Cancer Research Cancer Epigenetics Sign up for Cancer Research eTOC's
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Allen, T.
Right arrow Articles by Lobe, C. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Allen, T.
Right arrow Articles by Lobe, C. G.
[Cancer Research 66, 1294-1301, February 1, 2006]
© 2006 American Association for Cancer Research

Molecular Biology, Pathobiology, and Genetics

Grg1 Acts as a Lung-Specific Oncogene in a Transgenic Mouse Model

Thaddeus Allen1,3, Minke van Tuyl6, Pratibha Iyengar7, Serge Jothy4,8, Martin Post5,6, Ming-Sound Tsao4,7 and Corrinne G. Lobe1,2,3

1 Molecular and Cellular Biology Division, Sunnybrook and Women's College Health Science Centre; 2 Toronto-Sunnybrook Regional Cancer Centre; Departments of 3 Medical Biophysics, 4 Laboratory Medicine and Pathobiology, and 5 Physiology, University of Toronto; 6 Lung Biology Research Program, Hospital for Sick Children Research Institute; 7 Ontario Cancer Institute, University Health Network-Princess Margaret Hospital; and 8 St. Michael's Hospital, Toronto, Ontario, Canada

Requests for reprints: Corrinne G. Lobe, Molecular and Cellular Biology Division, Sunnybrook and Women's College Health Science Centre, S-236, 2075 Bayview Avenue, Toronto, ON, Canada M4N 3M5. Phone: 416-480-6100, ext. 3382; Fax: 416-480-5703; E-mail: corrinne.lobe{at}swri.on.ca.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Groucho proteins are transcriptional corepressors that are recruited to gene regulatory regions by numerous transcription factors. Long isoforms, such as Grg1, have all the domains of the prototype Drosophila Groucho. Short Groucho proteins, such as Grg5, have only the amino-terminal Q and G/P domains. We generated Grg1 and Grg5 transgenic mice and found that Grg1 overexpression induces lung adenocarcinoma, whereas Grg5 overexpression does not. Coexpression of Grg5 with Grg1 reduces tumor burden. Grg1 and Grg5 both diminish p53 protein levels; however, only Grg1 overexpression induces elevated levels of ErbB1 and ErbB2 receptor tyrosine kinases. The molecular and biological changes that accompany tumor progression in Grg1 transgenic mice closely reiterate events seen in human lung cancer. We also found that within a human lung tumor tissue array, a significant number of carcinomas overexpress Grg1/TLE1. Our data suggest that Grg1 overexpression contributes to malignancy in human lung cancers. (Cancer Res 2006; 66(3): 1294-301)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Human lung cancer is frequently associated with activation of oncogenes, such as K-ras, MYC, ErbB1, and ErbB2, and loss of tumor-suppressor genes, such as p53, Rb, and p16INK4A (1, 2). Some of these aberrations have been modeled in mice using transgenic and gene knockout strategies. Based on the high incidence of K-ras mutations in human lung adenocarcinoma, transgenic mice that express activated K-ras were generated (3–6). The K-ras–expressing mice develop lung tumors resembling human papillary adenocarcinoma. Loss of heterozygosity at 17p, the tp53 locus, correlates with tumor progression in human non–small cell lung cancer (NSCLC; ref. 7). Likewise, loss of p53 accelerates lung tumorigenesis in the mutant K-ras transgenic mice (3, 5). These animal models have validated the roles of K-ras and p53 gene mutations in lung cancer pathogenesis and represent valuable tools to study genetic modifiers and test therapeutic strategies. A significant number of lung tumors, however, do not harbor K-ras and/or p53 mutations.

This investigation reports the use of a novel Cre/loxP-based system for the generation of conditional transgenic mice. The system is based on the Z/AP Cre reporter (8) and through crosses with different Cre recombinase transgenic lines, it allows the generation of mice with either widespread or tissue-specific transgene expression. We made use of this system to overexpress Groucho-related genes (Grg) in the mouse. Groucho proteins are transcriptional corepressors that interact with a number of transcription factors, histone deacetylase complex (HDAC) molecules, and hypoacetylated histones (9–13). The Groucho family of proteins is encoded by Grg1-6 in mouse (14–19). They are called transducin-like enhancer-of-split (TLE) genes in humans (20). Two major protein isoforms have been characterized. Full-length Groucho proteins possess all the domains of the prototype Drosophila Groucho protein. Shorter isoforms, such as Grg5, and alternatively spliced variants, such as Grg1-S and Grg3b, encode only the amino-terminal Q domain and a G/P domain (17, 18, 21). The Q domain is a tetramerization domain (22) and it has been suggested that the short Groucho proteins inhibit the activity of the long proteins by forming nonfunctional complexes (23–25). The short Groucho proteins are not able to bind HDAC molecules (26) and in some cases have been shown to enhance, rather than repress, transcriptional activation (26, 27). Despite these findings, amino-terminal enhancer of split, the human homologue of Grg5, has been shown to function as a transcriptional corepressor with the androgen receptor and p65 subunit of nuclear factor-{kappa}B (28, 29). The function of short Groucho proteins may, therefore, be context dependent.

Here, we report that transgenic mice with widespread Grg1 overexpression specifically develop lung tumors starting at 1 month of age. Coexpression of Grg5 lowers Grg1-induced tumor burden, suggesting a mechanism that is sensitive to levels of long and short Groucho isoforms. Development of lung tumors in the Grg1-overexpressing mice is associated with altered status of p53 and up-regulation of the ErbB1 and ErbB2 receptor tyrosine kinases. Examination of the expression of TLE1, the human Grg1 homologue, on a human lung cancer tissue array revealed that TLE1 is overexpressed in a significant number of both squamous cell carcinomas and adenocarcinomas. Cumulatively, the results of this study show that Grg1 can act as an oncoprotein in the lung.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Construction of Grg1 and Grg5 conditional transgenic mice. The Cre-conditional expression vector iZ/AP was created to achieve controlled expression in transgenic mice. The iZ/AP vector is a modified version of the Z/AP reporter construct (8) in which the human alkaline phosphatase (hPLAP) coding sequence was replaced by an Internal Ribosomal Entry Site-hPLAP cassette (IRES-hPLAP). The Grg1 and Grg5 coding sequences were ligated to unique BglII/XhoI sites upstream of the IRES-hPLAP cassette to create the vectors iZ/AP-Grg1 and iZ/AP-Grg5. The R1 embryonic stem cell line was maintained essentially as described (30). Following electroporation with iZ/AP-Grg1 and iZ/AP-Grg5 DNA, embryonic stem cells were selected in G418 and subjected to a screening strategy previously used to select single-copy embryonic stem cell clones with strong transgene expression (8, 31, 32). Selected clones were used to make aggregation chimeras and males were mated to CD1 females to identify germ-line transmitters. Ubiquitous Cre-excision was achieved by crossing transgenic offspring to the pCX-NLS-Cre mouse line (33).

Mouse genotyping. Mice were genotyped by either lacZ or hPLAP staining of ear punch tissue (8) or by Southern blot analysis. The probe for Southern analysis of the Grg transgenes was an XhoI/SacII–excised fragment of the iZ/AP vector that overlapped the 3' region of the IRES-hPLAP cassette. The probe for Southern analysis of the pCX-NLS-Cre transgene was a XhoI/BglII fragment from the Cre coding sequence. 32P-labeled probes were created by random priming using the Megaprime DNA-labeling kit (Amersham Pharmacia Biotech, Oakville, Canada).

Quantitation of tumors. Lungs were dissected, fixed in 4% paraformaldehyde, and examined for the presence of tumors on the pleural surface under x10 magnification. Serial sectioning of paraffin-embedded lungs was done to examine for tumors not visible on the pleural surface. Five-micrometer sections were cut at 100 µm step intervals throughout the lungs and sections were stained with H&E. Sections were scanned for adenomatous/carcinomatous proliferation at x10 magnification. Mice with at least one tumorous lesion found in serial sections were counted as tumor positive. For quantification, 10 serial sections of the left pulmonary lobe were scored. Statistical significance was measured using a t test.

Histoenzymological staining. Fixation and staining of tissues, tissue culture plates and frozen sections for ß-galactosidase and alkaline phosphatase activity were essentially as described (8). All slides of frozen sections were counterstained with Nuclear Fast Red (Sigma-Aldrich, Oakville, Canada).

Immunohistochemistry. Paraffin sections (4-5 µm) were incubated with a mouse monoclonal antibody specific to thyroid transcription factor-1 (TTF-1; 1:50, LabVision/NeoMarkers, Fremont, CA), a rabbit polyclonal antibody for cytokeratin, wide spectrum screening (1:800, DakoCytomation, Inc., Mississauga, Canada), a mouse monoclonal antibody specific to vimentin (1:600, Sigma-Aldrich), or a mouse monoclonal antibody for proliferating cell nuclear antigen (PCNA, 1:200, Zymed Laboratories, Inc., Burlington, Canada). Primary dilutions were made in blocking solution (5% normal goat serum and 1% bovine serum albumin in PBS) and incubation was at room temperature overnight. Sections were subsequently incubated with biotinylated secondary antibodies, avidin-biotin complex and 3,3'-diaminobenzidine (DAB, Vector Laboratories, Inc., Burlington, Canada). For TTF-1 and PCNA, nickel was added to the DAB solution to enhance black nuclear staining. Sections were counterstained with methyl green (TTF-1 and PCNA) or hematoxylin (cytokeratin and vimentin).

In situ hybridization. PCR fragments of rat SP-C and CC10 were generated and cloned into the EcoRI site of the PCR2.1 vector (315 and 330 bp, respectively). Digoxigenin-labeled antisense mRNA complementary to SP-C and CC10 were produced by in vitro transcription according to the instructions of the manufacturer (Roche, Mississauga, Canada). In situ hybridization was done as described (34).

Electron microscopy. Fragments of tumors were fixed in glutaradehyde and postfixed in osmium tetroxide for electron microscopy according to standard procedures.

Western blot analysis. Western analysis was carried out using standard protocols. Rabbit polyclonal serum directed against Grg5 was created using a COOH-terminal 18 amino acid peptide.9 {alpha}-TLE1-specific rabbit polyclonal antiserum used in some of the experiments was the gift of Dr. Stefano Stifani (McGill University, Montreal, Quebec, Canada; ref. 35). Other primary antibodies were rabbit polyclonal {alpha}-TLE1 M-101, rabbit polyclonal {alpha}-ErbB1 C-20, rabbit polyclonal {alpha}-ErbB2 C-18 (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit polyclonal {alpha}-Mdm2 H221 (Santa Cruz Biotechnology), mouse monoclonal {alpha}-ß-galactosidase (Promega, Nepean, Canada), mouse monoclonal {alpha}-p53 Ab-1 (Calbiochem, Hornby, Canada), {alpha}-phospho-ErbB2-Y1248 rabbit IgG, {alpha}-pan-Ras mouse monoclonal antibody RAS10 (Upstate Biotechnology, Lake Placid, NY), mouse monoclonal anti-ß-actin (Sigma-Aldrich), mouse monoclonal {alpha}-phospho-p44/42 and rabbit polyclonal {alpha}-p44/42 (Cell Signaling, Danvers, MA). Horseradish peroxidase–conjugated secondary antibodies were purchased from Santa Cruz Biotechnology.

Transfection of NIH 3T3 cells. NIH 3T3 cells were maintained in DMEM, supplemented with 10% fetal bovine serum. Cells were grown at 37°C in a humidified atmosphere of 5% CO2. Standard calcium phosphate–mediated transfections were carried out for 16 hours. Transfections were carried out with a total of 5 µg DNA per well for six-well plates, using pCX-Grg1-IRES-hPLAP, pCX-Mdm2 expression constructs, and pBluescript DNA. Protein lysates were collected 72 hours following transfection.

Measurement of Ras activity. Ras activity in normal lung and lung tumors was measured using a Ras activation assay kit (Upstate Biotechnology) according to the instructions of the manufacturer. The antibodies in this assay do not distinguish between H-, K-, or N-Ras.

Human lung tumor analysis. Expression of TLE1 in human lung cancer samples was analyzed using tissue microarrays. Tissue microarrays were made using the Beecher Instruments tissue microarrayer (Silver Spring, MD) according to the instructions of the manufacturer. The tissue microarrays included 71 adenocarcinomas, 46 squamous cell carcinomas, 8 large cell undifferentiated carcinomas, 3 adenosquamous carcinomas, 2 carcinoid tumors, and 1 small cell carcinoma. Immunohistochemistry was done using rabbit polyclonal {alpha}-TLE1 M-101 using the indirect peroxidase-antiperoxidase technique with 1/150 dilution following microwave treatment of the tissue microarray sections. Immunoreactivity was graded as 0 (none), 1 (slight), 2 (moderate), or 3 (strong).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Transgenic mice with Cre-conditional expression of Groucho proteins. The transgene vector used for conditional Grg expression contained a loxP-flanked STOP sequence between the strong cytomegalovirus enhancer chicken ß-actin promoter (36) and the Grg coding sequence (Fig. 1A). In this configuration, the promoter drives expression of the STOP sequence, consisting of the ß-geo (ß-galactosidase-neomycinR) fusion gene and three polyadenylation sequences. When Cre is introduced, the STOP sequence is excised and the Grg gene is expressed. A second reporter, the hPLAP gene, is coexpressed with Grg from an IRES. The transgenic system is designed to take advantage of tissue-specific Cre recombinase–expressing mouse lines to tailor gene expression to particular tissues. Henceforth, in this text, we refer to the transgenic mice without Cre excision as Grg1lacZ and Grg5lacZ and after Cre excision as Grg1hPLAP and Grg5hPLAP, denoting the expression of histochemical markers before and following Cre excision.


Figure 1
View larger version (46K):
[in this window]
[in a new window]
  		Figure 1. Overexpression of Grg1 and Grg5 in transgenic mice. A, the iZ/AP transgene for Cre-conditional expression of Grg1 or Grg5. B, histoenzymological staining of tissues from Grg1 transgenic mice without Cre recombination (Grg1lacZ) and double-transgenic Grg1/pCX-NLS-Cre mice (Grg1hPLAP). C, Western analysis showing ß-galactosidase expression in tissues of Grg1lacZ mice and increased Grg1 protein in tissues of Grg1hPLAP mice. D, Western analysis for ß-galactosidase and Grg5 in Grg5lacZ and Grg5hPLAP mice.

 
To assess the fidelity of the transgenic system, we initially crossed Grg1lacZ and Grg5lacZ transgenic mice with the pCX-NLS-Cre transgenic line (33), which provides ubiquitous Cre excision. One mouse line each for Grg1 and Grg5 produced viable double-transgenic offspring in a normal Mendelian ratio. Southern blot analysis showed that the transgenes had been excised globally (data not shown). Further analysis by lacZ/hPLAP staining showed the single-transgenic Grg1 and Grg5 mice had widespread lacZ expression, whereas double-transgenic Grg1/Cre and Grg5/Cre mice had no lacZ expression but widespread hPLAP expression (Fig. 1B). Some tissues, such as liver, displayed hPLAP but not lacZ activity, as observed previously for Z/AP and Z/EG transgenic mice (8, 32), suggesting the ß-galactosidase protein is not well expressed or tolerated in those organs. Western blot analysis showed that the hPLAP reporter activity correctly reflected Grg1 and Grg5 protein expression in the double-transgenic mice both during embryonic development (data not shown) and in adults (Fig. 1C and D). Sublethal expression of Grg protein was not the case with every transgenic line produced. A second Grg5 line with stronger transgene expression had an embryonic lethal phenotype when crossed to either pCX-NLS-Cre or tissue-specific Cre recombinase mice. We elected to take advantage of the sublethal expression of the lines used in this study to examine the effects of continuous, widespread Grg expression in adult animals.

Grg1 expression leads to lung adenocarcinoma. Examination of adult transgenic mice revealed that Grg1hPLAP mice developed lung tumors with 100% penetrance (Fig. 2). Offspring of Grg1hPLAP mice that retain the Grg1 transgene but lack the Cre transgene have the same tumor phenotype. No tumors were observed in Grg1lacZ mice; thus, the tumor phenotype is not associated with the transgene insertion site. The Grg5lacZ, Grg5hPLAP, and double-transgenic Z/AP reporter/pCX-NLS-Cre control mice that express the hPLAP reporter but not Grg1 (8), also did not develop lung tumors. The lung neoplasms in the Grg1hPLAP mice first appeared at 1 month as focal proliferation of tall columnar epithelial cells on preexisting alveolar septae, thus resembling early noninvasive bronchioloalveolar carcinoma (arrow; Fig. 2B and F). At 3 to 5 months, visible tumor nodules were noted on the pleural surface (Fig. 2C and G). These tumors showed more extensive proliferation of these neoplastic epithelium but preserving their bronchioloalveolar growth pattern. Between 5 and 6 months, the number of lesions visible on the pleural surface increased substantially, from 3.2 ± 2.2 to 8.3 ± 6.0 per mouse. By 8 months, some of these lesions became confluent and appeared conspicuously as whitish solid tumor masses that are characteristic of invasive cancer. Such tumors replaced large portions of the normal lung parenchyma and showed areas of necrosis (Fig. 2E, I, and J). Our observations suggest that Grg1, but not Grg5, is able to act as a lung-specific oncogene in the mouse.


Figure 2
View larger version (82K):
[in this window]
[in a new window]
  		Figure 2. Time course of lung tumorigenesis in Grg1hPLAP mice. A, tabular summary of lung tumor burden in Grg overexpressing mice. B, at 1 month, no lesions are visible on the pleural surface of the lung, but in serial sections (F) small lesions can be seen that are internal to the lung (arrow). C, at 3 months, small lesions (arrowhead) have become visible on the pleural surface. These are small adenoma-like lesions (G). D, at 6 months, many glassy lesions are visible on the pleural surface that show a strong bronchioalveolar growth pattern (H). By 8 months, large adenocarcinomas are visible (E) and in some mice large regions of lung parenchyma have been replaced by tumor (I and J). K, higher magnification shows mucin-secreting adenocarcinoma (T) with papillary growth and destruction of preexisting alveolar architecture and stroma (S) invasion (arrow) by tumor cells, some of which show goblet cell appearance (arrowheads). Original magnifications of microimages: F, x100; I, x50; J, x16; K, x200.

 
The Grg1hPLAP lung tumors resemble human mucinous adenocarcinoma. Tumor cells were composed of tall columnar epithelial cells with basally located nuclei as well as goblet-like cells that contain periodic acid-Schiff–positive and diastase-resistant mucin in their apical cytoplasm (Fig. 2K). Tumor cells showed a mixture of papillary, acinar, as well as bronchioloalveolar growth patterns. In addition, there was prominent extracellular mucin secretion leading to destruction of preexisting alveolar septae. The overall histopathologic findings are consistent the equivalent of human lung mucinous (colloid) adenocarcinoma. Electron microscopy of tumor samples showed that the alveoli and terminal bronchioles were lined by dysplastic epithelial cells with pleomorphic and enlarged irregularly shaped nuclei (data not shown). A high percentage of tumor cells in the Grg1hPLAP mice stained positive for PCNA and cytokeratin but were negative for vimentin, CC10 (marker of bronchiolar Clara cells), and SP-C (marker of alveolar type II cells; refs. 32, 33), as well as for TTF-1/Nkx2.1 (data not shown). The absence of TTF-1 staining is frequently observed in human mucin-secreting pulmonary adenocarcinoma (37, 38).

Coexpression of Grg5 with Grg1 decreases tumor burden. Grg1hPLAP mice and Grg5hPLAP mice were both fertile, allowing for the generation of Grg1hPLAP/Grg5hPLAP double-transgenic mice. Southern blot analysis was used to confirm the inheritance of both transgenes and Western blot analysis showed that both proteins were abundantly expressed (data not shown). Aside from lung tumors, there were no additional overt phenotypic consequences associated with the coexpression of Grg1 and Grg5. We examined the lungs of 6-month-old coexpressing mice for lung tumor progression and found that lung tumor burden was reduced by the coexpression of Grg5 with Grg1 (Figs. 2A and 3). The percentage of mice with tumors visible on the pleural surface dropped from 96% (27 of 28 mice examined) to 50% (6 of 12 mice), whereas the average number of tumors visible on the pleural surface was reduced from 8.3 ± 6.0 to 2.4 ± 3.7 (P < 0.05; Fig. 2A). In addition to the decreased number of tumors, the tumors were substantially smaller in double-transgenic Grg1/Grg5 mice (Fig. 3A and B). These results suggest that the mechanism of tumorigenesis is sensitive to the ratio of long and short Grg protein isoforms.


Figure 3
View larger version (101K):
[in this window]
[in a new window]
  		Figure 3. Coexpression of Grg5 lowers Grg1-induced lung tumor burden. A, two large adenomas in the lung of a 6-month-old Grg1hPLAP mouse (arrows, H&E-stained section). B, in contrast, adenomas found in 6-month-old Grg1hPLAP/Grg5hPLAP lungs are substantially smaller (arrows). At higher magnification, the morphology of cells comprising the Grg1hPLAP (C) and Grg1hPLAP/Grg5hPLAP (D) lesions appear histologically similar.

 
Grg overexpression affects the tumor suppressor protein p53. The expression of proteins that are typically altered in human lung carcinogenesis was examined in Grg1-overexpressing mice. At 1 month of age, Grg1hPLAP lung lysates showed dramatically decreased levels of p53 protein (Fig. 4A). Other tissues examined also showed decreased p53 protein compared with control mice (data not shown). These changes were not due to direct transcriptional effects of Grg1 because Northern analysis revealed no alterations in p53 mRNA levels (Fig. 4B).


Figure 4
View larger version (61K):
[in this window]
[in a new window]
  		Figure 4. Altered levels of the p53 tumor suppressor and ErbB receptor tyrosine kinases in Grg-overexpressing mice. A, Western analysis of lung lysate from 1-month-old Grg1 mice. B, Northern analysis for p53 mRNA. Ribosomal bands from the ethidium bromide–stained gel are shown as a loading control. C, cotransfection of pCX-Grg1-IRES-hPLAP expression plasmid with pCX-Mdm2 cooperatively decreases p53 protein levels in NIH 3T3 cells. D, Western analysis of 3-month-old Grg1hPLAP and Grg5hPLAP mice.

 
An alternative mechanism for Grg1 reduction of p53 is through protein destabilization. Mdm2 is an E3 ubiquitin ligase that acts with HDAC1 to target p53 for proteosomal degradation. To investigate the possibility that Grg1 may collaborate with Mdm2, Grg1 and Mdm2 expression constructs were transfected into NIH3T3 cells (Fig. 4C). Cotransfection of Mdm2 and Grg1 lead to a substantial decrease in p53 protein, whereas the same amount of Mdm2 or Grg1 alone was ineffective at reducing p53.

We noted that the p53 protein levels are also reduced in Grg5-overexpressing mice (Fig. 4D). However, the Grg5 mice do not develop lung tumors; therefore, the reduction in p53 protein cannot solely account for the lung tumor phenotype in Grg1-overexpressing mice.

Increased ErbB1 and ErbB2 in the Grg1-overexpressing lung. At 1 month of age, we noted slightly increased levels of ErbB1 and ErbB2 in the lungs of Grg1-overexpressing mice (Fig. 4A). In samples collected from older mice, the increase was even more pronounced (Fig. 4D). Several observations suggest that ErbB1 and/or ErbB2 up-regulation plays a key role in Grg1-induced lung tumors. First, ErbB1 and ErbB2 were increased in the Grg1hPLAP lung but not consistently increased in other Grg1hPLAP tissues examined. In addition, ErbB1 and ErbB2 were not increased in Grg5hPLAP mice, which did not develop tumors (Fig. 4D). Finally, protein extracts from tumors consistently had even further elevated levels of ErbB2 protein compared with adjacent lung (Fig. 5A) and higher levels of phosphorylated ErbB2 and Erk1/Erk2 (p44/42) protein were observed in some tumors (Fig. 5A), suggesting enhanced activation of the ErbB signaling pathway.


Figure 5
View larger version (51K):
[in this window]
[in a new window]
  		Figure 5. Activation of ErbB2 and Erk1/2 in Grg1-induced tumors. A, Western analysis of lung and lung tumor lysates from 6-month-old Grg1hPLAP mice. Tumors have higher levels of ErbB2 compared with adjacent lung tissue. Several tumor lysates show increased phosphorylation of ErbB2 and an increase in the phosphorylation of Erk1 and Erk2 (p44/42). Levels of Ras protein are decreased specifically in tumors. B, levels of GTP-bound activated Ras (top) are reduced in tumors compared with adjacent lung and the lungs of control mice.

 
Altered Ras signaling in Grg1-induced tumors. Because the ErbB receptor tyrosine kinases act in part through activation of the Ras signaling pathway, we examined whether the Grg1hPLAP lung tumors had altered Ras status. Surprisingly, Western analysis showed decreased levels of total Ras protein in tumors relative to adjacent lung (Fig. 5A). Further measurement of total and activated GTP-bound Ras revealed that although Gg1-expressing lungs had reduced levels of total Ras, there was no change in the level of GTP-bound Ras, whereas within the tumors the lower levels of total Ras were accompanied by dramatically lowered levels of activated GTP-bound Ras (Fig. 5B). This is in stark contrast to lung tumors induced by mutant Ras in which high levels of constitutively active Ras are expressed (39).

Groucho is overexpressed in human lung tumors. We tested the possibility that Grg1/TLE1 overexpression may contribute to lung tumorigenesis in humans by screening a tumor tissue microarray for overexpression of TLE1, the human homologue of Grg1. In normal lung, the ciliated columnar epithelial cells lining bronchial mucosa showed moderate cytoplasmic and slight or moderate nuclear immunoreactivity for TLE1 (Fig. 6A) but endothelial cells predominantly showed strong positive nuclear staining (Fig. 6B). Using slight to moderate immunoreactivity of the normal bronchial epithelial cells as an internal control, nuclear staining of the tumor cells was evaluated. For each tumor sample, the scores from the multiple cores were averaged and tumors demonstrating a mean staining intensity of 3 were considered as showing TLE1 overexpression. The latter was found in 5 of 46 (11%) of the squamous cell carcinomas (Fig. 6C), 14 of 71 (20%) of the adenocarcinomas (Fig. 6D), 1 of 8 large cell undifferentiated carcinomas, and in 1 of 3 adenosquamous carcinomas. Both carcinoid tumors and SCLC showed only moderate staining. Thus TLE1 is overexpressed in a fraction of human squamous cell carcinomas and adenocarcinomas.


Figure 6
View larger version (42K):
[in this window]
[in a new window]
  		Figure 6. Analysis of TLE1 protein expression using a lung cancer tissue microarray. A representative tissue microarray consisting of 48 NSCLC samples, each represented by one to four tissue cores, was immunostained with a rabbit polyclonal antiserum to Grg1/TLE1. A, the majority of epithelial cells lining the normal bronchial mucosa show slight or no nuclear staining (open arrowhead), but occasional cells show moderate staining (closed arrowhead). B, an adenocarcinoma showing slight nuclear staining for TLE1 (open arrowhead). In contrast, endothelial cells show strong nuclear staining (arrow). C, a representative squamous cell carcinoma that shows strong nuclear immunoreactivity. D, a representative adenocarcinoma that shows strong nuclear TLE1 immunostaining.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This investigation uncovered a novel oncogenic function of the long isoform of Grg1 in the lung. Mice with Grg1 overexpression develop lung adenocarcinoma that initiates at 1 month as hyperplastic lesions resembling atypical adenomatous hyperplasia, the precursor lesion of lung adenocarcinoma in humans (7, 40). Grg5hPLAP mice do not develop tumors and coexpression of Grg5 with Grg1 is able to lessen the tumor burden induced by Grg1 overexpression. Mechanistically, this suggests that deregulated expression of long Grg isoforms contributes to lung neoplasia and confirms in an in vivo context that short isoforms can act in a dominant negative fashion to partially inhibit the tumor phenotype.

In examining the basis for the oncogenic effect of Grg1, we found that the level of the p53 tumor suppressor is dramatically decreased by Grg1 overexpression. The altered status of p53 protein in Grg1hPLAP mice may contribute to malignancy. However, a similar decrease in p53 was observed in Grg5hPLAP mice, suggesting a function shared by long and short Groucho isoforms and demonstrating that lower p53 levels are not solely responsible for the lung tumor phenotype of Grg1 mice. The decrease in p53 is likely a result of posttranslational regulation, as p53 transcript levels are not changed in Grg-overexpressing mice. The p53 protein is subject to numerous posttranslational modifications, including phosphorylation, acetylation, ubiquitination, and sumoylation (41). A possible mechanism for Grg regulation of p53 is through recruitment into HDAC1/Mdm2/p53–containing complexes. It is established that HDAC1 and Mdm2 deacetylate and ubiquitylate p53 on lysine residues, which targets p53 for degradation in the proteosome (42–44). As Groucho proteins directly bind with HDAC to deacetylate histone proteins (11, 26), Grg1 may also be part of the HDAC1 complex that deacetylates p53. This would represent a novel role for Groucho proteins in posttranslational regulation of nonhistone proteins. Grg5 would not be predicted to directly recruit HDAC activity; however, because Grg proteins function as a tetramer (22), Grg5 may form complexes with full-length Grg isoforms. Overall, the results suggest the effect of Grg overexpression on the p53 pathway is mediated by a posttranslational mechanism.

In the Grg1hPLAP lung, secondary genetic or epigenetic alterations must act coordinately with overexpressed Grg1, as the majority of Grg1-overexpressing lung epithelium cells do not become malignant. Grg1, but not Grg5, overexpression results in elevated levels of ErbB1 and ErbB2. This deregulation may predispose the Grg1 lung to tumorigenesis. Higher overall levels and phosphorylation of the ErbB2 receptor tyrosine kinase were noted in the Grg1-induced lung tumors, suggesting that ErbB2 may be a key growth regulator for these tumors. Abnormal ErbB2 expression has been reported in ~35% of human lung adenocarcinoma (45).

If ErbB receptor activation does mediate proliferation and/or survival in Grg1-induced tumors, it must be acting through pathways that do not activate Ras. Although mutation of K-ras occurs in a significant number of human lung adenocarcinomas (39), overactivation of Ras does not seem to be one of the cooperating events in Grg1-induced tumorigenesis. On the contrary, the lung tumors have decreased levels of GTP-bound Ras. Ras alleles that have been activated through mutation are often called dominant oncogenes as they are sufficient to transform transfected cells and induce tumor formation in vivo. However, evidence suggests that the dominance of activating Ras mutations depends on high levels of overexpression of mutant Ras (46) and loss of wild-type alleles. For example, heterozygosity for a null allele at the K-ras locus increases urethane-induced tumor susceptibility in the lung, whereas wild-type K-ras overexpression limits growth of lung tumor cells in nude mice (47). This suggests that wild-type K-ras has an antitumor role even in the presence of the activated allele. A decrease in wild-type Ras activation in Grg1-induced tumors may, therefore, be a contributing factor to growth of these malignancies. These mice may prove a valuable mouse model of lung tumorigenesis that is independent of hyperactive Ras signaling and the mucinous type lung adenocarcinoma.

In summary, we have shown the in vivo oncogenic potential of overexpressing a full-length Groucho protein, Grg1. Although Grg1 overexpression negatively affects the tumor suppressor p53, this alone cannot account for the tumor phenotype of Grg1, as Grg5 also decreases the level of p53 protein but is nontransforming in the mouse lung. Increased ErbB1 and ErbB2 may be key factors in Grg1-induced lung tumors. Importantly, Grg1 overexpression may contribute to a subset of human NSCLCs as it is overexpressed in 11% of squamous cell carcinomas and 20% of adenocarcinomas.


   Acknowledgments
 
Grant support: National Cancer Institute of Canada and Cancer Research Society.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Dr. Stefano Stifani for the {alpha}-TLE1 antibody; Dr. Jean Chamberlain for assistance in typing and subtyping of the murine lung tumors; Wahyuni Atmodjo for technical assistance with immunohistochemistry; Dr. Hao Ding for assistance with the Ras assays; Drs. Alex Swarbrick, Alana Welm, and Sue Kim for their critical reading of the manuscript; and Dr. Andras Nagy for his input on the Cre-conditional transgene system.


   Footnotes
 
9 C.G. Lobe, unpublished data. Back

Received 5/12/05. Revised 11/ 6/05. Accepted 11/16/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Rom WN, Hay JG, Lee TC, et al. Molecular and genetic aspects of lung cancer. Am J Respir Crit Care Med 2000;161:1355–67.[Free Full Text]
  2. Forgacs E, Zochbauer-Muller S, Olah E, et al. Molecular genetic abnormalities in the pathogenesis of human lung cancer. Pathol Oncol Res 2001;7:6–13.[Medline]
  3. Fisher GH, Wellen SL, Klimstra D, et al. Induction and apoptotic regression of lung adenocarcinomas by regulation of a K-Ras transgene in the presence and absence of tumor suppressor genes. Genes Dev 2001;15:3249–62.[Abstract/Free Full Text]
  4. Jackson EL, Willis N, Mercer K, et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev 2001;15:3243–8.[Abstract/Free Full Text]
  5. Johnson L, Mercer K, Greenbaum D, et al. Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature 2001;410:1111–6.[CrossRef][Medline]
  6. Meuwissen R, Linn SC, van der Valk M, et al. Mouse model for lung tumorigenesis through Cre/lox controlled sporadic activation of the K-Ras oncogene. Oncogene 2001;20:6551–8.[CrossRef][Medline]
  7. Yamasaki M, Takeshima Y, Fujii S, et al. Correlation between genetic alterations and histopathological subtypes in bronchiolo-alveolar carcinoma and atypical adenomatous hyperplasia of the lung. Pathol Int 2000;50:778–85.[CrossRef][Medline]
  8. Lobe CG, Koop KE, Kreppner W, et al. Z/AP, a double reporter for cre-mediated recombination. Dev Biol 1999;208:281–92.[CrossRef][Medline]
  9. Parkhurst S. Groucho: making its Marx as a trascriptional co-repressor. Trends Genet 1998;14:130–2.[CrossRef][Medline]
  10. Chen G, Courey AJ. Groucho/TLE family proteins and transcriptional repression. Gene 2000;249:1–16.[CrossRef][Medline]
  11. Chen G, Fernandez J, Mische S, et al. A functional interaction between the histone deacetylase Rpd3 and the corepressor Groucho in Drosophila development. Genes Dev 1999;13:2218–30.[Abstract/Free Full Text]
  12. Courey AJ, Jia S. Transcriptional repression: the long and the short of it. Genes Dev 2001;15:2786–96.[Free Full Text]
  13. Fisher A, Caudy M. Groucho proteins: transcriptional corepressors for specific subsets of DNA-binding transcription factors in vertebrates and invertebrates. Genes Dev 1998;12:1931–40.[Free Full Text]
  14. Dang J, Inukai T, Kurosawa H, et al. The E2A-HLF oncoprotein activates Groucho-related genes and suppresses Runx1. Mol Cell Biol 2001;21:5935–45.[Abstract/Free Full Text]
  15. Hartley DA, Preiss A, Artavanis-Tsakonas S. A deduced gene product from the Drosophila neurogenic locus, enhancer of split, shows homology to mammalian G-protein ß subunit. Cell 1988;55:785–95.[CrossRef][Medline]
  16. Koop KE, MacDonald LM, Lobe CG. Transcripts of Grg4, a murine Groucho-related gene, are detected in adjacent tissues to other murine neurogenic gene homologues during embryonic development. Mech Dev 1996;59:73–87.[CrossRef][Medline]
  17. Leon C, Lobe CG. Grg3, a murine Groucho-related gene, is expressed in the developing nervous system and in mesenchyme-induced epithelial structures. Dev Dyn 1997;208:11–24.[CrossRef][Medline]
  18. Mallo M, Franco Del Amo F, Gridley T. Cloning and developmental expression of Grg, a mouse gene related to the Groucho transcript of the Drosophila enhancer of split complex. Mech Dev 1993;42:67–76.[CrossRef][Medline]
  19. Miyasaka H, Choudhury BK, Hou EW, et al. Molecular cloning and expression of mouse and human cDNA encoding AES and ESG proteins with strong similarity to Drosophila enhancer of split Groucho protein. Eur J Biochem 1993;216:343–52.[Medline]
  20. Stifani S, Blaumueller CM, Redhead NJ, et al. Human homologs of a Drosophila enhancer of split gene product define a novel family of nuclear proteins. Nat Genet 1992;2:119–26.[CrossRef][Medline]
  21. Lepourcelet M, Shivdasani RA. Characterization of a novel mammalian Groucho isoform and its role in transcriptional regulation. J Biol Chem 2002;277:47732–40.[Abstract/Free Full Text]
  22. Chen G, Nguyen P, Courey A. A role for Groucho tetramerization in transcriptional repression. Mol Cell Biol 1998;18:7259–68.[Abstract/Free Full Text]
  23. Pinto M, Lobe CG. Products of the Grg (Groucho-related gene) family can dimerize through the amino-terminal Q domain. J Biol Chem 1996;271:33026–31.[Abstract/Free Full Text]
  24. Ren B, Chee KJ, Kim TH, et al. PRDI-BF1/Blimp-1 repression is mediated by corepressors of the Groucho family of proteins. Genes Dev 1999;13:125–37.[Abstract/Free Full Text]
  25. Roose J, Molenaar M, Peterson J, et al. The Xenopus Wnt effector XTcf-3 interacts with Groucho-related transcriptional repressors. Nature 1998;395:608–12.[CrossRef][Medline]
  26. Brantjes H, Roose J, van De Wetering M, et al. All Tcf HMG box transcription factors interact with Groucho-related co-repressors. Nucleic Acids Res 2001;29:1410–9.[Abstract/Free Full Text]
  27. Wang JC, Waltner-Law M, Yamada K, et al. Transducin-like enhancer of split proteins, the human homologs of Drosophila Groucho, interact with hepatic nuclear factor 3ß. J Biol Chem 2000;275:18418–23.[Abstract/Free Full Text]
  28. Tetsuka T, Uranishi H, Imai H, et al. Inhibition of nuclear factor-{kappa}B-mediated transcription by association with the amino-terminal enhancer of split, a Groucho-related protein lacking WD40 repeats. J Biol Chem 2000;275:4383–90.[Abstract/Free Full Text]
  29. Yu X, Li P, Roeder RG, et al. Inhibition of androgen receptor-mediated transcription by amino-terminal enhancer of split. Mol Cell Biol 2001;21:4614–25.[Abstract/Free Full Text]
  30. Wurst W, Joyner A. Production of targeted embryonic stem cell clones. In: Joyner A, editor. Gene targeting: a practical approach. Oxford: IRL Press; 1993. p. 33–62.
  31. Guo C, Yang W, Lobe CG. A Cre recombinase transgene with mosaic, widespread tamoxifen-inducible action. Genesis 2002;32:8–18.[CrossRef][Medline]
  32. Novak A, Guo C, Yang W, et al. Z/EG, a double reporter transgenic mouse line that expresses enhanced green fluorescent protein upon Cre-mediated recombination. Genesis 2000;28:147–55.[CrossRef][Medline]
  33. Nagy A. Cre recombinase: the universal reagent for genome tailoring. Genesis 2000;26:99–109.[CrossRef][Medline]
  34. Moorman AF, Houweling AC, de Boer PA, et al. Sensitive nonradioactive detection of mRNA in tissue sections: novel application of the whole-mount in situ hybridization protocol. J Histochem Cytochem 2001;49:1–8.[Abstract/Free Full Text]
  35. Husain J, Lo R, Grbavec D, et al. Affinity for the nuclear compartment and expression during cell differentiation implicate phosphorylated Groucho/TLE1 forms of higher molecular mass in nuclear functions. Biochem J 1996;317:523–31.[Medline]
  36. Niwa H, Yamamura K, Miyazaki J. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 1991;108:193–9.[CrossRef][Medline]
  37. Goldstein NS, Thomas M. Mucinous and nonmucinous bronchioloalveolar adenocarcinomas have distinct staining patterns with thyroid transcription factor and cytokeratin 20 antibodies. Am J Clin Pathol 2001;116:319–25.[Abstract/Free Full Text]
  38. Lau SK, Desrochers MJ, Luthringer DJ. Expression of thyroid transcription factor-1, cytokeratin 7, and cytokeratin 20 in bronchioloalveolar carcinomas: an immunohistochemical evaluation of 67 cases. Mod Pathol 2002;15:538–42.[CrossRef][Medline]
  39. Mills NE, Fishman CL, Rom WN, et al. Increased prevalence of K-ras oncogene mutations in lung adenocarcinoma. Cancer Res 1995;55:1444–7.[Abstract/Free Full Text]
  40. Mori M, Rao SK, Popper HH, et al. Atypical adenomatous hyperplasia of the lung: a probable forerunner in the development of adenocarcinoma of the lung. Mod Pathol 2001;14:72–84.[CrossRef][Medline]
  41. Bode AM, Dong Z. Post-translational modification of p53 in tumorigenesis. Nat Rev Cancer 2004;4:793–805.[CrossRef][Medline]
  42. Brooks CL, Gu W. Ubiquitination, phosphorylation and acetylation: the molecular basis for p53 regulation. Curr Opin Cell Biol 2003;15:164–71.[CrossRef][Medline]
  43. Ito A, Kawaguchi Y, Lai CH, et al. MDM2-HDAC1-mediated deacetylation of p53 is required for its degradation. EMBO J 2002;21:6236–45.[CrossRef][Medline]
  44. Nakamura S, Roth JA, Mukhopadhyay T. Multiple lysine mutations in the C-terminal domain of p53 interfere with MDM2-dependent protein degradation and ubiquitination. Mol Cell Biol 2000;20:9391–8.[Abstract/Free Full Text]
  45. Hirsch FR, Varella-Garcia M, Franklin WA, et al. Evaluation of HER-2/neu gene amplification and protein expression in non-small cell lung carcinomas. Br J Cancer 2002;86:1449–56.[CrossRef][Medline]
  46. Hua VY, Wang WK, Duesberg PH. Dominant transformation by mutated human ras genes in vitro requires more than 100 times higher expression than is observed in cancers. Proc Natl Acad Sci U S A 1997;94:9614–9.[Abstract/Free Full Text]
  47. Zhang Z, Wang Y, Vikis HG, et al. Wildtype Kras2 can inhibit lung carcinogenesis in mice. Nat Genet 2001;29:25–33.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 BloodHome page
F. Dayyani, J. Wang, J.-R. J. Yeh, E.-Y. Ahn, E. Tobey, D.-E. Zhang, I. D. Bernstein, R. T. Peterson, and D. A. Sweetser
Loss of TLE1 and TLE4 from the del(9q) commonly deleted region in AML cooperates with AML1-ETO to affect myeloid cell proliferation and survival
Blood, April 15, 2008; 111(8): 4338 - 4347.
[Abstract] [Full Text] [PDF]

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