse/Embryonic Stem Cell Shared Resource, and transgenic mice were created in the inbred FvB strain. Transgenic animals were initially identified by Southern blotting of genomic tail DNA and routinely screened by PCR using primers complementary to the region encoding the FLAG epitope tag (5'-GACTATAAGGACGATGATGAC-3') and the flexible linker joining the cyclin D1 and Cdk2 domains (5'-CCTCCAGAACCTCCACCACC-3'). Multiple lines of transgenic mice were obtained. Lines designated 34 and 44 were selected for further study.

Whole-mount staining and histologic analyses. Whole-mount preparation and staining with hematoxylin was done as described (16). Tissue samples were fixed with 4% paraformaldehyde in PBS for 14 h at 4°C and switched to 70% ethanol for 24 h, followed by an additional 24-h incubation in 70% ethanol at 4°C. The University of Florida Molecular Pathology Core embedded the tissue samples in paraffin, prepared 5-µm sections, and stained the sections with H&E or Trichrome.

Isolation and culture of cancer and tumor-associated fibroblast cell lines. Cancer cells were cultured from the tumors as described previously (16). Fibroblasts were removed from colonies of tumor cells by differential trypsinization and retained and cultured separately. Established cell lines were propagated in DMEM supplemented with 10% fetal bovine serum (FBS). Recombinant adenoviruses and retroviruses and the procedures used for infection and selection were described previously (14, 17). Roscovitine, a TGFß receptor I kinase inhibitor, and rapamycin were obtained from EMD Biosciences Inc. (La Jolla, CA). Recombinant human TGFß1 and recombinant human HGF were obtained from Chemicon International (Temecula, CA).

Immunoblot analysis of tumor samples and tumor-derived cell lines and Rb kinase assays. Preparation of tumor and cell lysates and subsequent immunoblot analysis were done as described previously (14, 17). Antibodies to the FLAG epitope (F-3165) and {alpha}-smooth muscle actin ({alpha}-SMA; A-2547) were obtained from Sigma-Aldrich, Inc. (St. Louis, MO). Her2/c-neu antibody (MS-730) was obtained from Neomarkers (Fremont, CA). E-cadherin antibody was obtained from BD Biosciences (San Jose, CA). P-cadherin antibody (sc-7893) was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Antibodies to c-Met, c-Met phosphorylated on tyrosine residues 1234/1235, Akt, Akt phosphorylated on threonine 308, signal transducers and activators of transcription 3 (STAT3), and STAT3 phosphorylated on Tyr705 were purchased from Cell Signaling Technologies (Beverly, MA). The anti-HIRA antibody (18) was generously provided by Dr. Peter Adams (Fox Chase Cancer Center, Philadelphia, PA). The sources of the other antibodies were listed previously (14, 17, 19, 20). Rb kinase assays of anti-FLAG and anti-Cdk4 immunoprecipitates were done as described previously (14, 17).

Transcriptional reporter assays and cell proliferation analyses. The Mv1Lu cell line with the plasminogen activator inhibitor-1 (PAI-1) promoter driving a luciferase reporter gene (PAI-Luc cells) was provided by Dr. D. Rifkin (New York University, New York, NY) and has been described previously (21). Conditioned cell culture medium was prepared by incubating equal numbers of cells with the same volume of medium over a period of 5 days. The medium was collected, and debris was removed by centrifugation, followed by passage through a 0.2-µm filter. PAI-Luc cells were incubated with conditioned medium samples for 24 h, luciferase assays were done, and the results were normalized to protein concentration as described (14).

3H-thymidine incorporation assays were done as described previously (17). Flow cytometry analysis of propidium iodide–stained nuclei and data analysis were done by the University of Florida Flow Cytometry Core Laboratory.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Characterization of MMTV-D1K2 mammary glands. Two transgenic mouse MMTV-D1K2 lines termed 34 and 44 engineered to express the cyclin D1-Cdk2 fusion protein in the mammary gland (Fig. 1A) were characterized. Mammary glands from 56-week-old virgin females were examined as hematoxylin-stained whole mounts and by H&E staining of tissue sections (Fig. 1B). Mammary glands from wild-type FvB females exhibit a fully developed ductal tree, but little side branching. Mammary glands from transgenic lines 34 and 44 exhibit a significant degree of side branching. H&E-stained mammary sections indicate that epithelial structures are sparse in wild-type glands, but dense assemblies of epithelial structures are present in the transgenic animals. Mammary glands from aged transgenic females exhibit a progressively more abnormal phenotype. Glands from 102-week-old line 44 MMTV-D1K2 animals exhibit hyperplastic lesions. Examination of H&E-stained sections of these mammary glands indicate that these lesions consist of ductal structures surrounded by fibroblasts (Fig. 1C). The strong blue staining by trichrome overlaps with the fibroblasts and is consistent with collagen deposition associated with the fibrosis. The hyperplastic lesions observed frequently progressed into tumors, and by 2 years of age, about 70% of the animals in the MMTV-D1K2 44 line had developed mammary tumors (Fig. 1D). Salivary gland tumors occurred less frequently and were not included in the data used to plot Fig. 1D. Several mammary and salivary tumors were also observed in the MMTV-D1K2 34 line, indicating that tumor formation in the MMTV-D1K2 animals is not an artifact of the transgene insertion site. No tumors of any kind were observed in wild-type virgin female littermates. Mammary tumors were not observed in male transgenic mice.

Characterization of MMTV-D1K2 tumors. MMTV-D1K2 mammary tumors exhibit a prominent stromal reaction (Fig. 2A ), and in many cases, a large fraction of the tumor bulk was made up of cells of mesenchymal morphology (arrows) distributed throughout the tumors. In one mouse, L44-7f, a mammary gland lesion, was observed that resembled human mammary sclerosing adenosis. These lesions are considered benign yet premalignant in humans and consist of atypical glandular structures that proliferate to various extents in a fibrotic stroma with proliferating myoepithelial cells. Tumors from MMTV-neu transgenic animals were examined in parallel because the human homologue of the rat neu gene, Her2 is overexpressed in ~30% of breast cancers (22, 23), cyclin D1 is thought to play an important role in neu-induced tumorigenesis (24, 25), and because we had previously isolated and characterized cancer cell lines derived from MMTV-neu tumors. The phenotype of MMTV-D1K2 mammary tumors differed from the MMTV-neu tumors, which exhibited a relatively homogenous center surrounded by a thin layer of stroma (arrow). Examination of H&E-stained tissue sections from several representative tumors indicated that the majority of the MMTV-D1K2 tumors were mammary gland adenocarcinomas ranging from low to high grade (Fig. 2B). Adenosquamous differentiation was observed in one tumor. Two tumors were observed in salivary glands, and one of these (L34-57f) seemed to be of salivary origin.


Figure 2
View larger version (61K):
[in this window]
[in a new window]

  		Figure 2. MMTV-D1K2 mammary and salivary tumors. A, H&E- and trichrome-stained histologic sections of MMTV-D1K2 and MMTV-neu mammary tumors. B, morphologic characterization of several representative tumors arising in the 34 (L34-) and 44 (L44-) transgenic lines. C, left, immunoblot analysis of extracts from four different MMTV-D1K2 and four different MMTV-neu mammary tumors demonstrating expression of the FLAG-tagged cyclin D1-Cdk2 transgene product (FLAG (D1K2)) and c-neu. Tumor lysates were also analyzed for the levels of E-cadherin, {alpha}-SMA, and actin as a loading control. C, center, immunoblot analysis demonstrating the hyperphosphorylation of Rb on multiple residues in a MMTV-D1K2 tumor extract relative to a MMTV-neu tumor extract. P-Rb and P-p130 represent the phospho-forms of Rb and p130, and Rb and p130 represent the corresponding unphosphorylated forms. Multiple products of E2F-dependent genes are up-regulated in the MMTV-D1K2 tumor relative to the MMTV-neu tumor including BRCA1, p107, and E2F1. Hira serves as a loading control. C, right, lysates from MMTV-neu and MMTV-D1K2 tumors were subjected to immunoprecipitation with anti-FLAG–agarose to isolate complexes containing the cyclin D1-Cdk2 fusion protein. Immunoblot analysis indicated that these complexes contain the cyclin D1-Cdk2 fusion protein (FLAG), p21, p27, and PCNA.

 
Immunoblot analysis of tumor lysates (Fig. 2C, left) showed that the FLAG-tagged cyclin D1-Cdk2 fusion protein is expressed in the MMTV-D1K2 tumors. MMTV-D1K2 tumors contained similar levels of E-cadherin and actin as MMTV-neu tumors but, on average, expressed higher levels of {alpha}-SMA. Higher levels of {alpha}-SMA could result from either a greater proportion of myofibroblasts or myoepithelial cells in the MMTV-D1K2 tumors than in the MMTV-neu tumors. Immunoblot analyses examining a MMTV-D1K2 tumor extract with a MMTV-neu tumor extract serving as the control (Fig. 2C, center) showed that Rb is hyperphosphorylated in the MMTV-D1K2 tumor, and that the protein products of the E2F-dependent genes BRCA1, p107, and E2F1 are up-regulated relative to the level observed in the MMTV-neu tumor lysate. Immunoprecipitation of these tumor lysates with anti-FLAG–agarose resin showed that the FLAG-tagged cyclin D1-Cdk2 fusion protein present in the MMTV-D1K2 tumor lysate was present in complexes with p21, p27, and proliferating cell nuclear antigen (PCNA) (Fig. 2C, right). It is likely that the fusion protein drives tumor formation by directly phosphorylating substrates and by sequestering p21 and p27.

Derivation and characterization of cancer cell lines from MMTV-D1K2 tumors. The large proportion of fibroblasts in the MMTV-D1K2 tumors made it difficult to study the biochemical properties of the cancer cells. We isolated a series of cancer cell lines and tumor-derived fibroblast cell lines to allow a detailed analysis of the properties of each cell type in isolation and to study how these two cell types might functionally interact in tumors. The neuT cell line was derived from an MMTV-neu tumor and serves as a reference for comparison with results obtained with cell lines derived from MMTV-D1K2 tumors. Five cancer cell lines were derived from five different MMTV-D1K2 tumors and termed D1K2-T1, D1K2-T2, D1K2-T3, D1K2-T4, and D1K2-T5. D1K2-T1 was isolated from a mammary tumor arising in the L44-25f MMTV-D1K2 transgenic mouse. D1K2-T3 was isolated from a salivary tumor arising in the L34-57f MMTV-D1K2 transgenic mouse. The D1K2-T1 and D1K2-T3 cell lines exhibit the cuboidal morphology typical of luminal epithelial cells and are similar in appearance to the neuT cells (Fig. 3A ). The D1K2-T2, D1K2-T4, and D1K2-T5 cell lines exhibit myoepithelial morphology and express markers of both the luminal and myoepithelial lineages. These cell lines are the subject of ongoing investigation and will not be described further here.


Figure 3
View larger version (47K):
[in this window]
[in a new window]

  		Figure 3. Isolation and characterization of MMTV-D1K2 cancer cell lines. A, phase contrast micrographs of cell lines derived from an MMTV-neu tumor (neuT) and two different MMTV-D1K2 tumors (D1K2-T1 and D1K2-T3). B, flow cytometry analysis of rapidly growing neuT, D1K2-T1, and D1K2-T3 cells stained with propidium iodide. C, left, immunoblot analysis of the neuT/Hyg and neuT/D1K2 cell lines prepared by retroviral transduction of the neuT cells with an empty retroviral vector or a vector encoding the cyclin D1-Cdk2 fusion protein, respectively, and the D1K2-T1 and D1K2-T3 cell lines. The results indicate the expression of the cyclin D1-Cdk2 fusion protein in the appropriate samples and show Rb and p130 hyperphosphorylation in the D1K2-T1 cell line. E-cadherin is expressed at similar levels in the four cell lines. Actin serves as a loading control. C, right, immunoblot analysis demonstrating the levels of expression of the cyclin D1-Cdk2 fusion protein (*) relative to the levels of endogenous Cdk2 (–), and phosphorylation of the cyclin D1-Cdk2 transgene product on the activating Thr160 (P-Cdk2(T160)) and inhibitory Tyr15 (P-Cdk2(Y15)) phosphorylation sites of the Cdk2 domain. C-Met is expressed at higher levels in the D1K2-T3 and D1K2-T1 cell lines than in the neuT cell line. D, assay of the kinase activity of the cyclin D1-Cdk2 fusion protein and endogenous Cdk4. Extracts of the indicated cell lines were immunoprecipitated using the FLAG antibody to isolate the cyclin D1-Cdk2 fusion protein or a Cdk4 antibody to isolate endogenous Cdk4. Immunoprecipitates were assayed for kinase activity using glutathione S-transferase-Rb as the substrate, and site-specific Rb phosphorylation was detected by immunoblotting with phosphospecific antibodies. Controls included the use of neuT cell extracts in FLAG immunoprecipitations, and omission of the Cdk4 antibody (No 1°).

 
We did propidium iodide staining followed by flow cytometry analysis to determine whether expression of the cyclin D1-Cdk2 fusion protein significantly altered the cell cycle profile (Fig. 3B). Rapidly growing neuT cells exhibit a typical cell cycle profile. In contrast, the D1K2-T1 cells and, to a lesser extent, the D1K2-T3 cells exhibited multiple peaks, suggesting that these cell lines are aneuploid. Similar observations were made with the D1K2-T4 and D1K2-T5 cell lines (data not shown). In all cases, each of the observed peaks of fluorescence intensity was shifted to exactly twice the fluorescence intensity by a 24-h treatment of the cells with the M-phase arresting agent nocodazole (data not shown). The results suggest that the peaks represent different populations of cancer cells with different states of ploidy. Overall, four of the five cell lines derived from MMTV-D1K2 tumors exhibited evidence of aneuploidy by flow cytometry. This observation is consistent with the observations that cyclin D1 overexpression induces aneuploidy (26), and that hyperactivation of Cdk2 induces aneuploidy (27). Thus, the induction of aneuploidy may be an additional mechanism by which the cyclin D1-Cdk2 fusion protein, and by extension cyclin D1/Cdk2 complexes, drive tumorigenesis.

Cell lines were derived from the neuT cells by transduction with a control retroviral vector (neuT/Hyg) or the same vector encoding the cyclin D1-Cdk fusion protein (neuT/D1K2). Immunoblot analysis of lysates from these cells showed that expression of the cyclin D1-Cdk2 fusion protein was higher in the D1K2-T1 cells than in the D1K2-T3 cells (Fig. 3C, left). The neuT/D1K2 cells express the cyclin D1-Cdk2 fusion protein at levels similar to those observed in the D1K2-T3 cells. Interestingly, Rb phosphorylation levels are similar in the neuT/Hyg, neuT/D1K2, and D1K2-T3 cell lines but higher in the D1K2-T1 cells. D1K2-T1 cells also exhibit altered p130 electrophoretic mobility consistent with p130 hyperphosphorylation. Together, the results indicate that high-level expression of the cyclin D1-Cdk2 fusion protein can occur in tumors and is associated with Rb hyperphosphorylation and cell cycle deregulation. However, lower levels of cyclin D1-Cdk2 expression that do not cause obvious biochemical perturbations as observed in the D1K2-T3 cells are apparently sufficient to initiate tumor formation. Immunoblot analysis of lysates from the neuT, D1K2-T1, and D1K2-T3 cell lines showed that expression of the HGF receptor c-Met is elevated in the D1K2-T1 and D1K2-T3 cell lines relative to the neuT cell line (Fig. 3C, right). Cdk2 immunoblots indicated that expression of the cyclin D1-Cdk2 fusion protein in the D1K2-T1 cells is about twice that of endogenous Cdk2 (–), whereas the fusion protein was barely detectable in the D1K2-T3 cells using the Cdk2 antibody. Analysis of the phosphorylation status of the Cdk2 domain of the fusion protein on the regulatory sites Tyr15 and Thr160 using phosphospecific antibodies indicated that the level of phosphorylation of the fusion protein on both sites is higher than that of endogenous Cdks. This is consistent with our previous observations indicating that the cyclin D1 domain of the fusion protein stimulates the phosphorylation of the Cdk2 domain through an intramolecular mechanism (14). The observation that inhibitory phosphorylation of Tyr15 of the fusion protein is enhanced suggested that the cyclin D1-Cdk2 fusion protein might not be catalytically active. We addressed this issue by performing kinase assays of the cyclin D1-Cdk2 fusion protein immunoprecipitated using the FLAG antibody (Fig. 3D). We also examined whether expression of the cyclin D1-Cdk2 fusion protein altered the activity of endogenous Cdk4. Although the fusion protein was heavily phosphorylated on both Tyr15 and Thr160, it phosphorylated Rb in vitro. This result suggests that a population of the fusion protein molecules exists in which the activating site, Thr160, is phosphorylated, but the inhibitory site, Tyr15, is not phosphorylated. The levels of Cdk4 kinase activity toward Rb were similar in the neuT, D1K2-T1, or D1K2-T3 cell lines, indicating that expression of the cyclin D1-Cdk2 fusion has minimal effects on endogenous Cdk4 activity. These results show that the cyclin D1-Cdk2 fusion protein expressed in the MMTV-D1K2 cancer cell lines is enzymatically active, is overexpressed only marginally relative to endogenous Cdk2 levels, and that the fusion protein displays enhanced regulatory phosphorylation relative to endogenous Cdks.

Derivation and characterization of fibroblast cell lines from MMTV-D1K2 tumors. Given the large proportion of fibroblasts in the MMTV-D1K2 tumors and the close proximity between the epithelial cells and fibroblasts in hyperplastic lesions (Fig. 1C) all the way to fully developed tumors (Fig. 2A), it is likely that tumor-associated fibroblasts have a significant influence on the progression and growth of MMTV-D1K2 tumors. We isolated fibroblast cell lines from the MMTV-D1K2 tumors to examine the properties of these cells. The tumor-derived fibroblast lines 1 and 2 (D1K2-TDF1 and D1K2-TDF2) exhibit the elongated morphology typical of fibroblasts and lack the extensive cell-cell contacts characteristic of colonies of epithelial cells (Fig. 4A ). Immunoblot analysis showed that D1K2-TDF1 and D1K2-TDF2 cells did not express proteins typically present in cancer cells of epithelial origin such as c-Met, E-cadherin, and P-cadherin (Fig. 4B). The tumor-derived fibroblast cell lines also did not express the cyclin D1-Cdk2 fusion protein, but expressed very high levels of {alpha}-SMA. {alpha}-SMA is expressed in fibroblasts that have differentiated into myofibroblasts. Such differentiation might arise as a result of cell culture in vitro. However, the presence of myofibroblasts in the MMTV-D1K2 tumors could explain the relatively high levels of {alpha}-SMA present in the tumors and could also explain the high levels of collagen detected by trichrome staining (Figs. 1 and 2) because myofibroblasts are thought to be the major cell type responsible for collagen deposition during fibrosis (28–30). The differentiation of stromal fibroblasts in the vicinity of tumors is thought to be caused by TGFß secreted by tumor cells because TGFß1 is capable of inducing the differentiation of mammary stromal fibroblasts to myofibroblasts in vitro (31, 32), fibroblast differentiation to myofibroblasts occurs in the vicinity of tumors in a graded manner (33), and tumors are known to secrete significant amounts of TGFß (34).


Figure 4
View larger version (60K):
[in this window]
[in a new window]

  		Figure 4. Isolation and characterization of tumor-derived fibroblast cell lines. A, phase contrast micrographs of fibroblast cell lines derived from two different MMTV-D1K2 tumors (D1K2-TDF1 and D1K2-TDF2). B, immunoblot analysis of extracts from the tumor-derived fibroblast cell lines D1K2-TDF1 and D1K2-TDF2, the MMTV-D1K2 tumor cell line D1K2-T1, and the BT549 human mammary carcinoma cell line. The results show a lack of E-cadherin, P-cadherin, c-Met, or the cyclin D1-Cdk2 transgene in the tumor-derived fibroblasts. The D1K2-TDF1 and D1K2-TDF2 cells express high levels of {alpha}-SMA. D1K2-T1 and BT549 lysates served as positive controls for the immunoblot analysis, and actin serves as a loading control. C, hypothetical model for the mechanisms by which D1K2 overexpression leads to the acquisition of stromal-derived mitogenic stimulation. D1K2 overexpression desensitizes cancer cells to autocrine TGFß inhibition. Cancer cell-derived TGFß either directly or indirectly increases the abundance of stromal myofibroblasts. Myofibroblasts secrete HGF, which stimulates the proliferation of cancer cells through the HGF receptor c-Met.

 
The large proportion of tumor-associated fibroblasts and c-Met expression in the tumor cells suggested the possible existence of a paracrine loop in the MMTV-D1K2 tumors as outlined in Fig. 4C. Secretion of HGF is thought to be an important function of tumor-associated fibroblasts, and myofibroblasts in particular (35–38), that drives tumor progression in a paracrine manner.

HGF and TGFß sensitivity of MMTV-D1K2 cancer cell lines. We were unable to significantly up-regulate c-Met expression in neuT cells upon overexpression of the cyclin D1-Cdk2 fusion protein by transient adenoviral transduction (Fig. 5A ). Likewise, treatment of the D1K2-T1 cells with the Cdk inhibitor roscovitine did not significantly decrease c-Met expression. These observations suggest that either higher levels of cyclin D1-Cdk2 fusion protein expression for extended periods are required to induce c-Met up-regulation, or that c-Met up-regulation provides a selective advantage for the MMTV-D1K2 tumors during progression but does not result directly from transgene expression. Figure 4B shows that the D1K2-T1 cells express c-Met at a level similar to that observed in the BT549 human mammary carcinoma cell line. c-Met expression was examined in a panel of human mammary carcinoma cell lines and noncancerous mammary epithelial cell lines (HBL100, NMuMG, and MCF10A; Fig. 5B). The levels of c-Met expression in BT549 cells were only exceeded by the levels observed in the MDA-MB-231 and MDA-MB-468 cell lines.


Figure 5
View larger version (40K):
[in this window]
[in a new window]

  		Figure 5. Receptor expression and responsiveness in tumor-derived cancer cell lines. A, neuT cells were transduced with adenoviruses encoding green fluorescent protein (Ad.GFP), the cyclin D1-Cdk2 fusion protein (Ad.D1K2), or a combination of adenoviruses encoding cyclin D1 (Ad.D1) and Cdk2 (Ad.Cdk2-His6) or cyclin D1 (Ad.D1) and Cdk4 (Ad.Cdk4-His6) as indicated. Cell lysates were prepared 48 h postinfection. D1K2-T1 and D1K2-T3 cells were washed and treated for 24 h with medium containing 0.1% FBS. The cells were either fed with medium containing 0.1% FBS for 24 h (Control) or treated with the same medium containing 2.5 ng/mL TGFß for 24 h, 50 ng/mL HGF for 15 min, treatment with 2.5 ng/mL TGFß for 24 h followed by a 15-min treatment with 50 ng/mL HGF (TGFß -> HGF), or treatment with 10 µmol/L roscovitine for 24 h. Lysates of the adenovirus or drug treated cells were prepared and analyzed by immunoblot for c-Met phosphorylated on tyrosine residues 1234/1235 (P-MET(1234/1235)), c-MET, Akt phosphorylated on Thr308 (P-Akt(308)), Akt, p70s6k phosphorylated on Thr389 (P-p70s6k(389)), p70s6k, c-neu (Her2/neu), STAT3 phosphorylated on Tyr705 (P-Stat3(705)), STAT3, plasminogen activator inhibitor-1 (Pai-1), the FLAG-tagged cyclin D1-Cdk2 fusion protein, or FLAG-tagged cyclin D1 (FLAG(cyclin D1)), the His6 affinity tag, or actin as a loading control. B, lysates of a panel of mammary epithelial (HBL100, NMuMG, and MCF10A) or human mammary carcinoma cell lines (T47D, MCF7, MDA-MB-361, MDA-MB-468, BT549, MDA-MB-231, MDA-MB-435S, and MDA-MB-436) were subjected to immunoblot analysis with the indicated antibodies. Actin serves as a loading control. C, D1K2-T1 or D1K2-T3 cells were treated for 24 h with 0.1% FBS-containing medium (Control) or the same medium containing 100 nmol/L rapamycin, 50 ng/mL HGF, or 50 ng/mL HGF + 100 nmol/L rapamycin and the rate of 3H-thymidine incorporation into DNA was measured. Columns, percent control 3H-thymidine incorporation; bars, SD of triplicate determinations.

 
Treatment of the D1K2-T1 line with HGF for 15 min stimulated c-Met tyrosine phosphorylation and activated downstream signaling events such as the phosphorylation of the protein kinases Akt and p70s6k on activating sites (Fig. 5A), showing that c-Met is functional in D1K2-T1 cells. D1K2-T3 cells exhibited a lower level of c-Met tyrosine phosphorylation in response to HGF treatment, and activation of downstream signaling was not observed. Consistent with these observations, HGF stimulated the proliferation of D1K2-T1 cells nearly 3-fold, but had only a modest effect on D1K2-T3 cells (Fig. 5C). HGF-induced proliferation of the D1K2-T1 cells was completely blocked by rapamycin, an inhibitor of the mTOR/p70s6k signaling pathway.

TGFß is often secreted by cancer cells and causes fibrosis and stromal fibroblast differentiation to myofibroblasts. TGFß potently inhibits the proliferation of nontransformed epithelial cells; therefore, inactivation of the TGFß growth inhibitory response is thought to be a key event in the progression of carcinomas arising in multiple tissues. Treatment of the D1K2-T1 cells with TGFß for 24 h resulted in the up-regulation of PAI-1 expression, a known transcriptional target of the TGFß signaling pathway (refs. 21, 39; Fig. 5A). This response was not observed in D1K2-T3 cells. Interestingly, TGFß treatment had no effect on Akt or p70s6k phosphorylation, but potentiated the HGF response observed. The most striking observation was the complete lack of STAT3 tyrosine phosphorylation in the D1K2-T1 cell line. STAT3 phosphorylation on Tyr705 causes dimerization and translocation of STAT3 to the nucleus, where it activates the expression of pro-proliferative and antiapoptotic genes. The reason for the lack of STAT3 tyrosine phosphorylation in the D1K2-T1 cells is unclear, but it was previously reported that STAT3 tyrosine phosphorylation levels correlate inversely with cellular Cdk2 activity (40). The observation that adenovirus-mediated overexpression of the cyclin D1-Cdk2 fusion protein in neuT cells partially suppresses STAT3 tyrosine phosphorylation suggests that this effect is a direct result of expression of the cyclin D1-Cdk2 fusion protein. Together, the results in Fig. 5A indicate that D1K2-T1 cells mount biochemical responses to TGFß and HGF and exhibit alterations in STAT3 regulation. This is significant given that this cell line was derived from a tumor exhibiting a dramatic desmoplastic reaction illustrated in Fig. 2A (left), and TGFß and HGF are thought to play key roles in the formation of paracrine loops resulting in desmoplasia. According to this model, the formation of large tumors with a significant stromal reaction is favored by a self-amplifying paracrine loop with the following characteristics: (a) TGFß secretion by the tumor cells; (b) tumor cells desensitized to TGFß antiproliferative actions; (c) TGFß-induced differentiation of stromal fibroblasts to myofibroblasts and stimulation of fibroblast proliferation or recruitment of additional stromal fibroblasts; and (d) secretion of paracrine factors, such as HGF by stromal fibroblasts and myofibroblasts that stimulate the proliferation of the cancer cells.

TGFß and HGF production by MMTV-D1K2 tumor-derived fibroblast and cancer cell lines. The cancer and tumor-derived fibroblast cell lines and medium conditioned by these cell lines were subjected to further analysis to determine whether they meet the necessary criteria for the maintenance of a TGFß/HGF paracrine loop between the tumor cells and the tumor fibroblasts. Medium conditioned by the neuT, D1K2-T1, and D1K2-T3 tumor cell lines and the D1K2-TDF1 and D1K2-TDF2 fibroblast cell lines, derived from two different MMTV-D1K2 tumors of the 44 line, was collected and assayed for the presence of TGFß and other growth regulatory factors secreted by the cells (Fig. 6A–D ). Mink lung epithelial cells (Mv1Lu) stably transfected with a TGFß-responsive PAI-1 luciferase reporter construct were treated with the conditioned medium. Medium conditioned by the neuT, D1K2-T1, and D1K2-T3 cell lines increased luciferase expression by several-fold relative to unconditioned medium (Fig. 6A). The addition of a TGFß type I receptor kinase inhibitor blocked the transcriptional response to medium conditioned by the cancer cell lines, suggesting that TGFß or a TGFß-like factor secreted by the cancer cells caused the transcriptional response. Medium conditioned by the fibroblast cell lines activated the PAI-1 promoter, suggesting that the tumor-derived fibroblasts also secrete TGFß. This is interesting in light of the observation that these fibroblasts tend to differentiate into myofibroblasts in culture (Fig. 4), and TGFß is known to induce the differentiation of mammary stromal fibroblasts into myofibroblasts (31, 33). Thus, some tumor fibroblasts may perpetuate their myofibroblastic differentiation via a TGFß autocrine loop after that loop is initiated by paracrine TGFß secreted by adjacent cancer cells. We next examined whether the levels of TGFß present in the conditioned medium samples were sufficient to inhibit the proliferation of the highly TGFß-sensitive Mv1Lu cells, or the D1K2-T1 cells (Fig. 6B). The TGFß type I receptor kinase inhibitor was used to distinguish TGFß growth inhibitory effects from other growth inhibitory effects such as depletion of nutrients from the conditioned medium. The inhibitor itself had no effect on the proliferation of Mv1Lu cells and a modest inhibitory effect on the proliferation of the D1K2-T1 cells. Medium conditioned by neuT, D1K2-T1, D1K2-T3, and D1K2-TDF1 cells strongly inhibited the proliferation of the Mv1Lu cells and, to a lesser extent, the D1K2-T1 cells. This effect was partially reversed by the TGFß type I receptor kinase inhibitor, suggesting that the concentration of TGFß present in the conditioned medium is sufficient to inhibit cell proliferation. Treatment of the cells with 0.5 ng/mL TGFß inhibited DNA synthesis in the Mv1Lu cells by about 90% and inhibited DNA synthesis in the D1K2-T1 cells by about 30%. This relative TGFß resistance of the D1K2-T1 cells is expected based on our previous observation that overexpression of the cyclin D1-Cdk2 fusion protein renders cells partially TGFß resistant (14). Interestingly, medium conditioned by the D1K2-TDF1 tumor-derived fibroblasts stimulated the proliferation of the D1K2-T1 cells, but inhibited the proliferation of the Mv1Lu cells. Figure 6A shows the presence of a low level of TGFß in this sample. It is likely that the D1K2-TDF1–conditioned medium contains sufficient TGFß to inhibit the more sensitive Mv1Lu cells, but not enough to inhibit the D1K2-T1 cells. In addition, the D1K2-TDF1–conditioned medium apparently contains factors that stimulate the proliferation of the neuT cells as well as the D1K2-T1 cells (Fig. 6C). This mitogenic effect is partially blocked by the TGFß type I receptor kinase inhibitor although this inhibitor increased the proliferation of the same cells treated with TGFß alone. When the conditioned media samples employed in Fig. 6B were used to stimulate D1K2-T1 cells for 15 min, the medium conditioned by the two fibroblast cell lines stimulated c-Met tyrosine phosphorylation to a greater extent than the 0.1% FBS control, or that observed upon stimulation with 10% FBS or D1K2-T1–conditioned medium (Fig. 6D). The tumor-derived myofibroblast–conditioned medium also increased the phosphorylation of Akt and p70s6k on activating sites over that observed with the unconditioned medium (0.1% FBS). The stimulation of c-Met tyrosine phosphorylation suggests that HGF is one of the mitogenic factors in tumor-derived fibroblast-conditioned medium. Other unidentified factors are also likely present in the medium because TGFß inhibited rather than potentiated HGF-induced proliferation of D1K2-T1 cells (data not shown) and because medium conditioned by tumor-derived fibroblast line 2 (TDF2) stimulated Akt phosphorylation to a much greater extent than 50 ng/mL HGF.


Figure 6
View larger version (38K):
[in this window]
[in a new window]

  		Figure 6. Characterization of tumor cell and tumor-associated fibroblast conditioned medium. A, mink lung epithelial cells containing a TGFß-responsive PAI-1 luciferase reporter construct were treated for 24 h with control medium (0.1% FBS-DMEM), or the same medium conditioned by neuT cells (neuT CM), D1K2-T1 cells (D1K2-T1 CM), D1K2-T3 cells (D1K2-T3 CM), D1K2 tumor-derived fibroblast line 1 (D1K2-TDF1 CM), or D1K2 tumor-derived fibroblast line 2 (D1K2-TDF2 CM). + TßRI Inh., the addition of 1 µmol/L TGFß receptor I kinase inhibitor to the conditioned medium. Columns, relative luminescence units normalized to micrograms of protein in the cell lysate analyzed; bars, SD of triplicate determinations. B, Mv1Lu or D1K2-T1 cells were treated for 24 h with the indicated conditioned medium and the rate of 3H-thymidine incorporation into DNA was measured. Columns, percent control 3H-thymidine incorporation of triplicate determinations; bars, SD. C, neuT cells were treated for 24 h with serum-free medium (SFM), 10% FBS, 10% FBS-containing medium conditioned by D1K2 tumor-derived fibroblast line 1 (D1K2-TDF1 CM), or medium conditioned by the neuT cells (neuT CM) and the rate of 3H-thymidine incorporation into DNA was measured as in (B). D, D1K2-T1 cells were stimulated for 15 min with 0.1% FBS, 0.1% FBS-containing medium conditioned by D1K2-T1 cells (D1K2-T1 CM), 0.1% FBS-containing medium conditioned by D1K2 tumor-derived fibroblast line 1 (TDF1 CM), 0.1% FBS-containing medium conditioned by D1K2 tumor-derived fibroblast line 2 (TDF2 CM), 10% FBS, or 50 ng/mL HGF. Cell lysates were prepared and analyzed by immunoblot with the indicated antibodies.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Cyclin D1/Cdk2 complexes are present in human breast cancer cell lines (12), and the levels of these complexes correlate well with the degree of cyclin D1 overexpression. It is unknown whether these complexes participate in the transforming effects of cyclin D1 overexpression. We constructed a gene encoding a cyclin D1-Cdk2 fusion protein to explore in a selective fashion the potential functions of cyclin D1/Cdk2 complexes in cell transformation and mammary tumorigenesis (14). The results presented here show that the expression of this cyclin D1-Cdk2 fusion protein (D1K2) in the mammary gland under the control of the MMTV promoter causes precocious lobuloalveolar differentiation of mammary glands in virgin mice. The extent of the phenotype increases with age and results in the formation of hyperplastic lesions and eventually in breast tumors. MMTV-D1K2 tumors induce a strong desmoplastic reaction as compared with MMTV-neu tumors. Biochemical analyses showed cell cycle deregulation in the tumors, including Rb and p130 hyperphosphorylation and up-regulation of E2F-dependent gene products. We hypothesize that c-Met expression in MMTV-D1K2 tumor cells, combined with their blunted growth inhibitory response to TGFß, render them particularly susceptible to the formation of a tumor cell/myofibroblast paracrine loop involving TGFß and HGF. In this model (Fig. 4C), tumor cells secrete TGFß that either directly or indirectly increases the abundance of stromal myofibroblasts. Myofibroblasts, in turn, drive the proliferation of cancer cells by secreting HGF and other growth factors.

Together, the results presented here suggest that the cell cycle deregulation caused by expression of the cyclin D1-Cdk2 fusion protein reprograms tumor cell responses to autocrine and paracrine growth factors to permit the formation of a self-amplifying cancer cell/myofibroblast paracrine cycle. The MMTV-D1K2 mice and cell lines derived from their tumors will be valuable tools for further delineating the mechanisms involved. A more thorough understanding of the mechanisms at work in this cycle is critical because of the high frequency with which similar desmoplastic reactions are observed in a wide variety of human tumors, the connection between tumor-induced stromal desmoplasia and tumor invasiveness and metastatic potential, and because several key elements of the cycle including c-Met (41, 42), TGFß receptors (43–45), and Cdk2 (9–11) are targets for agents under development as anticancer therapeutics.


   Acknowledgments
 
Grant support: NIH Grant R01-CA93651 (B. Law). We thank Dr. Harold L. Moses' laboratory (Vanderbilt-Ingram Cancer Center, Nashville, TN) for supplying MMTV-c-neu transgenic mice and the neuT tumor cell line.

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.

Received 10/16/06. Revised 12/29/06. Accepted 1/25/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Akli S, Keyomarsi K. Cyclin E and its low molecular weight forms in human cancer and as targets for cancer therapy. Cancer Biol Ther 2003;2:S38–47.[Medline]
  2. Xia W, Chen JS, Zhou X, et al. Phosphorylation/cytoplasmic localization of p21Cip1/WAF1 is associated with HER2/neu overexpression and provides a novel combination predictor for poor prognosis in breast cancer patients. Clin Cancer Res 2004;10:3815–24.[Abstract/Free Full Text]
  3. Winters ZE, Hunt NC, Bradburn MJ, et al. Subcellular localisation of cyclin B, Cdc2 and p21(WAF1/CIP1) in breast cancer. Association with prognosis. Eur J Cancer 2001;37:2405–12.[CrossRef][Medline]
  4. Catzavelos C, Bhattacharya N, Ung YC, et al. Decreased levels of the cell-cycle inhibitor p27Kip1 protein: prognostic implications in primary breast cancer. Nat Med 1997;3:227–30.[CrossRef][Medline]
  5. Han S, Park K, Kim HY, Lee MS, Kim HJ, Kim YD. Reduced expression of p27Kip1 protein is associated with poor clinical outcome of breast cancer patients treated with systemic chemotherapy and is linked to cell proliferation and differentiation. Breast Cancer Res Treat 1999;55:161–7.[CrossRef][Medline]
  6. Blain SW, Massague J. Breast cancer banishes p27 from nucleus. Nat Med 2002;8:1076–8.[CrossRef][Medline]
  7. Viglietto G, Motti ML, Bruni P, et al. Cytoplasmic relocalization and inhibition of the cyclin-dependent kinase inhibitor p27(Kip1) by PKB/Akt-mediated phosphorylation in breast cancer. Nat Med 2002;8:1136–44.[CrossRef][Medline]
  8. Coletta RD, Christensen K, Reichenberger KJ, et al. The Six1 homeoprotein stimulates tumorigenesis by reactivation of cyclin A1. Proc Natl Acad Sci U S A 2004;101:6478–83.[Abstract/Free Full Text]
  9. Shapiro GI. Preclinical and clinical development of the cyclin-dependent kinase inhibitor flavopiridol. Clin Cancer Res 2004;10:4270–5s.[CrossRef]
  10. Deans AJ, Khanna KK, McNees CJ, Mercurio C, Heierhorst J, McArthur GA. Cyclin-dependent kinase 2 functions in normal DNA repair and is a therapeutic target in BRCA1-deficient cancers. Cancer Res 2006;66:8219–26.[Abstract/Free Full Text]
  11. Wesierska-Gadek J, Schmid G. Dual action of the inhibitors of cyclin-dependent kinases: targeting of the cell-cycle progression and activation of wild-type p53 protein. Expert Opin Investig Drugs 2006;15:23–38.[CrossRef][Medline]
  12. Sweeney KJ, Swarbrick A, Sutherland RL, Musgrove EA. Lack of relationship between CDK activity and G1 cyclin expression in breast cancer cells. Oncogene 1998;16:2865–78.[CrossRef][Medline]
  13. Sutherland RL, Musgrove EA. Cyclin D1 and mammary carcinoma: new insights from transgenic mouse models. Breast Cancer Res 2002;4:14–7.[Medline]
  14. Chytil A, Waltner-Law M, West R, et al. Construction of a cyclin D1-2 fusion protein to model the biological functions of cyclin D1-2 complexes. J Biol Chem 2004;279:47688–98.[Abstract/Free Full Text]
  15. Matsui Y, Halter SA, Holt JT, Hogan BL, Coffey RJ. Development of mammary hyperplasia and neoplasia in MMTV-TGF {alpha} transgenic mice. Cell 1990;61:1147–55.[CrossRef][Medline]
  16. Ip MM, Asch BB, editors. Methods in mammary gland biology and breast cancer research. New York (NY), 10013: Kluwer Academic/Plenum Publishers; 2000. p. 317.
  17. Law BK, Chytil A, Dumont N, et al. Rapamycin potentiates transforming growth factor ß-induced growth arrest in nontransformed, oncogene-transformed, and human cancer cells. Mol Cell Biol 2002;22:8184–98.[Abstract/Free Full Text]
  18. Hall C, Nelson DM, Ye X, et al. HIRA, the human homologue of yeast Hir1p and Hir2p, is a novel cyclin-cdk2 substrate whose expression blocks S-phase progression. Mol Cell Biol 2001;21:1854–65.[Abstract/Free Full Text]
  19. Brown KA, Roberts RL, Arteaga CL, Law BK. Transforming growth factor-ß induces Cdk2 relocalization to the cytoplasm coincident with dephosphorylation of retinoblastoma tumor suppressor protein. Breast Cancer Res 2004;6:R130–9.[CrossRef][Medline]
  20. Law M, Forrester E, Chytil A, et al. Rapamycin disrupts cyclin/cyclin-dependent kinase/p21/proliferating cell nuclear antigen complexes and cyclin D1 reverses rapamycin action by stabilizing these complexes. Cancer Res 2006;66:1070–80.[Abstract/Free Full Text]
  21. Abe M, Harpel JG, Metz CN, Nunes I, Loskutoff DJ, Rifkin DB. An assay for transforming growth factor-ß using cells transfected with a plasminogen activator inhibitor-1 promoter-luciferase construct. Anal Biochem 1994;216:276–84.[CrossRef][Medline]
  22. Bacus SS, Gudkov AV, Esteva FJ, Yarden Y. Expression of erbB receptors and their ligands in breast cancer: implications to biological behavior and therapeutic response. Breast Dis 1999;11:63–75.
  23. Hulit J, Lee RJ, Russell RG, Pestell RG. ErbB-2–induced mammary tumor growth: the role of cyclin D1 and p27Kip1. Biochem Pharmacol 2002;64:827–36.[CrossRef][Medline]
  24. Lee RJ, Albanese C, Fu M, et al. Cyclin D1 is required for transformation by activated Neu and is induced through an E2F-dependent signaling pathway. Mol Cell Biol 2000;20:672–83.[Abstract/Free Full Text]
  25. Bowe DB, Kenney NJ, Adereth Y, Maroulakou IG. Suppression of Neu-induced mammary tumor growth in cyclin D1 deficient mice is compensated for by cyclin E. Oncogene 2002;21:291–8.[CrossRef][Medline]
  26. Nelsen CJ, Kuriyama R, Hirsch B, et al. Short-term cyclin D1 overexpression induces centrosome amplification, mitotic spindle abnormalities, and aneuploidy. J Biol Chem 2005;280:768–76.[Abstract/Free Full Text]
  27. Duensing A, Liu Y, Tseng M, Malumbres M, Barbacid M, Duensing S. Cyclin-dependent kinase 2 is dispensable for normal centrosome duplication but required for oncogene-induced centrosome overduplication. Oncogene 2006;25:2943–9.[CrossRef][Medline]
  28. Desmouliere A, Guyot C, Gabbiani G. The stroma reaction myofibroblast: a key player in the control of tumor cell behavior. Int J Dev Biol 2004;48:509–17.[CrossRef][Medline]
  29. Walker RA. The complexities of breast cancer desmoplasia. Breast Cancer Res 2001;3:143–5.[CrossRef][Medline]
  30. Ohtani H, Kuroiwa A, Obinata M, Ooshima A, Nagura H. Identification of type I collagen-producing cells in human gastrointestinal carcinomas by non-radioactive in situ hybridization and immunoelectron microscopy. J Histochem Cytochem 1992;40:1139–46.[Abstract]
  31. Ronnov-Jessen L, Van Deurs B, Nielsen M, Petersen OW. Identification, paracrine generation, and possible function of human breast carcinoma myofibroblasts in culture. In vitro Cell Dev Biol 1992;28A:273–83.
  32. Ronnov-Jessen L, Petersen OW. Induction of {alpha}-smooth muscle actin by transforming growth factor-ß 1 in quiescent human breast gland fibroblasts. Implications for myofibroblast generation in breast neoplasia. Lab Invest 1993;68:696–707.[Medline]
  33. Ronnov-Jessen L, Petersen OW, Koteliansky VE, Bissell MJ. The origin of the myofibroblasts in breast cancer. Recapitulation of tumor environment in culture unravels diversity and implicates converted fibroblasts and recruited smooth muscle cells. J Clin Invest 1995;95:859–73.[Medline]
  34. Beauchamp RD, Coffey RJ, Jr., Lyons RM, Perkett EA, Townsend CM, Jr., Moses HL. Human carcinoid cell production of paracrine growth factors that can stimulate fibroblast and endothelial cell growth. Cancer Res 1991;51:5253–60.[Abstract/Free Full Text]
  35. Neaud V, Faouzi S, Guirouilh J, et al. Human hepatic myofibroblasts increase invasiveness of hepatocellular carcinoma cells: evidence for a role of hepatocyte growth factor. Hepatology 1997;26:1458–66.[CrossRef]
  36. Guirouilh J, Castroviejo M, Balabaud C, Desmouliere A, Rosenbaum J. Hepatocarcinoma cells stimulate hepatocyte growth factor secretion in human liver myofibroblasts. Int J Oncol 2000;17:777–81.[Medline]
  37. De Wever O, Nguyen QD, Van Hoorde L, et al. Tenascin-C and SF/HGF produced by myofibroblasts in vitro provide convergent pro-invasive signals to human colon cancer cells through RhoA and Rac. FASEB J 2004;18:1016–8.[Abstract/Free Full Text]
  38. Lewis MP, Lygoe KA, Nystrom ML, et al. Tumour-derived TGF-ß1 modulates myofibroblast differentiation and promotes HGF/SF-dependent invasion of squamous carcinoma cells. Br J Cancer 2004;90:822–32.[CrossRef][Medline]
  39. Sato Y, Tsuboi R, Lyons R, Moses H, Rifkin DB. Characterization of the activation of latent TGF-ß by cocultures of endothelial cells and pericytes or smooth muscle cells: a self-regulating system. J Cell Biol 1990;111:757–63.[Abstract/Free Full Text]
  40. Steinman RA, Wentzel A, Lu Y, Stehle C, Grandis JR. Activation of Stat3 by cell confluence reveals negative regulation of Stat3 by cdk2. Oncogene 2003;22:3608–15.[CrossRef][Medline]
  41. Christensen JG, Burrows J, Salgia R. c-Met as a target for human cancer and characterization of inhibitors for therapeutic intervention. Cancer Lett 2005;225:1–26.[CrossRef][Medline]
  42. Christensen JG, Schreck R, Burrows J, et al. A selective small molecule inhibitor of c-Met kinase inhibits c-Met–dependent phenotypes in vitro and exhibits cytoreductive antitumor activity in vivo. Cancer Res 2003;63:7345–55.[Abstract/Free Full Text]
  43. Matsuyama S, Iwadate M, Kondo M, et al. SB-431542 and Gleevec inhibit transforming growth factor-ß–induced proliferation of human osteosarcoma cells. Cancer Res 2003;63:7791–8.[Abstract/Free Full Text]
  44. Halder SK, Beauchamp RD, Datta PK. A specific inhibitor of TGF-ß receptor kinase, SB-431542, as a potent antitumor agent for human cancers. Neoplasia 2005;7:509–21.[CrossRef][Medline]
  45. Ge R, Rajeev V, Ray P, et al. Inhibition of growth and metastasis of mouse mammary carcinoma by selective inhibitor of transforming growth factor-ß type I receptor kinase in vivo. Clin Cancer Res 2006;12:4315–30.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Pathol.Home page
E. Macias, P. L. Miliani de Marval, A. De Siervi, C. J. Conti, A. M. Senderowicz, and M. L. Rodriguez-Puebla
CDK2 Activation in Mouse Epidermis Induces Keratinocyte Proliferation but Does Not Affect Skin Tumor Development
Am. J. Pathol., August 1, 2008; 173(2): 526 - 535.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
X. Wang, B. Liu, N. Li, H. Li, J. Qiu, Y. Zhang, and X. Cao
IPP5, a Novel Protein Inhibitor of Protein Phosphatase 1, Promotes G1/S Progression in a Thr-40-dependent Manner
J. Biol. Chem., May 2, 2008; 283(18): 12076 - 12084.
[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 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 Corsino, P.
Right arrow Articles by Law, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Corsino, P.
Right arrow Articles by Law, B.

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