Cancer Research Prevention Award Metabolism
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 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 Hulit, J.
Right arrow Articles by Pestell, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hulit, J.
Right arrow Articles by Pestell, R. G.
[Cancer Research 66, 8529-8541, September 1, 2006]
© 2006 American Association for Cancer Research

Cell, Tumor, and Stem Cell Biology

p27Kip1 Repression of ErbB2-Induced Mammary Tumor Growth in Transgenic Mice Involves Skp2 and Wnt/ß-Catenin Signaling

James Hulit1, Richard J. Lee1, Zhiping Li2, Chenguang Wang2, Sanjay Katiyar2, Jianguo Yang2, Andrew A. Quong2, Kongming Wu2, Chris Albanese3, Robert Russell3, Dolores Di Vizio1, Andrew Koff3, Saiinder Thummala1, Hui Zhang4,6, Jennifer Harrell6, Hong Sun6, William J. Muller7, Giorgio Inghirami5, Michael P. Lisanti2 and Richard G. Pestell2

1 Department of Developmental and Molecular Biology, Department of Medicine, Albert Einstein College of Medicine, Bronx, New York; 2 Kimmel Cancer Center, Departments of Cancer Biology and Medical Oncology, Thomas Jefferson University, Philadelphia, Pennsylvania; 3 Department of Oncology, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, District of Columbia; 4 Molecular Biology Program, Memorial Sloan-Kettering Cancer Center, Sloan-Kettering Institute; 5 Department of Pathology and Kaplan Comprehensive Cancer Center, New York University School of Medicine, New York, New York; 6 Department of Genetics, Yale University School of Medicine, New Haven, Connecticut; and 7 Molecular Oncology Labs, Royal Victoria Hospital, McGill University, Montreal, Quebec, Canada

Requests for reprints: Richard G. Pestell, Departments of Cancer Biology and Medical Oncology, Thomas Jefferson University, 233 South 10th Street, Philadelphia, PA 19107. Phone: 215-503-5649; Fax: 215-923-9334; E-mail: Richard.Pestell{at}jefferson.edu or D_Scardino{at}mail.jci.tju.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Expression of the cyclin-dependent kinase (Cdk) inhibitor (p27Kip1) is frequently reduced in human tumors, often correlating with poor prognosis. p27Kip1 functions as a haploinsufficient tumor suppressor; however, the mechanism by which one allele of p27Kip1 regulates oncogenic signaling in vivo is not well understood. We therefore investigated the mechanisms by which p27Kip1 inhibits mammary tumor onset. Using the common background strain of FVB, p27Kip1 heterozygosity (p27+/–) accelerated ErbB2-induced mammary tumorigenesis. We conducted microarray analyses of mammary tumors developing in mice with genetic haploinsufficiency for p27Kip1 expressing a mammary-targeted ErbB2 oncogene. Global gene expression profiling and Western blot analysis of ErbB2/p27+/– tumors showed that the loss of p27Kip1 induced genes promoting lymphangiogenesis, cellular proliferation, and collaborative oncogenic signaling (Wnt/ß-catenin/Tcf, Cdc25a, Smad7, and Skp2). Skp2 expression was induced by ErbB2 and repressed by p27Kip1. Degradation of p27Kip1 involves an SCF-type E3 ubiquitin ligase, including Skp2. The Skp2 component of the SCFSKP2 complex that degrades p27Kip1 was increased in ErbB2 tumors correlating with earlier tumor onset. In both murine and human ErbB2-overexpressing breast cancers, p27Kip1 levels correlated inversely with Skp2. p27Kip1 haploinsufficiency activated Wnt/ß-catenin/hedgehog signaling. Reintroduction of p27Kip1 inhibited ß-catenin induction of Tcf-responsive genes (Siamosis, c-Myc, and Smad7). p27Kip1 is haploinsufficient for ErbB2 mammary tumor suppression in vivo and functions to repress collaborative oncogenic signals including Skp2 and Wnt/ß-catenin signaling. (Cancer Res 2006; 66(17): 8529-41)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cellular proliferation is positively regulated by cyclin/cyclin-dependent kinase (Cdk) complexes, which are opposed by the Cdk inhibitors. The Cdk inhibitors are classified into two families: p21 (p21Cip1, p27Kip1, and p57Kip2) and p16INK4a (p15, p16, p18, and p19). p27Kip1 inhibits most cyclin/Cdk complexes, in particular cyclin E/Cdk2. Expression of p27Kip1 in cell lines causes cell cycle arrest. Mitogen withdrawal and cadherin-mediated cell contact inhibition induce p27Kip1 binding to cyclin E/Cdk2 and cyclin A/Cdk2, correlating with the inhibition of cell cycle progression and DNA synthesis. p27Kip1 regulates cellular polarity and inhibits cell migration induced by cell-extracellular matrix contact. p27Kip1 abundance is regulated at the level of translation and protein turnover. Phosphorylation of p27Kip1 by Cdk2 creates a binding site for a specific SCF-type E3 ubiquitin-protein ligase, which promotes proteosome-mediated degradation of p27Kip1 (1–3). The F-box protein Skp2 is the substrate recognition factor of the SCF complex, which recognizes and binds to phosphorylated p27Kip1. Skp2 is required for G1-S phase transition in transformed cells and diploid fibroblasts (4) and Skp2-deficient cells exhibit increased p27Kip1 levels and polyploidy (5).

The mechanisms by which p27Kip1 regulates tumorigenesis are less well understood. In the majority of studies, reduced p27Kip1 levels in tumors, including colon and breast cancers, correlates with poor prognosis, although elevated levels are reported in a subset of tumors (6–9). Reduction in p27Kip1 expression in tumors correlates with tumor agressivenesss and dedifferentiation. Point mutations in the coding region of the p27Kip1 gene are rare in human tumors, although loss of heterozygosity has been observed (10–14). Despite the widespread implication that p27Kip1 may serve as a tumor suppressor in different human cancers (15, 16), inactivation of Cdkn1b (encoding p27Kip1) in mice does not result in enhanced spontaneous tumor onset, with the exception of pituitary adenomas (17–19). p27Kip1 deficiency accelerated gastrointestinal neoplasia induced by 1,2-dimethylhydrazine or by mutation of the Apc gene in the Min mice (20). Reduced p27Kip1 may be anticipated to cooperate as a component of multistep tumorigenesis with loss of other tumor suppressors or activating oncogenes. The inactivation of p27Kip1, for example, may be epistatic to the inactivation of pten, as the inactivation of one pten allele, together with the inactivation of one Cdkn1b allele, enhances the rate of prostate tumor onset (21).

The neu (c-ErbB2, HER-2) proto-oncogene encodes a receptor tyrosine kinase that is a member of a growth factor receptor family, which is overexpressed in 20% to 30% of human breast tumors (22). When under the control of the murine mammary tumor virus (MMTV) long terminal repeat, overexpression of either wild-type or activated ErbB2 induces mammary adenocarcinomas in transgenic mice with high frequency (23, 24). Several cell cycle components regulated by ErbB2 have been implicated in mammary tumorigenesis. Cyclin D1 is overexpressed in MMTV-ErbB2 mammary tumors (25). In contrast to cyclin E or cyclin A, cyclin D1 is selectively transcriptionally induced by ErbB2 (25) and cyclin D1 antisense is sufficient to abrogate MMTV-ErbB2 induced cellular growth in nude mice (25). Furthermore, cyclin D1–/– mice are resistant to tumor induction by ErbB2 (26). In cultured cells, heregulin engagement and activation of ErbB2 correlate with a reduction in p27Kip1 protein abundance (27). Conversely, ErbB2 inhibitor antibodies increase p27Kip1 abundance, thus implicating p27Kip1 in ErbB2 mammary epithelial cell proliferation.

In these studies, we have examined the in vivo functions of p27Kip1 in mammary gland tumorigenesis by mating mice homozygously deleted of Cdkn1b (p27Kip1) with transgenic mice expressing mammary-targeted oncogenic ErbB2. We show that p27Kip1 is haploinsufficient for ErbB2 mammary tumor suppression in vivo. The loss of a single p27Kip1 allele resulted in a disproportionate decrease (~80%) in p27Kip1 abundance and increased Skp2. Global gene expression profiling and Western blot analysis of ErbB2/p27+/– tumors showed that the loss of p27Kip1 induced genes promoting Wnt/ß-catenin/hedgehog signaling (Dhh, Wnt3, and Lef1), lymphangiogenesis (Flt3l and EphA2), cellular proliferation, and collaborative oncogenic signaling (Cdc25A, Smad7, and Skp2). Reintroduction of p27Kip1 into p27Kip1-deficient cells inhibited expression of Skp2, Smad7, and Wnt/ß-catenin responsive genes. Thus, p27Kip1 inhibits Skp2, Smad7, and Wnt/ß-catenin responsive genes through a mechanism that involves repression of target gene promoter activity.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals
MMTV-neu mice (24) and mice homozygously deleted of the gene encoding p27Kip1 were previously described (17). The MMTV-neu mice were in the FVB strain and the p27Kip1–/– mice on the C57BL/6 background were backcrossed for three generations into FVB and then backcrossed into the MMTV-neu mice of FVB strain. Animal care was conducted in accordance with the standards set forth by the Institute for Animal Studies of the Albert Einstein College of Medicine. For genomic analyses of the transgenic mice, weaned animals were identified through a system of toe tagging. One centimeter of tail section was removed to make genomic DNA. Tail sections were incubated overnight at 50°C to 55°C in 700 µL of a proteolytic solution [50 mmol/L Tris (pH 8.0), 100 mmol/L EDTA, and 0.5% SDS] plus 20 µL of 20 mg/mL proteinase K solution. DNA was purified by phenol/chloroform extraction and ethanol precipitation, followed by resuspension in 250 µL of 0.1x TE (10 mmol/L Tris, 1 mmol/L EDTA). Virgin female neu-p27Kip1 mice were finger palpated over the length of their ventral side biweekly. On detection of a mass, individual mice of each specific p27Kip1 genotype were scored against the entire population of that genotype which remained at that time point and charted.

PCR
PCR primers for p27Kip1 from the p27Kip1-null mice were SW39, 5'-ATATTGCTGAAGAGCTTGGCGG; SW40, 5'-TCAAACGTGAGAGTGTCTAACGG; and SW41, 5'-AGGGGCTTATGATTCTGAAAGTCG. For detection of the wild-type and heterozygote p27Kip1 alleles, the primers SW39 (a forward primer that binds nucleotides 1,420-1,441 of PMC1POLA), SW40 (a forward primer binding nucleotides 4-26 of p{lambda}-KIP1-34-1), and SW41 (a reverse primer binding nucleotides 209-186 of p{lambda}-KIP1-34-1) were used. When used in conjunction, SW40 and SW41 amplify a region of 206 bp from the wild-type locus, whereas SW40 and SW39 produce a 298-bp product from the mutant locus. The reaction for each genomic DNA sample included 1 µg of genomic DNA, 100 pmol SW40, 50 pmol SW39, 50 pmol SW41, 1.5 units of Taq polymerase (Fisher Scientific, Pittsburgh, PA), 200 µmol/L of each deoxynucleotide triphosphate (dNTP), and PCR buffer [40 mmol/L NaCl, 10 mmol/L Tris (pH 8.9), 1.5 mmol/L MgCl2, and 0.01% gelatin]. Genomic DNA (denatured at 99°C, 4 minutes; 4°C, 1 minute) was subjected to PCR (35 cycles at 94°C, 60°C, and 72°C, 1 minute each).

PCR primer sequences used for neu from the MMTV-neu mice were 5'-CGGAACCCACATCAGGCC and 5'-TTTCCTGCAGCCTACGC. For detection of the neu transgene, the same buffer conditions were used as above, with the forward and reverse primers specific for neu. The PCR program was 30 cycles of 94°C, 1 minute; 92°C, 30 seconds; 58°C, 30 seconds; and 72°C, 1 minute.

Southern Blots
For detection of the neu transgene, 10 µg of genomic DNA were digested with BamH1 overnight at 37°C in 200 µL of total volume until complete digestion was confirmed with 5% of the reaction mixture. DNA was ethanol precipitated, resuspended in 15 to 20 µL of volume, and electrophoresed in a 1% agarose gel using proper controls and markers. After image verification, the gel was denatured for at least 45 minutes in 1.5 mol/L NaCl and 0.4 N NaOH, rinsed with water, and neutralized for at least 45 minutes in 1.5 mol/L NaCl and 0.5 mol/L Tris (pH 7.6). DNA was transferred overnight onto Hybond-N (charged nylon) membrane [equilibrated in 20x SSC (175.3 g of NaCl and 88.2 g of sodium citrate in 1 liter of water)]. DNA was then UV cross-linked to the nylon membrane. The neu-specific probe was prepared by gel purification of an 800-bp fragment from a BamH1-digested pJ4{Omega}NeuN plasmid. Twenty-five nanograms of probe were labeled with [{alpha}-32P]dCTP using the Amersham Redi-Prime kit protocol and the probe solution was prepared with RapidHyb solution according to the protocol of the manufacturer (Amersham Pharmacia Biotech, Piscataway, NJ). The membrane was hybridized with probe at 65°C for 3 hours with shaking, washed twice for 5 minutes each using 2x SSC and 0.1% SDS at room temperature, twice for 5 minutes with 0.2x SSC and 0.1% SDS at room temperature, and twice for 5 minutes with 0.2x SSC and 0.1% SDS at 65°C. Blots were then exposed to film or phosphorimaging screen (Molecular Dynamics, Sunnyvale, CA).

Whole-Mount Mammary Gland Preparation
Fourth (abdominal) mammary glands were surgically removed, stretched, and mounted onto glass slides, followed by fixation in 75% ethanol/25% acetic acid overnight. Tissues were then washed in 70% ethanol, followed by 5 minutes in distilled water, and stained overnight in an aluminum-carmine dye solution (1 g carmine dye, 2.5 g potassium alum; Sigma Chemical, St. Louis, MO) diluted in 500 mL of water. The solution was brought to a boil, followed by addition of a few thymol crystals (Sigma) while stirring. The solution was filtered and stored at 4°C. Following staining, mammary glands were dehydrated through a graded series of ethanol solutions (15 minutes each at 70%, 90%, and 100%), defatted in xylenes (Electron Microscopy Sciences, Fort Washington, PA), and stored in methyl salicylate (Sigma). Ductal branch points in the mammary gland whole-mount preparations were measured from the nipple area to the tip of the three longest ducts passing through the lymph node. The numbers of branches represent the mean branching number along the three longest ducts. Epithelial cell area was determined from the cross-sectional area of ductal end points. Area measurements were assessed in 10 separate animals (age, 165 days) using 181 fields and analyzed with NIH image software, with the data being expressed as total pixel units. Statistical evaluations were done with either two-tailed Student's t test or Mann-Whitney U test in cases where the data could not be assumed to be Gaussian.

Nucleus Size Analysis
H&E-stained mammary tumor sections were analyzed by microscopy using Nikon model Diaphop 300 system (Nikon Instruments, Inc., Melville, NY) at 40x objective. Images were captured with the SPOT charge-coupled device digital camera (model 1.5.0, Diagnostic Instruments, Inc., Sterling Heights, MI) fitted with a Nikon 0.45x HRD045-NIK lens. Digital imaging, capture, and editing were done with Adobe Photoshop (Adobe Systems, Inc., San Jose CA). Nucleus counting and measuring were done with Image-Pro Plus version 3.0 (Media Cybernetics, Silver Spring, MD). Nucleus number and area data were graphically analyzed and displayed using Microsoft Excel (Microsoft Corporation, Redmond, WA). Briefly, three fields were selected from each section; the images were captured and saved as gray scale TIFF files. Images were then transformed into a two-color display (black nuclei and white background). Files were then imported into the Image-Pro software program. Nuclei were then analyzed for both number and area (area displayed as total number of pixels).

Western Blots
The abundance of mitogen-activated protein kinase and cell cycle–, survival-, and apoptosis-related proteins was determined by Western blot analysis as previously described (25). Antibodies included anti-cyclin D1 (Ab3; NeoMarkers, Fremont, CA), anti-p27Kip1 (F-8; Santa Cruz Biotechnology, Santa Cruz, CA), anti-p21Cip1 (187; Santa Cruz Biotechnology), anti-cyclin E (HE111; Santa Cruz Biotechnology), anti-Cul-1 (C20; Santa Cruz Biotechnology), anti-Akt and anti-phospho-Akt (9272 and serine 473; Cell Signaling Technology, Danvers, MA), and a guanine nucleotide dissociation inhibitor antibody (a generous gift from Perry Bickel, Washington University, St. Louis, MO).

Immunohistochemistry
For analysis of murine mammary tumors, paraffin-embedded sections were stained and visualized with a 3,3'-diaminobenzidine (DAB) method (Dako Kit LSAB+, peroxidase) as detailed in a protocol provided by the manufacturer (Dako Corporation, Carpinteria, CA). Antibodies used were directed to cyclin D1 (25) or p27Kip1 (1:500 dilution; Transduction Laboratories, Inc., BD Biosciences, Franklin Lakes, NJ). Briefly, after deparaffination, antigens were retrieved by microwave irradiation in 0.01 mol/L trisodium citrate buffer (pH 6.0). Slides were washed and incubated with primary antibodies for 1 hour at room temperature, followed by incubation with secondary antibodies. Positive signals were revealed by DAB chromogen according to the conditions of the supplier. Slides were then counterstained with Harris hematoxylin (Fisher Scientific). For analysis of human breast cancer tumors, immunohistochemical staining for Skp2 was done with a mix of four different affinity-purified monoclonal antibodies (mAb) as previously described (28). At least 20 high-power fields were chosen at random and 2,000 cells were counted. Immunostaining was done on formalin-fixed, paraffin-embedded tissues with the avidin-biotin-peroxidase complex method and a semiautomated immunostainer (DAKO or Ventana System) as described (29). Antigen retrieval was done by microwaving the slides for 20 minutes in 10 mmol/L citrate buffer (pH 8.0 for Skp2; Zymed, South San Francisco, CA) or for 16 minutes in Trilogy buffer (Cell Marque, Los Angeles, CA; cb11, Novacastra, New Castle, United Kingdom). Immunostaining was evaluated by a pathologist blinded to the conditions. At least 10 high-power fields were chosen randomly and >100 cells per field were counted. The tumors were scored as percentage of positive cells for each antigen. The determination of HER-2 protein overexpression was based on the membrane staining only. Cytoplasmic staining was considered nonspecific. Immunohistochemical values for HER-2 overexpression were expressed as negative (0 and +), weakly positive (++), and strongly positive (+++).

Plasmids and reporter constructs. Luciferase reporter genes c-Myc-LUC, c-Myc {Delta}Tcf-LUC (30), Siamosis-LUC, Siamosis {Delta}Tcf-LUC (31), Engr-LUC, Engr-{Delta}Tcf (30), Smad7 (–4,600 to +672), Smad7{Delta} (–303 to +672; ref. 32), pCGT-Skp2 construct expression vector (33), CMV-Skp2 (34), ß-catenin S33 (35), and PSV2 Neu NT (25). A 1.87-kb Skp2 promoter fragment was cloned into pGL3-LUC to form –1870 Skp2-LUC.

Cell culture and reporter assays. The p27Kip1–/–3T3 and p27Kip+/+ 3T3 were maintained in DMEM with 10% FCS and 1% penicillin/streptomycin at 37°C under 10% CO2. Luciferase assay procedures were done as previously described (36). Cells were plated in 12-well dishes (Falcon) and seeded at 50% to 70% confluency the night before. Transfections were carried out with Polyfect Transfection reagent (Qiagen, Valencia, CA). Transfection protocol was based on the instructions of the manufacturer. Vectors were individually aliquoted into 1.5-mL tubes (Eppendorf) and diluted to 150 µL with serum- and antibiotic-free DMEM. To the DNA/DMEM mixture was added 10 µL of Polyfect reagent per reaction tube. After 10 minutes of incubation, 1.0 mL of full mediun (serum plus antibiotics) was added to the DNA/Polyfect mixture and immediately transferred onto cells that had been washed with PBS. Luciferase activity was determined 24 to 36 hours posttransfection. Renilla luciferase (TK-LUC) was cotransfected (in a 1:10 ratio with reporter vector) as an internal control for transfection efficiency. Luciferase assays were done at room temperature with Autolumat LB 953 (EG&G Berthold, Berthold Technologies, Oak Ridge, TN). Measurements were made over 10 seconds to assess luciferase content. Data values were expressed in arbitrary light units. Background activity from cell extracts was typically <100 arbitrary light units/10 s. Statistical analyses were done with the Mann-Whitney U test and significant differences were established as P < 0.05. Luciferase buffers and reagents were purchased from Promega (Madison, WI).

cDNA and Oligonucleotide Microarray Analyses
RNA was isolated from freshly dissected mammary tumors from MMTV-ErbB2 transgenic mice of either p27+/– or p27+/+ allele type. The mRNA was examined with the cDNA microarray glass slides or Affymetrix oligonucleotide arrays to expand the number of genes examined. Affymetrix Mu11K A/B GeneChips were used for all oligonucleotide microarray experiments. The targets for Affymetrix DNA microarray analysis were prepared as described by the manufacturer. Briefly, double-stranded cDNA was synthesized from total RNA (10 µg starting material) isolated from tissue culture harvests. Biotin-labeled cRNA was generated by in vitro transcription from the DNA. The cRNA was fragmented before hybridization. A hybridization cocktail was prepared that included the fragmented cRNA, probe array controls, bovine serum albumin, and herring sperm DNA. The cRNA was hybridized to the array oligonucleotide probes for a 16-hour incubation at 45°C. Immediately after the hybridization, the hybridized probe array underwent automated washing and staining on Affymetrix Fluidics Station. The DNA chips were scanned with the Affymetrix GeneChip scanner and the signals processed with the GeneChip expression analysis algorithm v.2 (Affymetrix, Santa Clara, CA). Metagene regression analysis of expression data was conducted as previously described (25, 37). Affymetrix U74 Av2 data sets were processed with Affymetrix MAS5.0 to compute signal values for each probe set and resulting data sets for each specimen. Details on the methods of normalization, calculation of geometric fold changes, t test analyses, and hierarchical clustering were previously described (38).

Unique human expressed sequence tags (16,580) from Research Genetics (Huntsville, AL) were used as cDNA probes, which were spotted on two separate glass slides: 5H arrays contained 7,873 non-sequence-verified probes and H1 arrays contained 8,707 sequence-verified probed.8 For each hybridization, cDNA targets were prepared from the RNA sample (Cy3 labeled) obtained from ErbB2/p27+/– cells and the reference RNA sample (Cy3-labeled ErbB2 p27+/+). RNA was subjected to in vitro transcription in the presence of biotinylated Cy3-UTP or Cy5-UTP (Amersham Pharmacia Biotech). The target was hybridized for 12 hours at 50°C to the arrays. Independent images were obtained from Cy3 and Cy5 fluorescence emitted from hybridized microarrays using a custom-built dual channel laser scanning microscope.9 ScanAlyze version 2.44 software (M. Eisen, Stanford University, Palo Alto, CA)10 was used as described in the ScanAlyze manual to generate raw data files containing measurements of signal and background fluorescence emissions of Cy3 and Cy5, respectively, for each element. Genes were identified after eliminating genes that failed to vary in expression level within an experiment by a factor of 2 and an absolute value of 100 and normalizing within experiments to a mean of 0 and an SD of 1.

Genes were selected based on the criteria that P < 0.05 from a two-tailed t test and the fold change was >1.8 over the three replicates. This set of genes is listed in Fig. 5 as a heat map, where data from each probe are represented by rows and each experiment is shown as a column. Red and green denote increased and decreased expression levels, respectively, with the intensity reflecting the magnitude of change.


Figure 5
View larger version (35K):
[in this window]
[in a new window]
  		Figure 5. Microarray analysis of MMTV-ErbB2-p27+/– versus MMTV-ErbB2-p27+/+ tumors. A, hierarchical cluster analysis for genes found to be differentially regulated. Red, up-regulation; green, down-regulation; black, no change. Each column represents a different tumor sample grouped together as those from p27Kip1 wild-type animals (left) and those from p27Kip1 heterozygote animals (right). Each row represents a different gene found to be differentially regulated. The tree branches define relationships between the various genes and between the tumors. The shorter the tree branch, the closer the similarity of the objects, as determined by hierarchical clustering (http://141.161.133.103/graph). B, multidimensional scaling analysis (MDS) for ErbB2/p27Kip1 wild-type and heterozygote tumors. Distance between points on the three-dimensional scatter plot represents dissimilarities between tumors as determined from expression levels of 183 differentially expressed genes in each tumor sample. Graph plot positions are determined by applying the multidimensional scaling algorithm and are calculated from the Spearman's correlation measure (r) as 1 – r. C, Western blot analysis of Tcf4. Columns, mean for p27+/– 10 wild-type and 10 ErbB2/p27+/– tumors; bars, SE. D, quantitative RT-PCR analysis of several genes altered in expression by microarray.

 
PCR, RNA Extraction, Real-time PCR, Comparative Calculation, and Determination of Relative Expression Levels
Validation of microarray gene expression results by real-time RT-PCR. cDNA made from RNA extracted from animal tumor samples using standard guanidine isothiocyanate-phenol-chloroform method was subjected to GeneChip Mouse Expression Set 430 gene expression arrays (Affymetrix). The same RNA sample was RQ1 DNase I (Promega) treated and phenol-chloroform extracted before proceeding to validation of gene expression by real-time RT-PCR analysis. Equal quantities of DNA-free RNA were used for reverse transcription reactions for making cDNA using Iscript Reverse transcriptase kit (Bio-Rad, Hercules, CA). SYBR green–based real-time PCR reactions were done with QuantiTect SYBR Green PCR Kit and p27 primers (Prime Express 5.1, Applied Biosystems Inc., Foster City, CA; ref. 39) prevalidated QuantiTect Primer assays (Qiagen) for Cdc25a, Pglyrp1, Yy1, Col2a1, Cd97, Serpini1, Mmp16, Dusp9, Map3k8, Nfatc2, Mras, Tiam1, Epha2, and 18S RNA (internal control) on an 7900HT real-time PCR system (Applied Biosystems) using Sequence Detection System 2.1 with manufacturer-prescribed thermal cycling variables and real-time data collections points. All assays were set up for relative quantitation method where mean Ct values from p27+/+ samples were used as calibrator for data analysis for p27+/– samples. The relative fold change of gene expression between p27+/+ versus p27+/– was calculated using the standard 2–{Delta}{Delta}Ct method.

Fluorescence-Activated Cell Sorting Analysis
Small pieces of formalin preserved tumor samples were cut (~25-50 mg). Samples were rinsed (1-hour incubations) twice in water for a total of 2 hours at room temperature. Tissue was incubated overnight at 37°C with gentle agitation in a solution of pepsin and sodium chloride [0.5% pepsin, 0.9% NaCl (pH 1.5)]. Following incubation, the digested material was centrifuged at 5,000 x g for 5 minutes. The pellet was washed in PBS and repelleted, leaving 0.5 mL of PBS covering the cells. Citric acid buffer (96 parts sodium phosphate 0.2 mol/L and 4 parts citric acid 0.1 mol/L) was added to PBS in a 1:1 ratio and samples were incubated for 5 minutes. Cells were repelleted and the buffer was decanted. Cells were resuspended in a propidium iodide/RNase/PBS solution (final concentrations of 10 µg/mL propidium iodide and 60 µg/mL RNase). DNA content was then measured by flow cytometry using FACScan (Becton Dickinson Immunocytometry Systems, San Jose CA). Data were collected with ScanQuest software (Becton Dickinson) and analyzed with ModFit LT (Verity Software House, Inc., Topsham, ME).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Rapid onset of Neu-induced mammary tumors in p27Kip1 heterozygote mice. To examine the molecular mechanisms by which p27Kip1 functions as a tumor suppressor with haploinsufficiency, transgenic mice were engineered. Experiments were conducted using transgenic mice lacking functional p27Kip1 (17). Heterozygote p27 mice were first bred into the FVB background to diminish any potential effects arising from the known, but poorly characterized, tumor-resistant phenotype of the C57BL/6 mice (40–42). Litters were then bred with MMTV-neu transgenic animals (FVB strain; ref. 24) to generate MMTV-neu-p27+/+, MMTV-neu-p27+/–, and MMTV-neu-p27–/– mice. Compared with the MMTV-neu-p27+/+ mice, MMTV-neu-p27+/– mice were intermediate in size (data not shown), similar to previously published reports on p27Kip1-deficient mice (17). The females from each genotype were monitored and palpated weekly for the onset of tumors. On detection, mammary tumors were periodically measured to determine growth rates before being carefully dissected free of surrounding tissues. MMTV-neu-p27+/– mice developed mammary gland tumors earlier and more rapidly than their littermates, with 50% having developed tumors (T50) by 45 weeks of age (Fig. 1 ). In contrast, MMTV-neu-p27+/+ mice exhibited a T50 of 69 weeks. By the Cox proportional hazards model, MMTV-neu-p27+/– animals developed tumor onset significantly faster than the wild-type mice (P = 0.014, compared with MMTV-neu-p27+/+ mice). Previous studies have suggested a correlation between reduced p27Kip1 levels and higher pathologic grade (8). Using the Annapolis classification system, histopathologic analysis was carried out to compare the ErbB2-p27+/+ and ErbB2-p27+/– mammary tumors (43). The majority of the tumors were classified as undifferentiated adenocarcinomas with features of both glandular and squamous differentiation. The nuclei were slightly smaller than normal mammary epithelial nuclei but uniform in size with delicate chromatin and abundant pink cytoplasm, characteristic of the ErbB2 histopathologic "signature" (43). As nuclear size is reported to correlate with specific genotypic and oncogenic changes (43), we formally measured nuclear size distribution within the ErbB2 mammary tumors from 30 separate animals. There was no significant alteration in nuclear size when tissues from the p27+/– and p27+/+ tumors were compared (n = 30 mice, 6 x 104 cells analyzed per sample; data not shown) and there was no difference in grade between the p27+/– and p27+/+ mammary tumors (data not shown). DNA content analysis revealed that mammary tumors had remained diploid and that no significant differences in aneuploidy existed in the tumors due to the loss of a p27Kip1 allele (n = 40; data not shown).


Figure 1
View larger version (8K):
[in this window]
[in a new window]
  		Figure 1. In vivo incidence of mammary tumor formation and survival in MMTV-ErbB2/p27+/– mice. A, mammary tumor onset rate in the MMTV-ErbB2/p27+/+ and MMTV-ErbB2/p27+/– genotypes.

 
The reduction in ErbB2-induced tumor onset of p27+/– versus p27+/+ may have resulted from altered development of the mammary gland. The mammary gland architecture was therefore carefully analyzed by whole-mount mammary preparation, with measurements taken of ductal branching and length. Epithelial development was determined by cross-sectional area and total mammary gland size. Analysis was done at 166 days, at which time mammary gland maturation is complete. Comparison between the MMTV-ErbB2 p27+/+ and MMTV-ErbB2 p27+/– animals revealed no significant difference in mammary gland length (3.66 ± 0.3 versus 3.78 ± 0.16 cm; n = 10) and no significant decrease in the number of mammary duct branch points per centimeter of duct in the p27+/– mammary gland (p27+/+ 33.9 ± 0.18 versus 30.70 ± 1.9 branches/cm duct; Fig. 2A ). The mammary epithelium area of 10 separate animals was assessed with NIH image software and revealed no significant differences (3.69 x 104 ± 0.37 versus 3.72 x 104 ± 0.22 arbitrary pixel units; Fig. 2B).


Figure 2
View larger version (87K):
[in this window]
[in a new window]
  		Figure 2. Mammary gland structure of ErbB2-p27+/– transgenic mice. Mammary gland whole mounts from MMTV-ErbB2/p27+/+ and MMTV-ErbB2/p27+/– mice (n = 10) were examined for ductal branching (mammary ducts/cm), mammary gland length (A) and mammary epithelial area (B), as described in Materials and Methods. C, genomic Southern blot analysis of mammary tumors derived from MMTV-ErbB2-p27+/+ or MMTV-ErbB2-p27+/– mice (all tumors are heterozygous for the p27Kip1 allele).

 
To determine whether the increase in tumor rate in p27+/– mice was secondary to either reduced expression of the ErbB2 transgene or loss of the remaining p27Kip1 allele, analysis was done on mammary tumors. Comparison was made between mammary tumor tissues derived from p27+/+ and p27+/– transgenic mice (Fig. 2C). Genotyping for the p27Kip1 allele and the MMTV-ErbB2 transgene was done with genomic Southern blotting and PCR analysis (data not shown). Analysis revealed that the p27Kip1 allele status of the tumors had remained unchanged from that of the animal from which they were derived (Fig. 2C).

Abundance of the SCF complex in MMTV-ErbB2 p27+/– mammary tumors. Cyclin D1 is required for ErbB2-induced tumor growth in nude mice (25). We therefore examined the abundance and activity of the Cdks and Cdk inhibitors in the p27+/+ and p27+/– ErbB2 mammary tumors. The abundance of these components was expressed as 100% in the p27Kip1 wild-type for comparative purposes. Previous studies have shown an increased abundance of cyclin D1 in ErbB2 mammary tumors (25). Although cyclin D1 levels were increased compared with normal mammary epithelium, we found that the levels of cyclin D1 were similar between the p27+/– and p27+/+ mammary tumors [protein abundance normalized to 100; 100 ± 5 (p27+/+) versus 97 ± 6 (p27+/–); n = 20; Fig. 3A ]. In the p27+/– mammary tumors, cyclin E levels were reduced by 60% [100 ± 23 (p27+/+) versus 40 ± 14 (p27+/–); n = 20; Fig. 3A], p27Kip1 levels were reduced by 80% [100 ± 25 (p27+/+) versus 20 ± 5.9 (p27+/–); n = 18], and p21Cip1 levels were reduced by 27% [100 ± 25 (p27+/+) versus 73 ± 19 (p27+/–); n = 20; Fig. 3B].


Figure 3
View larger version (23K):
[in this window]
[in a new window]
  		Figure 3. Loss of one p27Kip1 allele increases Skp2 level in MMTV-ErbB2 mammary tumors. MMTV-ErbB2 mammary tumor lysates were analyzed by Western blotting for cyclin D1 and cyclin E (n = 40; A) and p27Kip1 and p21Cip1 (n = 40; B). Columns, mean protein abundance; bars, SE. C, Skp2 abundance determined by Western blotting of mammary tumors. Columns, mean (n = 20); bars, SE. D, Skp2 abundance is increased in tumors with earlier onset.

 
p27Kip1 is regulated at the level of translation and protein turnover, with p27Kip1 ubiquitination by the SCF complex resulting in a 26S proteosome–mediated degradation. As p27Kip1 levels were reduced by >50% in the ErbB2-p27+/– tumors when compared with the p27+/+, we examined the abundance of the SCFSKP2 components (Skp1, Cul1, and Skp2) as depletion of any specific factor prevents p27Kip1 polyubiquitination and subsequent degradation (1–3, 44). The abundance of Skp1 and Cul1 were similar in the tumors from the p27+/+ and p27+/– animals (data not shown). Skp2 levels were increased 5-fold in the p27+/– tumors (Fig. 3C). Tumors with increased Skp2 positively correlated with an earlier tumor onset time (Fig. 3D).

Skp2 levels were increased in the p27+/– tumors, correlating inversely with p27Kip1 levels in murine ErbB2 tumors (Fig. 4A ). To assess whether the relationship between Skp2 and p27Kip1 advance in murine mammary tumors reflected changes in human breast cancer, studies were conducted. For analysis of human breast cancer tumors, immunohistochemical staining for Skp2 was done with a mix of four different affinity-purified mAbs as previously described (28). At least 20 high-power fields were chosen at random and 2,000 cells were counted. Skp2 levels were inversely concluded with p27Kip1 abundance in human ErbB2-positive tumors (Fig. 4B). These findings suggest the increase correlation between Skp2 and p27Kip1 levels in murine tumors may reflect changes in human ErbB2 breast cancer. We examined the relationship between Skp2 and p27Kip1 levels and tumor onset with a three-dimensional pictograph. The three-dimensional modeling showed that high Skp2 and low p27Kip1 levels correlated with earlier tumor onset in MMTV-ErbB2 tumors (Fig. 4C). Skp2 is induced during DNA synthesis. This prior observation raised the possibilities that the relationship between high Skp1 abundance and earlier tumor onset may simply reflect more rapidly proliferating tumors. The DNA synthetic phase fraction was therefore assessed in tumor samples with fluorescence-activated cell sorting analysis (45). The earlier-onset tumors did not have increased S-phase fractions (Fig. 4D); therefore, the increased Skp2 levels do not seem to result from increased DNA synthesis. Collectively, these analyses of murine and human tumors raised the possibility that p27Kip1 may regulate Skp2 abundance.


Figure 4
View larger version (30K):
[in this window]
[in a new window]
  		Figure 4. The SCF complex in the MMTV-ErbB2-p27+/– tumors. A, Skp2 and p27Kip1 levels are inversely correlated in MMTV-ErbB2 mammary tumors and in human ErbB2 positive human breast cancers (Supplementary data; n = 35). B, Skp2 and p27Kip1 levels are inversely correlated in the human ErbB2-positive human breast cancers (n = 35). C, three-dimensional pictograph of p27Kip1, Skp2, and time of tumor onset, showing inverse correlation between Skp2 and p27Kip1. D, the S phase of the ErbB2 tumors is shown with time of onset. E, the relative activity of the Skp2 promoter luciferase reporter was transfected into 3T3 cells in the presence of an activating mutant of ErbB2 (neu-NT) and p27Kip1 or the expression vector for Skp2 (G) or p27Kip1 (F) alone. Columns, mean (n = 8 separate transfections); bars, SE.

 
To determine whether p27Kip1 regulated Skp2 expression, the Skp2 promoter linked to a luciferase reporter gene was examined. Coexpression of an activating ErbB2 mutation enhanced Skp2 promoter activity by 50%. Coexpression of p27Kip1 inhibited ErbB2 activation of Skp2 (Fig. 4E). Coexpression of Skp2 enhanced Skp2 promoter activity 5-fold (Fig. 4F). Coexpression of p27Kip1 repressed Skp2 promoter activity 2-fold (Fig. 4G). Together these studies show that the Skp2 promoter is directly induced by ErbB2 and repressed by p27Kip1.

To investigate further the role of p27Kip1 in tumor onset in the mammary gland of the mouse, we determined the molecular genetic signature regulated by one allele of p27Kip1 in mammary tumors using genome-wide expression analysis. Microarray analysis was conducted on ErbB2 p27+/– versus ErbB2 p27+/+ mammary tumors. Tumors were taken from age-matched controls with similar tumor sizes. Both cDNA microarray and Affymetrix oligonucleotide microarray were used to expand the number of genes examined. We hypothesized that a subset of genes that were relevant to the earlier tumor formation would be different between these two genetically distinct sets of tumors and would include genes that are components of pathways involved in both growth promotion and growth inhibition. RNA was extracted from tumors that developed with the same T50. Using the Affymetrix MG-U74AV2 mouse chip array (12,422 genes represented), microarray hybridization was done for each of six (three wild-type and three heterozygote) tumors. The mean expression levels of each gene were calculated for either the three wild-type or three heterozygote tumors. Mean differences that were statistically different (P < 0.05, Student's t test) and showed a >1.8-fold change were scored. Using the p27Kip1 heterozytote tumors as the reference, 72 identifiable genes were found to be differentially regulated (42 up-regulated and 30 down-regulated; Fig. 5 ; Table 1 ). Within the up-regulated group of genes, nine were found to be involved in collaborative oncogenesis (Cdc25a, Wnt/ß-catenin signaling; i.e., Wnt, Epha 2, matrix metalloproteases11; ref. 46) and cell growth/suvival, five in the immune response, four in membrane transport, three each in protein processing/trafficking and DNA or amino acid metabolism, and two genes were found to be associated with development.


View this table:
[in this window]
[in a new window]
  		Table 1. Differentially expressed genes

 
Using a mouse cDNA chip array (8,969 genes represented), two microarray hybridizations were done for each of three wild-type versus p27Kip1 heterozygote tumor comparisons (tumors from six different animals). For each hybridization, differences that were >1.8-fold were scored. Scored genes from each of the two hybridizations were then pooled. Genes considered to represent consistent differences were defined as differences that were present in two of the three comparisons. With the p27Kip1 heterozygote tumors as a reference, 37 identifiable genes (Fig. 5A; Table 2 ) were found to be differentially regulated (34 up-regulated and 3 down-regulated). The cDNA array analysis extended the microarray analysis by detecting genes not represented or below the level of detection on the affymetrix MG-U74AV2 chip. Of the 34 up-regulated genes, 9 were categorized as functioning in cell growth, 4 were components of the immune response, and 2 were involved in cellular differentiation and protein or RNA processing. Comparison with previous studies indicated that the 19 genes that increased in the ErbB2 p27+/– tumors were previously identified to be increased in estrogen receptor–negative tumors (47, 48), and several genes induced in the ErbB2 p27+/– tumors were similar to previously described ß-catenin-responsive genes (EphA2, Claudin, aquaporin, matrix metalloproteinase, Tcf/Lef1, p27–/– ß-catenin, and Engrailed2; ref. 46).11 To examine further the expression of genes altered by micoarray, we conducted Western blot analysis of Tcf4 as a marker for activation of ß-catenin signaling. The Tcf4 gene is induced by ß-catenin and contributes to the induction of ß-catenin signaling of downstream target genes (49). Western blot analysis was conducted on 20 tumors and the data are shown as mean densitometry indicating a >2-fold increase in Tcf4 and abundance in ErbB2/p27+/– tumor compared with ErbB2/p27+/– (Fig. 5C). Furthermore, analysis of several genes altered in expression by microarray was confirmed by RT-PCR analysis of mRNA from the mammary tumors (Fig. 5D). Collectively, the data of both array systems were consistent with a model in which p27 inhibited several collaborative oncogenic signaling pathways, including Cdc25a and Wnt/ß-catenin signaling.


View this table:
[in this window]
[in a new window]
  		Table 2. Differentially expressed genes: cDNA

 
To examine the possibility that the induction of ß-catenin-responsive genes in the ErbB2/p27+/– mammary tumors may be a function of reduced p27Kip1 levels, we examined the possibility that p27Kip1 may repress ß-catenin signaling. Expression studies were conducted in p27+/– versus p27+/– cells. The transcription factor Siamosis is a ß-catenin-responsive gene involved in the early dorsal ventral patterning (50). The Siamosis promoter was induced 5-fold by ß-catenin S33 (Fig. 6A ). Coexpression of p27Kip1 repressed ß-catenin-induced Siamosis luciferase reporter activity in wild-type and p27–/– cells (Fig. 6B and C). The Siamosis {Delta}Tcf mutant, the c-myc Tcf mutant, and the Smad7 {Delta}Tcf mutant were not repressed by p27Kip1 (data not shown). The c-myc promoter was induced 4-fold by coexpression of ß-catenin S33 as previously shown (51). The point mutant of the Tcf site was not induced by ß-catenin S33 (Fig. 6D). Coexpression of p27Kip1 inhibited ß-catenin-induced c-myc promoter activity in both p27–/– and p27+/+ 3T3 cells (Fig. 6E and F). Deletion of the Smad7 Tcf site increased luciferase activity 3-fold (Fig. 6G). The transfection of the activating ß-catenin S33 enhanced activity of the Smad7 promoter 5-fold. Coexpression of p27Kip1 repressed ß-catenin-induced Smad7 expression in a dose-dependent manner (Fig. 6H). In contrast, the Smad7 {Delta}Tcf mutant was not activated by ß-catenin. p27Kip1 cotransfection repressed Smad7 promoter activity in p27Kip1 wild-type 3T3 cells (Fig. 6I). Together these studies show that p27Kip1 inhibits ß-catenin signaling to multiple distinct target genes.


Figure 6
View larger version (23K):
[in this window]
[in a new window]
  		Figure 6. p27Kip1 inhibits ß-catenin signaling. A, the Siamosis-LUC reporter was assessed for ß-catenin induction and p27Kip1 regulation. Columns, mean relative light units x 103 (n = 9 separate experiments); bars, SE. D to F, the c-Myc-LUC reporter and c-Myc {Delta}Tcf-LUC reporter were assessed for ß-catenin responsiveness and regulation by cotransfected p27Kip1. G to I, the Smad7 promoter luciferase or Smad7 {Delta}Tcf mutant was assessed for luciferase activity in the presence of activating ß-catenin S33 and a transfected p27Kip1 expression vector (G). Transfections were conducted in p27+/+ (I) or p27–/– 3T3 (H) cells as indicated.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Biallelic tumor-specific mutations have provided strong evidence of a causal link between tumorigenesis and several key tumor suppressors including pRb, p53, and INK4a. In the current studies, p27Kip1 (Cdkn1b) functioned as a suppressor of ErbB2-induced mammary gland tumorigenesis. Analysis of the mammary tumors from the p27+/– mice showed that the remaining allele was expressed, suggesting that p27Kip1 is haploinsufficient for tumor suppression. The ability of p27Kip1 to function as a suppressor of ErbB2-induced mammary gland tumorigenesis with haploinsufficiency in vivo extends previous findings in which p27Kip1 was haploinsufficient for suppression of chemical carcinogens or {gamma}-irradiation-induced tumors and APC-mediated colonic carcinogenesis (52). The p27+/– mice in the FVB strain of the current studies developed mammary tumors with a T50 not dissimilar to that of mice in the Bl6/FVB mixed strain in a prior study (53). Therefore, mice heterozygous for deletion of p27Kip1, as has also been reported for transforming growth factor ß(TGF-ß; ref. 54) and Dmp1 (55), show varying degrees of haploinsufficiency and do not conform to Knudson's two-hit model of tumor suppressors. Although only a handful of definitively haploinsufficient tumor suppressor genes have been identified, it has been recently argued that haploinsufficiency in some form may be a hallmark of essentially all tumor suppressors, although the mechanisms involved are unclear (56). The current studies showed that the loss of a single p27Kip1 allele results in disproportionate loss of p27Kip1 protein abundance and activates multiple distinct collaborative oncogenic signaling pathways. Global gene expression profiling and Western blot analysis of ErbB2/p27+/– tumors showed that the loss of p27Kip1 induced genes promoting Wnt/ß-catenin/hedgehog signaling (Dhh, Wnt3, and Lef1), lymphangiogenesis (Flt3l and EphA2), cellular proliferation, and collaborative oncogenic signaling (Cdc25A, Smad7, and Skp2). Reintroduction of p27Kip1 into p27Kip1-deficient cells inhibited expression of Skp2, Smad7, and Wnt/ß-catenin responsive genes. Thus, p27Kip1 inhibits Skp2, Smad7, and Wnt/ß-catenin responsive genes through a mechanism that involves repression of target gene promoter activity.

Analysis of the molecular basis for the p27Kip1 haploinsufficient mammary tumor suppression showed that the loss of one p27Kip1 allele was associated with reduced p27Kip1 levels correlating with earlier time of tumor onset. The accelerated loss of p27Kip1 abundance on loss of a single allele may contribute to the acceleration of tumor onset in ErbB2/p27+/– mammary tumors. Reduced p27Kip1 abundance correlates with poor prognosis in human breast cancer (6–9). p27Kip1 degradation is mediated by the SCF complex following p27Kip1 phosphorylation by cyclin E/Cdk2 (57). Herein, increased Skp2 abundance correlated with earlier onset of MMTV-ErbB2 tumors. Skp2 levels were substantially increased in the p27+/– tumors. Increased Skp2 correlated with reduced p27Kip1 levels in both murine and human breast tumors that overexpress ErbB2 (Fig. 4A and B), consistent with recent findings in human lymphomas, colorectal carcinoma, and oral squamous cell carcinomas (28, 58). Although Skp2 is induced during DNA synthesis, it is unlikely that the increase in Skp2 levels in the p27+/– tumors is a result of altered DNA synthesis per se, as S-phase fractions were similar between the p27+/+ and p27+/– tumors. Herein Skp2 expression was repressed by p27Kip1. Skp2 is a ubiquitylated target of Cul1, and Cul1 inhibits Skp2 expression (59). However, in the current studies, Cul1 levels were unchanged in mammary tumors of mice genetically heterozygous for p27Kip1 (data not shown). As Skp2 abundance is normally induced by DNA synthesis and inhibited by Cul1, Skp2 abundance seems to be uncoupled from normal regulation in ErbB2 tumors. Skp2 promotes contact-independent growth (60); thus, such uncoupling with anomalous p27Kip1 degradation would be predicted to contribute to the aberrant growth advantage. As cyclin E/Cdk2 kinase levels were 40% higher in the p27+/– tumors (not shown), it would be anticipated that Skp2-mediated p27Kip1 degradation would be more efficient in these tumors. Thus, it is unlikely that the increase in Skp2 is compensating for reduced ability to degrade its substrate. An alternative possibility is that Skp2 is induced by oncogenic stimuli and is contributing to the enhanced rate of tumorigenesis observed in the p27+/– tumors. In support of this hypothesis, increased Skp2 levels were found in murine mammary tumors with increased ErbB2 abundance, and oncogenic ErbB2 induced the Skp2 promoter. The current studies suggest that p27Kip1 regulates its own abundance in the presence of ErbB2. The ability of p27Kip1 to regulate Skp2 may contribute to the disproportionate loss of p27Kip1 protein in the p27Kip1 heterozygote mammary tumors.

Analysis of MMTV-ErbB2 p27+/– tumors showed that the genetic deletion of one p27Kip1 allele induced multiple distinct collaborative oncogenic pathways (Dhh, Wnt/ß-catenin, Cdc25A, Skp2, and Smad7). Cdc25A, Smad7, and Skp2 have each been shown to promote contact-independent growth (61–63). DUSP9, which encodes two NH2-terminal CH2 domains homologous to Cdc25A (64), was, like Cdc25A, also induced in ErbB2/p27+/– tumors. p27Kip1 deficiency increased expression of Dhh/Wnt/ß-catenin responsive genes (Tcf, Ephrin receptor, Claudin, Wnt, aquaporin, and matrix metalloproteinases).11 Semiquantitative RT-PCR analysis showed a 50% mean decrease in the mRNA abundance of p27Kip1. Abundance of the ß-catenin responsive gene Tcf4 was increased >2-fold in ErbB2/p27+/– tumor, consistent with the model in which Tcf4 is itself induced by ß-catenin signaling. Transcriptional activity of several ß-catenin responsive promoters was increased in p27–/– cells in a Tcf site–dependent manner. Coexpression of p27Kip1 repressed ß-catenin-dependent activity. Ephrin receptors represent a large family of tyrosine kinases. The interactions between Eph receptors and their ligands involved direct cell-to-cell interactions, frequently resulting in repulsion and contributing to epithelial polarity (65). Ephrin receptor A2, which was induced in ErbB2/p27+/– mammary tumors, promotes tumor vascularization (66). Induction of ß-catenin signaling promotes mammary tumorigenesis and induction of this pathway may have contributed to the more rapid onset of p27Kip1-deficient ErbB2 mammary tumors.

Herein, tumors derived from mice genetically deleted of p27Kip1 showed the induction of Dhh/Wnt/ß-catenin signaling. ß-Catenin is oncogenic in mouse models of tumorigenesis (67, 68) and ß-catenin overrides density-dependent growth inhibition and cooperates with Ras in cellular transformation (69). How might p27Kip1 inhibit expression of the Wnt/ß-catenin pathway? The repression of ß-catenin signaling by p27Kip1 in mammary epithelial cells is consistent with findings that p21Cip1 interrupts the ß-catenin signaling pathway in intestinal cells (65). Activation of the Dhh/Wnt/ß-catenin pathway is thought to expand the mammary gland stem cell compartment that contributes to tumorigenesis (70). In the current studies, p27Kip1 repressed ß-catenin signaling assayed in transient expression studies of multiple distinct Tcf target genes. We showed that the repression of ß-catenin signaling by p27Kip1 required the presence of the Tcf site in the promoter of the target genes assessed. The inhibition of ß-catenin signaling by p27Kip1 may likely contribute an antiproliferative component, the loss of which may in turn enhance the rate of onset of mammary tumorigenesis in the p27Kip1 heterozygote mice.


   Acknowledgments
 
Grant support: NIH Cancer Center Core grant P30 CA56036-08; Canadian Breast Cancer Initiative and Medical Research Council of Canada Scientist award (W.J. Muller); NIH grants R01CA70896, R01CA75503, R01CA86072, and R01CA93596; Pfeiffer Foundation; Susan G. Komen Breast Cancer Foundation (R.G. Pestell); NIH grants AG20337C (C. Albanese) and CA536340 (J.C. Hulit); NIH training grants T32 DK 07513 (R.J. Lee), CA76642, and CA14462 (G. Inghirami); and Breast Cancer Alliance Inc. (A.A. Quong).

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 Drs. E. Bottinger (Department of Medicine, Icahn Medical Institute, Mt. Sinai School of Medicine, New York, NY) and W. Tansey (Cold Spring Harbor Laboratories, Cold Spring Harbor, NY) for plasmids and Dr. M. Pagano for helpful discussions.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

J. Hulit and R.J. Lee contributed equally to the manuscript.

8 Detailed descriptions of microarray hardware and procedures are available from http://www.aecom.yu.edu/home/molgen/facilities.html. Back

9 See http://www.aecom.yu.edu/home/molgen/facilities.html for specifications. Back

10 http://www.microarrays.org/software.html. Back

11 http://www.stanford.edu/~rnusse/wntwindow.html. Back

Received 1/16/06. Revised 5/31/06. Accepted 6/27/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Sutterluty H, Chatelain E, Marti A, et al. p45SKP2 promotes p27Kip1 degradation and induces S phase in quiescent cells. Nat Cell Biol 1999;1:207–14.[CrossRef][Medline]
  2. Rolfe M, Chiu MI, Pagano M. The ubiquitin-mediated proteolytic pathway as a therapeutic area. J Mol Med 1997;75:5–17.[CrossRef][Medline]
  3. Carrano A, Eytan E, Hershko A, Pagano M. SKP2 is required for ubiquitin-mediated degradation of the Cdk inhibitor p27. Nat Cell Biol 1999;1:193–9.[CrossRef][Medline]
  4. Zhang H, Kobayashi R, Galaktionov K, Beach D. p19Skp1 and p45Skp2 are essential elements of the cyclin A-CDK2 S phase kinase. Cell 1995;82:915–25.[CrossRef][Medline]
  5. Nakayama K, Nagahama H, Minamishima YA, et al. Targeted disruption of Skp2 results in accumulation of cyclin E and p27(Kip1), polyploidy and centrosome overduplication. EMBO J 2000;19:2069–81.[CrossRef][Medline]
  6. 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]
  7. Sgambato A, Zhang Y-J, Arber N, et al. Deregulated expression of p27Kip1 in human breast cancers. Clin Cancer Res 1997;3:1879–87.[Abstract]
  8. Fredersdorf S, Burns J, Milne AM, et al. High level expression of p27kip1 and cyclin D1 in some human breast cancer cells: Inverse correlation between the expression of p27kip1 and degree of malignancy in human breast and colorectal cancers. Proc Natl Acad Sci U S A 1997;94:6380–5.[Abstract/Free Full Text]
  9. Porter PL, Malone KE, Heagerty PJ, et al. Expression of cell-cycle regulators p27Kip1 and cyclin E, alone and in combination, correlate with survival in young breast cancer patients. Nat Med 1997;3:222–5.[CrossRef][Medline]
  10. Ponce-Castañeda MV, Lee M-H, Latres E, et al. p27Kip1: chromosomal mapping to 12p12-12p13.1 and absence of mutations in human tumors. Cancer Res 1995;55:1211–4.[Abstract/Free Full Text]
  11. Kawamata N, Morosetti R, Miller CW, et al. Molecular analysis of the cyclin-dependent kinase inhibitor gene p27/Kip1 in human malignancies. Cancer Res 1995;55:2266–9.[Abstract/Free Full Text]
  12. Pietenpol JA, Bohlander SK, Sato Y, et al. Assignment of the human p27Kip1 gene to 12p13 and its analysis in leukemias. Cancer Res 1995;55:1206–10.[Abstract/Free Full Text]
  13. Stegmaier K, Pendse S, Barker GF, et al. Frequent loss of heterozygosity at the TEL gene locus in acute lymphoblastic leukemia of childhood. Blood 1995;86:38–44.[Abstract/Free Full Text]
  14. Hatta Y, Takeuchi S, Yokota J, Koeffler HP. Ovarian cancer has frequent loss of heterozygosity at chromosome 12p12.3-13.1 (region of TEL and Kip1 loci) and chromosome 12q23-ter: evidence for two new tumour-suppressor genes. Br J Cancer 1997;75:1256–62.[Medline]
  15. Slingerland J, Pagano M. Regulation of the cdk inhibitor p27 and its deregulation in cancer. J Cell Physiol 2000;183:10–7.[CrossRef][Medline]
  16. Weinstein IB. Disorders in cell circuitry during multistage carcinogenesis:the role of homeostasis. Carcinogenesis 2000;21:857–64.[Abstract/Free Full Text]
  17. Kiyokawa H, Kineman RD, Manova-Todorova KO, et al. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27Kip1. Cell 1996;85:721–32.[CrossRef][Medline]
  18. Fero ML, Rivkin M, Tasch M, et al. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27Kip1-deficient mice. Cell 1996;85:733–44.[CrossRef][Medline]
  19. Nakayama K, Ishida N, Shirane M, et al. Mice lacking p27Kip1 display increased body size, multiple organ hyperplasia, retinal dysplasia, and pituitary tumors. Cell 1996;85:707–20.[CrossRef][Medline]
  20. Philipp-Staheli J, Kim KH, Payne SR, et al. Pathway-specific tumor suppression. Reduction of p27 accelerates gastrointestinal tumorigenesis in Apc mutant mice, but not in Smad3 mutant mice. Cancer Cell 2002;1:355–68.[CrossRef][Medline]
  21. Di Cristofano A, De Acetis M, Koff A, Cordon-Cardo C, Pandolfi PP. Pten and p27 KIP1 cooperate in prostate cancer tumor suppression in the mouse. Nat Genet 2001;27:222–4.[CrossRef][Medline]
  22. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987;235:177–82.[Abstract/Free Full Text]
  23. Muller WJ, Sinn E, Pattengale PK, Wallace R, Leder P. Single step induction of mammary adenocarcinoma in transgenic mice bearing the activated c-neu oncogene. Cell 1988;54:105–15.[CrossRef][Medline]
  24. Guy CT, Webster MA, Schaller M, Parson TJ, Cardiff RD, Muller WJ. Expression of the neu proto-oncogene in the mammary epithelium of transgenic mice induces metastatic disease. Proc Natl Acad Sci U S A 1992;89:10578–82.[Abstract/Free Full Text]
  25. 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]
  26. Yu Q, Geng Y, Sicinski P. Specific protection against breast cancers by cyclin D1 ablation. Nature 2001;411:1017–21.[CrossRef][Medline]
  27. Yang HY, Zhou BP, Hung MC, Lee MH. Oncogenic signals of HER-2/neu in regulating the stability of the cyclin-dependent kinase inhibitor p27. J Biol Chem 2000;275:24735–9.[Abstract/Free Full Text]
  28. Latres E, Chiarle R, Schulman BA, et al. Role of the F-box protein Skp2 in lymphomagenesis. Proc Natl Acad Sci U S A 2001;98:2515–20.[Abstract/Free Full Text]
  29. Chiarle R, Budel LM, Skolnik J, et al. Increased proteasome degradation of cyclin-dependent kinase inhibitor p27 is associated with adecreased overall survival in mantle cell lymphoma. Blood 2000;95:619–26.[Abstract/Free Full Text]
  30. Albanese C, Wu K, D'Amico M, et al. IKK{alpha} regulates mitogenic signaling through transcriptional induction of cyclin D1 via Tcf. Mol Biol Cell 2003;14:585–99.[Abstract/Free Full Text]
  31. Brannon M, Gomperts M, Sumoy L, Moon RT, Kimelman D. A ß-catenin/XTcf-3 complex binds to the siamois promoter to regulate dorsal axis specification in Xenopus. Genes Dev 1997;11:2359–70.[Abstract/Free Full Text]
  32. von Gersdorff G, Susztak K, Rezvani F, Bitzer M, Liang D, Bottinger EP. Smad3 and Smad4 mediate transcriptional activation of the human Smad7 promoter by transforming growth factor ß. J Biol Chem 2000;275:11320–6.[Abstract/Free Full Text]
  33. Mendez J, Zou-Yang H, Kim S-Y, Hidaka M, Tansey WP, Stillman B. Human origin recognition complex large subunit is degraded by ubiquitin-mediated proteolysis after initiation of DNA replication. Mol Cell 2002;9:481–91.[CrossRef][Medline]
  34. Kim SY, Herbst A, Tworkowski KA, Salghetti SE, Tansey WP. Skp2 regulates Myc protein stability and activity. Mol Cell 2003;11:1177–88.[CrossRef][Medline]
  35. Shtutman M, Zhurinsky J, Simcha I, et al. The cyclin D1 gene is a target of the b-catenin/LEF-1 pathway. Proc Natl Acad Sci U S A 1999;96:5522–7.[Abstract/Free Full Text]
  36. Watanabe G, Howe A, Lee RJ, et al. Induction of cyclin D1 by simian virus 40 small tumor antigen. Proc Natl Acad Sci U S A 1996;93:12861–6.[Abstract/Free Full Text]
  37. Bearss DJ, Lee RJ, Troyer DA, Pestell RG, Windle JJ. Differential effects of p21(WAF1/CIP1) deficiency on MMTV-ras and MMTV-myc mammary tumor properties. Cancer Res 2002;62:2077–84.[Abstract/Free Full Text]
  38. Huang E, Ishida S, Pittman J, et al. Gene expression phenotypic models that predict the activity of oncogenic pathways. Nat Genet 2003;34:226–30.[CrossRef][Medline]
  39. Sakamaki T, Casimiro MC, Ju X, et al. Cyclin d1 determines mitochondrial function in vivo. Mol Cell Biol 2006;26:5449–69.[Abstract/Free Full Text]
  40. Drinkwater NR, Bennett LM. Genetic control of carcinogenesis in experimental animals. Prog Exp Tumor Res 1991;33:1–20.[Medline]
  41. Donehower LA, Godley LA, Aldaz CM, et al. Deficiency of p53 accelerates mammary tumorigenesis in Wnt-1 transgenic mice and promotes chromosomal instability. Genes Dev 1995;9:882–95.[Abstract/Free Full Text]
  42. DiGiovanni J. Genetic factors controlling responsiveness to skin tumor promotion in mice. Prog Clin Biol Res 1995;391:195–212.[Medline]
  43. Cardiff RD, Anver MR, Gusterson BA, et al. The mammary pathology of genetically engineered mice: the consensus report and recommendations from the Annapolis meeting. Oncogene 1999;19:968–88.
  44. Tsvetkov LM, Yeh KH, Lee SJ, Sun H, Zhang H. p27(Kip1) ubiquitination and degradation is regulated by the SCF(Skp2) complex through phosphorylated Thr187 in p27. Curr Biol 1999;9:661–4.[CrossRef][Medline]
  45. Krishan A. Rapid flow cytofluorometric analysis of mammalian cell cycle by propidium iodide staining. J Cell Biol 1975;66:188–95.[Abstract/Free Full Text]
  46. van de Wetering M, Sancho E, Verweij C, et al. The ß-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer cells. Cell 2002;111:241–50.[CrossRef][Medline]
  47. van de Vijver MJ, He YD, van't Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002;347:1999–2009.[Abstract/Free Full Text]
  48. van't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530–6.[CrossRef][Medline]
  49. Nateri A, Spencer-Dene B, Behrens A. Interaction of phosphorylated c-Jun with TCF4 regulates intestinal cancer development. Nature 2005;437:281–5.[CrossRef][Medline]
  50. Fan MJ, Gruning W, Walz G, Sokol SY. Wnt signaling and transcriptional control of Siamois in Xenopus embryos. Proc Natl Acad Sci U S A 1998;95:5626–31.[Abstract/Free Full Text]
  51. He TC, Sparks AB, Rago C, et al. Identification of c-MYC as a target of the APC pathway. Science 1998;281:1509–12.[Abstract/Free Full Text]
  52. Fero ML, Randel E, Gurley KE, Roberts JM, Kemp CJ. The murine p27Kip1 is haplo-insufficient for tumour suppression. Nature 1998;396:177–80.[CrossRef][Medline]
  53. Muraoka RS, Lenferink AEG, Simpson J, et al. Cyclin-dependent kinase inhibitor p27 Kip1 is required for mouse mammary gland morphogenesis and function. J Cell Biol 2001;153:917–31.[Abstract/Free Full Text]
  54. Tang B, Bottinger EP, Jakowlew SB, et al. Transforming growth factor-ß1 is a new form of tumor suppressor with true haploid insufficiency. Nat Med 1998;4:802–7.[CrossRef][Medline]
  55. Inoue K, Zindy F, Randle DH, Rehg JE, Sherr CJ. Dmp1 is haplo-insufficient for tumor suppression and modifies the frequencies of Arf and p53 mutations in Myc-induced lymphomas. Genes Dev 2001;15:2934–9.[Abstract/Free Full Text]
  56. Quon KC, Berns A. Haplo-insufficiency? Let me count the ways. Genes Dev 2001;15:2917–21.[Free Full Text]
  57. Montagnoli A, Fiore F, Eytan E, et al. Ubiquitination of p27 is regulated by Cdk-dependent phosphorylation and trimeric complex formation. Genes Dev 1999;13:1181–9.[Abstract/Free Full Text]
  58. Hershko D, Bornstein G, Ben-Izhak O, et al. Inverse relation between levels of p27(Kip1) and of its ubiquitin ligase subunit Skp2 in colorectal carcinomas. Cancer 2001;91:1745–51.[CrossRef][Medline]
  59. Wirbelauer C, Sutterluty H, Blondel M, et al. The F-box protein Skp2 is a ubiquitylation target of a Cul1-based core ubiquitin ligase complex: evidence for a role of Cul1 in the suppression of Skp2 expression in quiescent fibroblasts. EMBO J 2000;19:5362–75.[CrossRef][Medline]
  60. Pagano M, Tam SW, Theodoras AM, et al. Role of the ubiquitin-proteasome pathway in regulating abundance of the cyclin-dependent kinase inhibitor p27. Science 1995;269:682–5.[Abstract/Free Full Text]
  61. Carrano AC, Pagano M. Role of the F-box protein Skp2 in adhesion-dependent cell cycle progression. J Cell Biol 2001;153:1381–90.[Abstract/Free Full Text]
  62. Galaktionov K, Lee AK, Eckstein J, et al. CDC25 phosphatase as potential human oncogenes. Science 1995;269:1575–7.[Abstract/Free Full Text]
  63. Liu X, Lee J, Cooley M, Bhogte E, Hartley S, Glick A. Smad7 but not Smad6 cooperates with oncogenic ras to cause malignant conversion in a mouse model for squamous cell carcinoma. Cancer Res 2003;63:7760–8.[Abstract/Free Full Text]
  64. Christie GR, Williams DJ, Macisaac F, Dickinson RJ, Rosewell I, Keyse SM. The dual-specificity protein phosphatase DUSP9/MKP-4 is essential for placental function but is not required for normal embryonic development. Mol Cell Biol 2003;25:8323–33.
  65. Batlle E, Henderson JT, Beghtel H, et al. ß-Catenin and TCF mediate cell positioning in the intestinal epithelium by controlling the expression of EphB/EphrinB. Cell 2002;111:251–63.[CrossRef][Medline]
  66. Ogawa K, Pasqualini R, Lindberg RA, Kain R, Freeman AL, Pasquale EB. The ephrin-A1 ligand and its receptor, EphA2, are expressed during tumor neovascularization. Oncogene 2000;19:6033–42.[CrossRef][Medline]
  67. Gat U, DasGupta R, Degenstein L, Fuchs E. De novo hair follicle morphogenesis and hair tumors in mice expressing a truncated ß-catenin in skin. Cell 1998;95:605–14.[CrossRef][Medline]
  68. Chan TA, Hermeking H, Lengauer C, Kinzler KW, Vogelstein B. 14-3-3 Sigma is required to prevent mitotic catastrophe after DNA damage. Nature 1999;401:616–20.[CrossRef][Medline]
  69. Damalas A, Kahan S, Shtutman M, Ben-Ze'ev A, Oren M. Deregulated ß-catenin induces a p53- and ARF-dependent growth arrest and cooperates with Ras in transformation. EMBO J 2001;20:4912–22.[CrossRef][Medline]
  70. Li Y, Welm B, Podsypanina K, et al. Evidence that transgenes encoding components of the Wnt signaling pathway preferentially induce mammary cancers from progenitor cells. Proc Natl Acad Sci U S A 2003;100:15853–8.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Nutr.Home page
P. A. Davis, M. Jenab, J. P. Vanden Heuvel, T. Furlong, and S. Taylor
Tree Nut and Peanut Consumption in Relation to Chronic and Metabolic Diseases Including Allergy
J. Nutr., September 1, 2008; 138(9): 1757S - 1762S.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
Z. Li, C. Wang, X. Jiao, S. Katiyar, M. C. Casimiro, G. C. Prendergast, M. J. Powell, and R. G. Pestell
Alternate Cyclin D1 mRNA Splicing Modulates p27KIP1 Binding and Cell Migration
J. Biol. Chem., March 14, 2008; 283(11): 7007 - 7015.
[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 Hulit, J.
Right arrow Articles by Pestell, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hulit, J.
Right arrow Articles by Pestell, R. G.

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