Skip Navigation


Carcinogenesis Advance Access originally published online on August 8, 2007
Carcinogenesis 2007 28(12):2589-2596; doi:10.1093/carcin/bgm136
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/12/2589    most recent
bgm136v1
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 ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Pandey, J.
Right arrow Articles by Jakowlew, S. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pandey, J.
Right arrow Articles by Jakowlew, S. B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published by Oxford University Press 2007.

Modulation of tumor induction and progression of oncogenic K-ras-positive tumors in the presence of TGF-β1 haploinsufficiency

Jyotsna Pandey, Sarah M. Umphress, Yang Kang, Jerry Angdisen, Alena Naumova, Kim L. Mercer1, Tyler Jacks1 and Sonia B. Jakowlew*

Cell and Cancer Biology Branch, National Cancer Institute, 9610 Medical Center Drive, Suite 300, Rockville, MD 20850, USA
1 Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

* To whom correspondence should be addressed. Tel: +301 594 7280; Fax: +301 402 4472;Email: jakowles{at}mail.nih.gov


   Abstract
Top
 Abstract
Introduction
 Material and methods
 Results
 Discussion
 Funding
 References
 
Oncogenic K-ras is one of the most common genetic alterations in human lung adenocarcinomas. In addition, inactivation of clusters of tumor suppressor genes is required to bring about classical characteristics of cancer including angiogenesis as a prelude to invasion and metastasis. Transforming growth factor-β (TGF-β) 1 is a tumor suppressor gene that is implicated in lung cancer progression. Although in vitro studies have shown that TGF-β1 and Ras pathways cooperate during tumorigenesis, the biology of interaction of TGF-β1 and Ras has not been studied in in vivo tumorigenesis. We hypothesized that inactivation of TGF-β1 in addition to oncogeneic activation of K-ras would lead to early initiation and faster progression to lung adenocarcinoma and invasion and metastasis. Heterozygous (HT) TGF-β1 mice were mated with latent activatable (LA) mutated K-ras mice to generate TGF-β1+/+, K-ras LA (wild-type (WT)/LA) and TGF-β1+/–, K-ras LA (HT/LA) mice. Both HT/LA and WT/LA mice developed spontaneous lung tumors, but HT/LA mice progressed to adenocarcinomas significantly earlier compared with WT/LA mice. In addition, WT/LA adenocarcinomas had significantly higher angiogenic activity compared with HT/LA adenocarcinomas. Thus, while oncogenic K-ras mutation and insensitivity to the growth regulatory effects of TGF-β1 is essential for initiation and progression of mouse lung tumors to adenocarcinoma, a full gene dosage of TGF-β1 is required for tumor-induced angiogenesis and invasive potential. This study identifies a number of genes not previously associated with lung cancer that are involved in tumor induction and progression. In addition, we provide evidence that progression to invasive angiogenic lesions requires TGF-β1 responsiveness in addition to Ras mutation.

Abbreviations: Erk, extracellular signal-regulated kinase; HT, heterozygous; LA, latent activatable; MMP, matrix metalloproteinase; PCR, polymerase chain reaction; TGF-β, transforming growth factor-β; TGF-β RII, transforming growth factor-β type II receptor; WT, wild-type


   Introduction
Top
 Abstract
 Introduction
Material and methods
 Results
 Discussion
 Funding
 References
 
Lung cancer is a leading cause of cancer-related death throughout the western world (1,2). The genetic changes in lung tumor tissue reflect mainly activation of a number of oncogenes and inactivation of clusters of tumor suppressor genes (1,3,4) that together bring about classical characteristics of cancer. These include sustained angiogenesis as a prelude to invasion and metastasis (5). Identification of molecular footprints during cancer progression and correlation with particular stages of tumor development would improve our ability to find novel ways of controlling tumor invasion and metastasis (6). Thus, understanding the molecular biology and various interactions will help to develop targeted and less toxic therapies.

Investigator-initiated gene-targeted mouse models that develop lung cancer spontaneously to study aberrations in intracellular network signaling molecules of tumor tissues are a valuable tool in understanding tumor progression (7). K-ras activation is an early and tumor-initiating event in lung tumorigenesis (8–12). Mutations at codon 12 of K-ras account for 85% of all the K-ras mutations in human lung tumors (1) and in lung adenocarcinomas of mice (12,13). In order to study this further, mouse strains carrying oncogenic alleles of K-ras that can be activated by a spontaneous recombination event in the whole animal were generated. These mice are highly predisposed to early onset lung lesions (14). As mutation of K-ras is only one of a number of mutations necessary for progression of tumors to malignancy (2), it is important to identify other genes that K-ras may interact with to bring about enhanced tumor invasion and metastasis.

Expression of oncogenic forms of Ras in normal cells is known to inhibit proliferation by activating tumor suppressor activity as a response to inappropriate proliferation signals (15,16). One of the tumor suppressor genes implicated in the pathogenesis of lung cancer is transforming growth factor-β (TGF-β) and its type II receptor (TGF-β RII) (17–19). The growth-inhibitory effect of TGF-β1 in epithelial cells is associated with a direct and rapid stimulation of guanosine triphosphate-bound ras p21 that is active in signal transduction (20). Oncogenic Ras inhibits TGF-β1 signaling by down-regulating TGF-β RII and modulating Smad-dependent transcription (21,22) and growth-inhibitory response to TGF-β1 (23–25). In vitro studies have shown that TGF-β1 can induce epithelial to mesenchymal transdifferentiation that may lead to invasion. The process requires cooperation between Ras–mitogen-activated protein kinase and the TGF-β-signaling pathway (26,27). The interaction of TGF-β1 and Ras mutation has not been studied in vivo.

We previously examined the role of TGF-β in mouse lung tumorigenesis and reported reduced expression of TGF-β RII in lung tumors (28,29) and that TGF-β1 shows true haploid insufficiency in its ability to protect against chemically induced carcinogenesis (30). TGF-β1 null mice have considerable uterine lethality by 3 weeks after birth (31,32). In contrast, TGF-β1 heterozygous (HT) mice have a phenotype indistinguishable from their wild-type (WT) littermates. They are susceptible to chemically induced carcinogenesis, but do not develop spontaneous tumors. Hence, TGF-β1 HT mice are more suitable for studying investigator-initiated compound (mutation of an oncogene and a tumor suppressor gene) tumor biology.

To investigate the tumor biology resulting from interaction of TGF-β1 and K-ras in vivo, we mated TGF-β1 HT mice with mice harboring latent activatable (LA) oncogenic alleles of K-ras and generated TGF-β1 HT, K-ras LA compound mice. We hypothesized that the heterozygosity of the TGF-β1 gene would provide a growth advantage to a neoplastic cell population harboring spontaneous K-ras mutation and, thus, would lead to early progression and invasion. We tested this hypothesis by comparing latency, incidence, multiplicity and load and invasive potential (angiogenic index) of lung lesions of TGF-β1 HT/K-ras LA (HT/LA) compound mice with TGF-β1 wild-type/K-ras LA (WT/LA) mice. We find that TGF-β1 heterozygosity causes an early progression to a malignant phenotype. In addition, contrary to expectation, we find that a functional TGF-β pathway in addition to K-ras mutation is required for developing an invasive (angiogenic tumor) phenotype.


   Material and methods
Top
 Abstract
 Introduction
 Material and methods
Results
 Discussion
 Funding
 References
 
Generation of TGF-β1+/–, K-ras LA mice: novel congenic mouse model of lung cancer
To determine whether the loss of one TGF-β1 allele combined with mutational activation of K-ras has any oncogenic effect over and above the oncogenic effect of K-ras mutation alone, we generated mouse genotypes that included TGF-β1+/+, K-ras LA (WT/LA) and TGF-β1+/–, K-ras LA (HT/LA), respectively. C57BL/6NCr mice that have HT deletion of a TGF-β1 allele (TGF-β1+/–) (30,32) were mated with C57BL/6/129/sv mice with a LA K-ras mutation (14). Confirmation of genotypes was done using tail sample DNA and polymerase chain reaction (PCR) amplification with specific oligonucleotide primers for wild-type and mutated alleles of both TGF-β1 and K-ras. All mice were maintained in a pathogen-free barrier facility and were treated in accordance with the guidelines for animal care and use established by the National Institutes of Health.

In order to determine whether there was a significant difference in morbidity and mortality, wild-type and congenic mice were allowed to live until their health was compromised or natural death. Additional wild-type and congenic mice were killed at monthly intervals between 1 and 4 months of age using CO2 inhalation, and the lungs were quickly removed. Tumor number and size were determined, and lungs were fixed in 10% neutral buffered formalin or 70% ethanol at 4°C for 20 h, dehydrated, embedded in paraffin for immunohistochemical staining or put on ice for microdissection and isolation of lung lesions, which were then stored in RNALater at –20°C (Qiagen, Germantown, MD) or flash frozen and stored at –80°C until used.

Western blot and immunohistochemistry and gelatin zymogram analysis
Total protein was extracted from frozen tissues (lung lesions and wild-type lungs) in modified RadioImmunoPrecipitation Assay buffer. Antibodies used included pErk1/2 (E-4, sc-7383, Santa Cruz Biotechnology, Santa Cruz, CA), Raf-1 (C-12, sc-133, Santa Cruz Biotechnology), K-ras (F234, sc-30, Santa Cruz Biotechnology), Smad3 (51-1500, Zymed Laboratories, San Francisco, CA), Smad4 (B-8, sc-7966, Santa Cruz Biotechnology), Smad7 (H-79, sc-11392, Santa Cruz Biotechnology), TGF-β RII (L-21, sc-400, Santa Cruz Biotechnology) and TGF-β1 (G1221, Promega, Madison, WI), Bax (N-20, sc-493, Santa Cruz Biotechnology), Bcl2 (N-19, sc-492, Santa Cruz Biotechnology) and horseradish peroxidase-conjugated secondary antibody (Amersham, Life Science, Arlington Heights, IL). Densitometric analysis of protein bands was performed and the same membrane was probed with antibodies for actin (C-11, sc-1615, Santa Cruz Biotechnology) after using stripping buffer (Bio-Rad, Hercules, CA). Immunohistochemistry was done on paraffin sections of mouse lungs using the horseradish peroxidase–avidin–biotin complex technique (Vector Laboratories, Burlingame, CA). Gelatin zymograms were done using an Invitrogen (Carlsbad, CA) kit and reagents according to the manufacturer's protocol.

Assesment of apoptotic activity with tissue terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling assay
Apoptotic cells were visualized by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling using a TACS-XL DAB kit (R&D Systems, Minneapolis, MN). Counterstaining was performed using methyl green nuclear stain. The apoptotic index (number of apoptotic cells per 100 cells) was determined by microscopic examination of randomly selected microscopic fields at x40 magnification. At least 500 cells or seven microscopic fields were counted for every lesion. A minimum of eight lesions was counted for each type of pathology.

Quantitative real-time reverse transcription–PCR analysis
Frozen tissues (lung lesions and wild-type lungs) stored in RNALater were weighed and total RNA was isolated using a RNeasy Midi kit (Qiagen). Two microliter of total RNA was converted to first-strand cDNA in a total reaction of 50 µl using oligo(dT)20 primers with a two-step reverse transcription–PCR SYBR green kit (Qiagen) and real-time reverse transcription–PCR was performed using 40 amplification cycles of 94°C for 30 s, 55°C for 30 s and 72°C for 45 s in a Bio-Rad iCycler. Glyceraldehyde-6-phosphate dehydrogenase was co-amplified as an internal control. Single amplification band was confirmed by melt curve analysis at the end of amplification cycles.

cDNA and oligonucleotide array labeling and hybridization
Gene expression was analyzed using gene expression cDNA microarrays for profiling genes involved in angiogenesis (#MM-009), mitogen-activated protein kinase (#MM-017) and TGF-β (#MM-023) pathways from SuperArray Biosciences (Walkersville, MD). Oligonucleotide gene expression arrays were used for apoptosis (#OMM-012) and cell cycle (#OMM-020) genes from SuperArray Biosciences. The RNA for oligonucleotide arrays was labeled using biotin-16-uridine triphosphate (#11388908910; Roche Applied Science, Indianapolis, IN) and TrueLabeling-AMP 2.0 Kit (SuperArray Biosciences). RNA for cDNA arrays was labeled using a biotin-16-deoxyuridine triphosphate (#1093070; Roche Applied Biosciences) and Ampolabeling-LPR Kit (SuperArray Biosciences). Each membrane was hybridized overnight using a GEAhyb hybridization kit (SuperArray Biosciences). A horseradish peroxidase enzyme/substrate reaction kit was used for chemiluminescent detection (SuperArray Biosciences), and a digital 16 bit tiff image was generated and stored for analysis. Analysis was performed using GEarray expression analysis suite software (SuperArray Biosciences). The primers for confirmation of expression of differentially expressed genes were obtained pre-manufactured from SuperArray Biosciences and are listed in Tables I and II.


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

  		Table I. Differential gene expression patterns confirmed by quantitative reverse transcription–PCR for benign lung lesions of TGF-β1/K-ras mice

 


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

  		Table II. Differential gene expression patterns confirmed by quantitative reverse transcription–PCR for malignant lung lesions of TGF-β1/K-ras mice

 
Analysis of microvessel density to determine the angiogenesis index
Tumor vascularity was measured in order to provide a histological assessment of tumor angiogenesis. Microvessels were counted in 10 adenocarcinomas of HT/LA and WT/LA mice as described previously (33), according to criteria established by international consensus protocol (34). Briefly, microvessel count was calculated manually by selecting two areas of highest microvessel density containing the highest number of capillaries (microvessels) per area (i.e. areas of intense vascularity ‘hot spot’). A single countable microvessel was defined as any endothelial cells or group of cells that was clearly separate from other vessels and surrounded by tumor tissue by identifying Griffonia simplicifolia lectin positivity, without the necessity of having a vessel lumen or red blood cell within the lumen. Individual microvesels were counted at x200 (area: 0.7386 mm2 per field), within each hot spot identified. Data were averaged to give the microvessel count for each sample.

Statistics
Statistical analysis of data was performed using SPSS13.0 software. The t-test or Mann–Whitney Rank sum test was used for analysis and a P value of 0.05 or less was considered to be significant.


   Results
Top
 Abstract
 Introduction
 Material and methods
 Results
Discussion
 Funding
 References
 
Mouse phenotype, survival and tumor biology
The phenotypes of HT/LA and WT/LA mice were indistinguishable from WT/WT mice at birth. However, the survival (life span) of congenic mice with K-ras mutation was significantly reduced (P < 0.0001) compared with WT/WT mice (Figure 1A). All mice harboring a LA mutation of K-ras (HT/LA and WT/LA mice) had multiple lung tumors ranging in size from 1 to 5 mm in diameter (Figure 1B). The HT/WT and WT/WT mice did not develop lung lesions (Figure 1B). Total tumor load was significantly higher in HT/LA mice at 1 month of age only (Figure 2F). Tumors were located mainly on the surface of the lung, with very few lesions in the central region of the lung. The tumors were classified according to criteria of the Mouse Models of Human Cancer Consortium (35). Lesions ranged from alveolar hyperplasia, adenoma, atypical adenoma to adenocarcinoma. For simplicity, atypical adenomas were classified with adenocarcinomas as invasive lesions. Lung tumor cells were determined to be positive for surfactant protein-C and negative for Clara cell-specific protein-10 by immunohistochemistry (data not shown) and confirm results reported in other animal models expressing K-ras oncogenes (14,36). Tumors at 1 month of age were mostly benign lesions consisting of adenomas and hyperplasias in both HT/LA and WT/LA mice (Figure 2A and B). However, HT/LA mouse tumors occasionally had foci of adenocarcinoma within adenomatous lesions by 1 month of age. While WT/LA mouse lungs had a linear increase in adenomas over time (Figure 2B), the lungs of HT/LA mice exhibited a similar linear progression in the number of adenocarcinomas over the same time period (Figure 2C). Hence, there is an exaggerated conversion and progression to the malignant phenotype in HT/LA mice compared with WT/LA mice (Figure 2D and E). A switch to predominantly malignant tumor load occurs between the ages of 1–2 months in HT/LA mice and 2–3 months in WT/LA mice (Figure 2C). The earlier progression of HT/LA tumors to a malignant phenotype suggests a more profound accumulation of mutations resulting from K-ras mutation-induced proliferation. In order to see if TGF-β1 heterozygosity further added to the process by lack of apoptosis and preservation of clones with accumulated mutations, we looked at the apoptosis index in tumors of both WT/LA and HT/LA mice.


Figure 1
View larger version (53K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. (A) Mice with a LA K-ras mutation have a significantly reduced life span compared with their wild-type littermates. Kaplan–Meier survival curve of TGF-β1/K-ras mice shows that TGF-β1 HT/K-ras LA (HT/LA) F1 mice (median survival age of 65 days for n = 25 mice) and TGF-β1 WT/K-ras LA (WT/LA) mice (median survival age of 120 days for n = 24 mice) have significantly decreased survival (P < 0.0001) compared with both TGF-β1 HT/K-ras WT (HT/WT) mice (median survival age of 350 days for n = 25 mice) and wild-type TGF-β1 WT/K-ras WT (WT/WT) mice (median survival age of 381 days for n = 20 mice). Mice were followed for their natural life span till death or euthanized if their health was severely compromised. (B) Mice harboring a LA K-ras mutation develop multiple spontaneous lung tumors. TGF-β1 HT/K-ras LA (HT/LA) and TGF-β1 WT/K-ras LA (WT/LA) mouse lungs show multiple tumors in various stages of development on the surface of lungs at 4 months of age. TGF-β1 HT/K-ras WT (HT/WT) and wild-type TGF-β1 WT/K-ras WT (WT/WT) mice do not develop spontaneous tumors.

 


Figure 2
View larger version (14K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Lung lesion histopathology, incidence and multiplicity in TGF-β1/K-ras mice show that HT/LA mouse tumors progress to adenocarcinoma earlier and have a significantly higher adenoma to adenocarcinoma progression frequency due to effective abrogation of apoptosis compared with WT/LA tumors. The average numbers of lung (A) hyperplasias, (B) adenomas and (C) adenocarcinomas in TGF-β1 WT/K-ras LA (WT/LA) and TGF-β1 HT/K-ras LA (HT/LA) mice were counted in lungs at 1, 2, 3 and 4 months of age. (A) The average numbers of hyperplasias are similar for all time points in HT/LA mouse lungs, but there is a significant increase in the number of hyperplasia lesions at 3 and 4 months. (B) The lung adenomas show statistically significant and linear increases over 1–4 months in WT/LA lungs (P < 0.001). The average numbers of adenomas are few and unchanged in HT/LA lungs in contrast. (C) A linear increase in the average number of adenocarcinomas is seen in HT/LA lungs over the 1- to 4-month period. The mean numbers of adenocarcinomas are significantly higher in HT/LA lungs at 1 and 2 months of age compared with WT/LA lungs (P < 0.001). Taken together, this implies that there is faster and earlier progression to adenocarcinomas in HT/LA lungs. This is further confirmed by tumor progression frequencies for the two genotypes. There is no significant difference between hyperplasia to adenoma conversion in the two genotypes (D), but there is significantly higher progression of adenomas to adenocarcinomas in HT/LA mouse lungs (E). (F) The total tumor load is significantly higher for HT/LA mice at 1 month of age only. The tumor progression frequency was determined by dividing the total number of adenomas at each time point by the total numbers of hyperplasia for hyperplasia to adenoma conversion and the total number of adenocarcinomas was divided by adenomas for adenoma to adenocarcinoma conversion. All data are the means ± SEMs, with 10–12 mice at each age point. P value < 0.01. White bars: TGF-β1 WT/K-ras LA; Black bars: TGF-β1 HT/K-ras LA.

 
Apoptosis abrogation in lesions
We reported earlier that apoptosis is significantly decreased in the bronchioalvelolar epithelium of mice that are HT for TGF-β1 (30). This is consistent with a role for endogenous TGF-β1 in regulating apoptosis in the lung. There was a significant decrease in apoptosis index in lung adenomas of HT/LA mice compared with WT/LA mice (Figure 3A), although the apoptosis indices in hyperplasias of the two genotypes were similar. Bax and Bcl2 expression patterns (Figure 3B) are in concordance with the apoptotic indices, indicating that abrogation of apoptosis may be an important factor in transition to the malignant phenotype. There was an almost complete absence of apoptosis in adenocarcinomas of both HT/LA and WT/LA mice.


Figure 3
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Suppression of apoptosis is more pronounced in HT/LA mouse tumors. (A) The tumor apoptosis index is significantly (P < 0.001) abrogated in adenomas of HT/LA mice, suggesting that this may be an important factor resulting in earlier and faster progression to adenocarcinoma in HT/LA mouse lungs. (B) There is down-regulation of Bax levels with attendant sustained Bcl2 expression starting at 2 months of age in HT/LA mice and 3 months of age in WT/LA mice. This suggests that apoptosis may be an important factor in tumor progression and that a Bcl2-mediated mechanism may be involved in the suppression of the apoptotic cell death in tumor cells. All data are the means ± SDs. P value < 0.01. White bars: TGF-β1 WT/K-ras LA; black bars: TGF-β1 HT/K-ras LA.

 
Ras/Raf/extracellular signal-regulated kinase pathway
Ras transmits extracellular signals via sequential phosphorylation of pivotal kinases including Raf and extracellular signal-regulated kinase (Erk) 1/2, which culminates in Ras-mediated cell transformation, proliferation and cell survival (26). Western blot analysis shows that expression of K-ras increased dramatically in lung lesions of HT/LA and WT/LA mice as lung tumor load became prominent (Figure 4A). A second band recognized by the K-ras antibody appeared at the point of switch to malignant phenotype (i.e. between 1–2 month lesions of HT/LA mice and 2–3 month lesions of WT/LA mice). Raf-1 and Erk1/2 expression were not altered significantly in tumors, but showed a trend for increase similar to that of K-ras (Figure 4A).


Figure 4
View larger version (62K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Expression patterns of proteins involved in the Ras-mitogen-activated protein kinase and TGF-β1 pathways in lung tumors. (A) Pooled protein from five mouse lung tumors of TGF-β1/K-ras mice was analyzed on the same gel for comparison purposes. K-ras expression increases with tumor load increase (see text), whereas Raf-1 and Erk1/2 are not differentially regulated between TGF-β1 WT/K-ras LA (WT/LA) and TGF-β1 HT/K-ras LA (HT/LA) tumors. Total TGF-β RII is down-regulated with progression of tumors in both genotypes, whereas pSmad2 is not differentially regulated with tumor development. Smad3 is down-regulated in early tumorigenesis, but is up-regulated with tumor progression. Smad4 is up-regulated in TGF-β1 WT/K-ras LA (WT/LA) mice in late-stage tumors with a predominant malignant phenotype, and c-myc expression inversely correlates with Smad3 and Smad4 expression patterns in TGF-β1 WT/K-ras LA (WT/LA) tumors. The classical TGF-β pathway expression patterns suggest that while tumors of both TGF-β1 HT/K-ras LA (HT/LA) and TGF-β1 WT/K-ras LA (WT/LA) mice show down-regulation of TGF-β1 responsiveness in early lesions, only TGF-β1 HT/K-ras LA (HT/LA) mouse tumors show this resistance in later predominantly malignant lesions. (B) Immunohistochemical staining shows patchy loss and down-regulation of TGF-β1 and TGF-β RII in WT/LA adenomas, whereas HT/LA adenomas have a homogenous down-regulation. Adenocarcinoma lesions also show patchy and partial loss of TGF-β1 and TGF-β RII in WT/LA tumors and there is a homogenous loss of positive expression in HT/LA tumors. The top panel shows hematoxylin and eosin staining for adenoma and adenocarcinoma in HT/LA mouse lung. The cellular morphology is similar in WT/LA tumors (data not shown). (C) Quantitative real-time reverse transcription–PCR amplification analysis of TGF-β pathway components in lung lesions of TGF-β1/K-ras mice shows that RNA levels correlate with protein levels. Total RNA from lung lesions of 1 month old (adenomas) and 4 month old (adenocarcinomas) TGF-β1 WT/K-ras LA (WT/LA) and TGF-β1 HT/K-ras LA (HT/LA) mice (n = 4 each) was converted to cDNA and used in real-time reverse transcription–PCR amplification for TGF-β RII, Smad2, Smad3, Smad4 and Smad7 messenger ribonucleic acids. The values were normalized to control samples from wild-type littermate lungs. All data are the means ± SDs.

 
TGF-β-signaling pathway
Cells harboring oncogenic activation of Ras have been shown to exhibit loss of the anti-proliferative response to TGF-β1 and leads to tumor progression (24,25). There was a progressive reduction in the expression pattern of both TGF-β1 and TGF-β RII proteins in lesions ranging from hyperplasia and adenoma to adenocarcinoma in WT/LA and HT/LA mice as seen by immunohistochemistry (Figure 4B). Significantly reduced levels of both TGF-β1 and TGF-β RII proteins were detected in adenocarcinomas compared with hyperplasias and adenomas in HT/LA and WT/LA mice. However, loss of TGF-β RII was generally patchy in tumors of WT/LA mice, suggesting at least partial TGF-β responsiveness in these animals (Figure 4B). Western blot analysis for TGF-β RII expression shows that TGF-β RII decreased progressively as tumors progressed in both HT/LA and WT/LA animals (Figure 4A). Transforming growth factor-β type I receptor expression was similar for all tumors (data not shown).

The pattern of expression of pSmad2 was similar to that of TGF-β RII and decreased as tumors progressed in HT/LA and WT/LA mice, whereas Smad3 expression increased in 2- and 3-month old mouse lung lesions of HT/LA and WT/LA mice, respectively, corresponding to the appearance of adenocarcinomas (Figure 4A and C). Smad4 had a differential expression pattern, increasing dramatically when the lesion load shifted from hyperplasia and adenoma (benign) to adenocarcinoma (malignant) in WT/LA mice at 3 and 4 months of age. This suggests restoration of TGF-β responsiveness in WT/LA malignant tumors. Surprisingly, expression of Smad7 was found to be below the level of detection in tumor lysates of HT/LA and WT/LA mice. The RNA levels correlated with protein expression (Figure 4C). There was down-regulation of Smad2, Smad3, Smad4 and Smad7 transcripts in benign tumors of both HT/LA and WT/LA mice, whereas malignant tumors of WT/LA mice showed a reversal of this pattern (Figure 4C). Earlier reports have shown that c-myc repression is required for TGF-β1-induced growth arrest and that oncogenic K-ras leads to loss of TGF-β1 responsiveness and increased c-myc expression (37). c-myc expression levels were high at early stages of tumorigenesis in benign tumors of both HT/LA and WT/LA mice and decreased in invasive lesions so that c-myc expression was barely detectable in lung lesions of 3- and 4-month old HT/LA and WT/LA mice, respectively (Figure 4A).

Adenocarcinoma tumors that have wild-type TGF-β1 (WT/LA mice) showed increases in Smad2, Smad3 and Smad4 and suppression of c-myc (Figure 4A), suggesting some amount of TGF-β1-responsive activity. In contrast, adenocarcinoma lesions of mice that are HT for TGF-β1 (HT/LA) showed persistent abrogation of TGF-β signaling as evidenced by decreased expression of TGF-β RII and Smad4 (Figure 4). In order to delineate how this differential abrogation pattern for TGF-β responsiveness in adenocarcinoma affected the biology of the tumor, we examined the gene expression pattern using a small subset of genes involved in apoptosis, cell cycle regulation and tumor-induced angiogenesis.

Pathway-specific microarray analyses
Pathway-specific microarrays were probed using RNA isolated from lung lesions of 1- and 4-month old HT/LA and WT/LA mice representing hyperplasia/adenoma and adenocarcinomas, respectively. In benign tumors (hyperplasias and adenomas), 109 genes (58 up-regulated and 51 down-regulated) were differentially expressed; of these, 91 genes were involved in the process of apoptosis (40 genes) or cell cycle regulation (51 genes). This confirms that abrogation of apoptosis and cell cycle regulation is an important factor in the early development of lesions and progression. Adenocarcinomas of HT/LA and WT/LA mice had 179 genes (77 up-regulated and 102 down-regulated) differentially expressed; of these, a majority of the genes regulate the processes of apoptosis (48 genes) and cell cycle (56 genes) as expected. However, a large number of differentially regulated genes are known to be involved in tumor-induced angiogenesis (56 genes). Thirty genes from benign tumor lesions and 31 genes from malignant tumor lesions were selected to confirm microarray data expression patterns by quantitative real-time reverse transcription–PCR. On normalizing gene expression profiles in lesions to expression profiles of normal lung from wild-type mice, the number of differentially expressed genes between HT/LA and WT/LA lesions was further reduced to five genes in hyperplasia/adenoma lesions and 17 genes in adenocarcinomas (Tables I and II). Most of the differentially expressed genes in hyperplasias/adenomas are involved in cell cycle regulation or apoptosis as expected, whereas those in adenocarcinomas are involved in tumor-induced angiogenesis. The WT/LA adenocarcinomas had up-regulation of a number of angiogenic genes, suggesting that a full gene dose of TGF-β1 and TGF-β responsiveness is required for tumor-induced angiogenesis in the presence of K-ras mutation. In order to see if this translated into the biology of the tumors, we determined the angiogenesis index for adenocarcinomas of both HT/LA and WT/LA mice.

Angiogenesis index based on microvessel density
Adenocarcinomas of WT/LA mouse lungs had a significantly higher microvessel density compared with HT/LA mouse lung adenocarcinomas (Figure 5A). As tumor-induced angiogenesis is a prelude to invasion and metastatic spread, this implies a differential invasive potential for WT/LA mouse adenocarcinomas. The increased angiogenesis index in WT/LA mice suggests that for late-stage invasive tumors and possibly metastasis, a full gene dosage of TGF-β1 and TGF-β1 responsiveness is required. Expression of matrix metalloproteinase (MMP)-9 and MMP-2 were also up-regulated in 4 month WT/LA mouse tumors as compared with HT/LA tumors (Figure 5B). MMP-9 levels have been correlated with tumor-induced angiogenesis and tumor invasion (33,38). However, we did not see any metastases. The absence of metastasis can be explained by the fact that K-ras-mutated mice (both HT/LA and WT/LA) have a very short life span of 4–6 months and die before tumors can spread.


Figure 5
View larger version (40K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. The WT/LA mouse adenocarcinomas may have a higher invasive potential as compared with HT/LA mouse tumors. (A) WT/LA adenocarcinomas have a significant increase in angiogenesis index compared with HT/LA adenocarcinomas. Mean microvessel counts were performed in the areas of maximum vessel density for lung adenocarcinomas of TGF-β1 WT/K-ras LA (WT/LA) and TGF-β1 HT/K-ras LA (HT/LA) mice according to the international consensus protocol (see text). All data are the means ± SDs. P value < 0.01. White bar: TGF-β1 WT/K-ras LA; black bar: TGF-β1 HT/K-ras LA. (B) Gelatin zymograms were done for three sets of pooled lung tumors from 4-month-old mice. The WT/LA mouse lung tumors have increased expression of both MMP-9 and MMP-2 as compared with HT/LA mouse lung tumors.

 

   Discussion
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
Funding
 References
 
This study examines the effects of interactions between K-ras and TGF-β pathways on tumor biology, using a mouse model that harbors TGF-β1 haploinsufficiency and develops a random spontaneous K-ras oncogenic mutation. We find that presence of a LA oncogenic K-ras gene leads to induction of spontaneous lung tumors. In the presence of TGF-β1 heterozygosity (HT/LA) in addition to K-ras mutation, there is faster progression to a malignant phenotype as compared with tumors that have a wild-type functional gene dosage of TGF-β1. We also find that a functional TGF-β pathway, in addition to K-ras mutation, is required for tumor-induced angiogenesis during late-stage tumor progression. Thus, our data provide in vivo evidence and validation of earlier reported in vitro data that there is a functional correlation between oncogenic K-ras and TGF-β-mediated cellular responses during transformation and tumor progression (21,23–27,39). We provide evidence that oncogenic K-ras alone is not sufficient to promote a malignant phenotype and that additional abrogation and/or suppression of a tumor suppressor pathway is required for early progression to a malignant phenotype. In addition, we identify a number of genes involved in the process of apoptosis, cell cycle regulation and tumor-induced angiogenesis that lead to tumor progression in mouse lung tumors.

Activated Ras proteins act through several effector mechanisms and pathways to promote cell transformation and tumor induction. Ras transmits extracellular signals via sequential phosphorylation of pivotal kinases, including Raf and Erk1/2, which culminate in Ras-mediated cell transformation, proliferation and cell survival (26). At tumor initiation, K-ras activity levels may be similar to those in normal tissue (40), and loss of the normal K-rasv12 allele is not a prerequisite for early stages of tumor development (16). High levels of c-myc gene expression have been shown to occur prior to activation of the c-ki-ras gene and have been hypothesized to be a key predisposing feature that may lead to transformation of type II pneumocytes (41). c-myc repression is required for TGF-β1 responsiveness and oncogenic K-ras leads to increased c-myc expression and, thus, loss of TGF-β1 responsiveness (37). Hence, suppression of TGF-β1-mediated effects may be sufficient to push alveolar type II pneumocytes into hyperplasia. Our results support this premise (Figures 2A and 3) as hyperplasia induction is not seen to be dependent on TGF-β1 gene dosage.

In spite of a number of studies conducted over the years concerning K-ras in human cancer, the role of K-ras in early stages of tumorigenesis is not fully understood. It is thought that mere over-expression of an oncogene-like K-ras is not sufficient to promote transformation without a number of cooperating genetic events and suppression of tumor suppressor genes (42–44). Our model suggests that effective suppression of apoptosis through abrogation of TGF-β-dependent signaling pathways may be one of the primary mechanisms responsible for progression of tumors, following activational mutation of K-ras as TGF-β inhibits proliferation of both normal and tumor epithelial cells (17).

There is an exaggerated progression to a malignant phenotype in HT/LA mice when TGF-β1 is haploid. Decreased expression of TGF-β receptor proteins has been associated with loss of TGF-β1 sensitivity and enhanced tumor progression (18,19,28,29,45). We found that TGF-β RII expression was down-regulated in benign lung adenomas of both WT/LA and HT/LA mice. Abrogation of TGF-β1-mediated responses, like apoptosis, occurs in the early transformation phase of tumors in both HT/LA and WT/LA mice. However, a significant and sustained decrease in apoptosis is seen in adenomas of only HT/LA tumors. This appears to be mediated by a Bcl2/Bax-mediated mechanism that is more pronounced in the HT/LA tumors. Thus, our model suggests that an effective suppression of apoptosis through in the presence of TGF-β1 heterozygosity is one of the mechanisms responsible for progression to malignant tumors following an oncogenic mutation of K-ras.

A distinct group of genes is consistently and specifically altered during early and late progression stages of tumors. We have identified a number of differentially expressed novel genes that have not previously been associated with lung cancer. A majority of these genes are involved in cell cycle, apoptosis (in benign tumors) and angiogenesis (in malignant tumors). Apoptosis has been shown to be due to the convergence of multiple pathways from numerous different initiating events and insults. Down-regulation of one or more of the pathways will affect the sum total of the apoptotic index in a tissue mass and will lead to earlier tumor progression and/or acquisition of a malignant phenotype due to accumulation of mutations. This, in conjunction with factors that lead to cell cycle progression or loss of growth inhibition, is also likely to promote accumulation of mutations and earlier tumor progression. The significant abrogation of apoptosis in adenomas of HT/LA mice compared with WT/LA tumors indicates that there is a modulation of apoptotic response due to TGF-β1 haploinsufficiency. Further, in vitro data have shown that strong activation of Raf can provide protection from apoptotic stimuli and allows the cell to respond to TGF-β with increased invasiveness (46). Raf, which is downstream of the activated K-ras, may be one of the factors involved. In addition, the loss of the growth-inhibitory response to TGF-β in transformed cells has been attributed to mutations that directly inactivate TGF-β RII, Smad2 and Smad4 (39). Oncogenic Ras can also interfere with TGF-β signaling by inhibiting nuclear accumulation of Smad2 and Smad3 (21) and early termination of Smad signaling by Smad4 degeneration (47). This is accomplished by Ras acting through the Erk signaling intermediate and leading to specific phosphorylations on Smad2/3 that link the DNA-binding domain and transactivational domain and preventing nuclear translocalization (21). Oncogenic Ras may also repress TGF-β signaling by down-regulation of Smad4 and subverting its tumor suppressor activity (47). These two processes appear to be operating in lung lesions in HT/LA mice where there is a more profound abrogation of growth-inhibitory response in early tumors.

It has been reported that Ras activation and mutation may also have effects via TGF-β pathway components and may alter TGF-β signaling with different outcomes (21). TGF-β can contribute to the invasive phenotype at later stages of tumorigenesis by actions such as suppressing immune surveillance, fostering tumor invasion and promoting development of metastases (17–19). Ras-transformed cells not only exhibit a limited growth-inhibitory response to TGF-β (24,25) but also may respond to TGF-β with invasive activity (48). The tumor biology of WT/LA malignant tumors supports this premise. We report that oncogenic activation of Ras alone and in the presence of TGF-β1 haploinsufficiency produces different outcomes in tumors, not only at the gene transcription level but also in translation into biological effects. Our data indicate that while Ras transformation suppresses TGF-β-induced growth inhibition in early tumorigenesis, in later malignant lesions, Ras subverts the TGF-β pathway into stimulating invasion. Over-expression of Smad4, in vitro, has been shown to result in restoration of TGF-β signaling, despite the presence of oncogenic Ras (47). The data presented here support this premise, and we propose that over-expression of Smad3 and Smad4 leads to transcriptional activation of angiogenic genes seen in WT/LA adenocarcinoma tumors. Thus, tumor-induced angiogenesis seen here is a result of a TGF-β-dependent mechanism as malignant lesions of HT/LA mouse do not show up-regulation of either angiogenic genes or increase in angiogenesis index. This happens despite an early progression to adenocarcinoma in HT/LA mice. Our microarray analyses using predominantly adenocarcinoma lung tumors showed that all but 6 of 19 differentially expressed (WT/LAversus HT/LA) genes are involved in the process of angiogenesis. Both TGF-β1 and K-ras have either a direct or indirect interaction with all but four (Sesn2, Rnase4, Max and Ccna1) of these genes (Table II). Thus, this supports the premise that TGF-β1 effector pathways are responsible for up-regulation of genes involved in the process of tumor-related/induced angiogenesis. For example, connective tissue growth factor, a growth factor that is known to have a role in cellular proliferation, was found to be differentially regulated in our mouse model. Connective tissue growth factor homogenous knockout mice show decreased expression levels of several pro-angiogenic factors like vascular endothelial growth factor and MMP-9 and reduced angiogenesis (38). In addition, connective tissue growth factor expression is increased by both Smad3 (49) and TGF-β1 (50). Other genes that have been implicated in angiogenesis and are differentially expressed in our mouse model include Cxcl1, Dsip1, Ets2, Fgfr2, Flt1, Hif1{alpha}, Igf1, Igfbp3, Il12a, Max, Pdgfb, Ptgs2 and Vegf-c. The evidence presented here indicates the importance of these genes in tumor progression and additional studies will be required to understand the interactions between these genes and TGF-β1 and K-ras in lung tumorigenesis.

Finally, the novel changes in gene expression pattern identified in this study may provide significant insights into late-stage tumor progression in lung tumors. The use of the compound oncogene–tumor suppressor mouse model described here provides identification of previously unknown targets, which may, in the future, act as new targets for cancer therapy that are important in late-stage tumor progression, invasion and metastasis. They may also help to identify changes that are associated with early tumor progression in some of the intricate intersecting functional pathways that are known to be affected by carcinogenesis.


   Funding
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Funding
References
 
Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research.


   Acknowledgments
 
We thank Drs L.Anderson and K.Flanders (National Cancer Institute) for their comments during preparation of this article.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Material and methods
 Results
 Discussion
 Funding
 References
 

  1. Sekido Y, et al. Molecular genetics of lung cancer. Ann. Rev. Med. (2003) 54:73–87.[CrossRef][Medline]
  2. Yokota J, et al. Molecular footprints of human lung cancer progression. Cancer Sci. (2004) 95:197–204.[CrossRef]
  3. Breuer R, et al. Molecular pathology of non-small-cell lung cancer. Respiration (2005) 72:313–330.[CrossRef][Web of Science][Medline]
  4. Johnson BE, et al. Epidermal growth factor receptor mutations in patients with non-small cell lung cancer. Cancer Res. (2005) 65:7525–7529.[Abstract/Free Full Text]
  5. Hanahan D, et al. The hallmarks of cancer. Cell (2000) 100:57–70.[CrossRef][Web of Science][Medline]
  6. Mitsuuchi Y, et al. Cytogenetics and molecular genetics of lung cancer. Am. J. Med. Genet. (2002) 115:183–188.[CrossRef][Web of Science][Medline]
  7. Tuveson D, et al. Modeling human lung cancer in mice: similarities and shortcomings. Oncogene (1999) 18:5318–5324.[CrossRef][Web of Science][Medline]
  8. You M, et al. Activation of the Ki-ras protooncogene in spontaneously occurring and chemically induced lung tumors of the strain A mouse. Proc. Natl Acad. Sci. USA (1989) 86:3070–3074.[Abstract/Free Full Text]
  9. Belinsky S, et al. Role of clara cells and type II cells in the development of pulmonary tumors in rats and mice following exposure to a tobacco-specific nitrosamine. Exp. Lung Res. (1991) 17:263–278.[Web of Science][Medline]
  10. Malkinson A. Primary lung tumors in mice: an experimentally manipulable model of human adenocarcinoma. Cancer Res. (1992) 52:2670s–2676s.[Abstract/Free Full Text]
  11. Kawano R, et al. Role of point mutation of the K-ras gene in tumorigenesis of B6C3F1 mouse lung lesions induced by urethane. Jpn. J. Cancer Res. (1995) 86:802–810.[CrossRef][Web of Science]
  12. Horio Y, et al. Ki-ras and p53 mutations are early and late events, respectively, in urethane-induced pulmonary carcinogenesis in A/J mice. Mol. Carcinog. (1996) 17:217–223.[CrossRef][Web of Science][Medline]
  13. Li J, et al. RASSF1A promoter methylation and Kras2 mutations in non small cell lung cancer. Neoplasia (2003) 5:362–366.[Web of Science][Medline]
  14. Johnson L, et al. Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature (2001) 410:1111–1116.[CrossRef][Medline]
  15. Weintraub S. Dormant tumor suppressor pathway in tumors. Am. J. Respir. Cell Mol. Biol. (1999) 20:541–542.[Free Full Text]
  16. Zhang Z, et al. Wildtype Kras2 can inhibit lung carcinogenesis in mice. Nat. Genet. (2001) 29:25–33.[CrossRef][Web of Science][Medline]
  17. Elliott RL, et al. Role of transforming growth factor beta in human cancer. J. Clin. Oncol. (2005) 23:2078–2093.[Abstract/Free Full Text]
  18. Glick A. TGF beta: back to the future: revisiting its role as a transforming growth factor. Cancer Biol. Ther. (2004) 3:276–283.[Web of Science][Medline]
  19. Sun L. Tumor-suppressive and promoting function of transforming growth factor beta. Front. Biosci. (2004) 9:1925–1935.[Web of Science][Medline]
  20. Mulder KM. Role of Ras and Mapks in TGFbeta signaling. Cytokine Growth Factor Rev. (2000) 11:23–35.[CrossRef][Web of Science][Medline]
  21. Kretzschmar M, et al. A mechanism of repression of TGF beta/Smad signaling by oncogenic Ras. Genes Dev. (1999) 13:804–816.[Abstract/Free Full Text]
  22. Bhowmick NA, et al. Transforming growth factor-beta1 mediates epithelial to mesenchymal transdifferentiation through a RhoA-dependent mechanism. Mol. Biol. Cell (2001) 12:27–36.[Abstract/Free Full Text]
  23. Longstreet M, et al. Loss of transforming growth factor beta 1 (TGF-beta 1)-induced growth arrest and p34cdc2 regulation in ras-transfected epithelial cells. Oncogene (1992) 7:1549–1556.[Web of Science][Medline]
  24. Filmus J, et al. Development of resistance mechanisms to the growth-inhibitory effects of transforming growth factor-beta during tumor progression. Curr. Opin. Oncol. (1993) 5:123–129.[Medline]
  25. Oft M, et al. TGF-beta1 and Ha-Ras collaborate in modulating the phenotypic plasticity and invasiveness of epithelial tumor cells. Genes Dev. (1996) 10:2462–2477.[Abstract/Free Full Text]
  26. Zavadil J, et al. TGF-beta and epithelial-to-mesenchymal transitions. Oncogene (2005) 24:5764–5774.[CrossRef][Web of Science][Medline]
  27. Roberts AB, et al. Smad3 is key to TGF-beta-mediated epithelial-to-mesenchymal transition, fibrosis, tumor suppression and metastasis. Cytokine Growth Factor Rev. (2006) 17:19–27.[CrossRef][Web of Science][Medline]
  28. Jakowlew S, et al. Reduction in transforming growth factor-beta type II receptor in mouse lung carcinogenesis. Mol. Carcinog. (1998) 22:46–56.[CrossRef][Web of Science][Medline]
  29. Kang Y, et al. Enhanced tumorigenesis and reduced transforming growth factor-beta type II receptor in lung tumors from mice with reduced gene dosage of transforming growth factor-beta1. Mol. Carcinog. (2000) 29:112–126.[CrossRef][Web of Science][Medline]
  30. Tang B, et al. Transforming growth factor-beta1 is a new form of tumor suppressor with true haploid insufficiency. Nat. Med. (1998) 4:802–807.[CrossRef][Web of Science][Medline]
  31. Shull MM, et al. Targeted disruption of the mouse transforming growth factor-(beta)1 gene results in multifocal inflammatory disease. Nature (1992) 359:693–699.[CrossRef][Medline]
  32. Kulkarni A, et al. Transforming growth factor beta1 null mutation in mice causes excessive inflammatory response and early death. Proc. Natl Acad. Sci. USA (1993) 90:770–774.[Abstract/Free Full Text]
  33. Pandey J, et al. The Edpm5 locus prevents the ‘angiogenic switch’ in an estrogen-induced rat pituitary tumor. Carcinogenesis (2004) 25:1829–1838.[Abstract/Free Full Text]
  34. Vermeulen P, et al. Second international consensus on the methodology and criteria of evaluation of angiogenesis quantification in solid human tumours. Eur. J. Cancer (2002) 38:1564–1579.[CrossRef][Web of Science][Medline]
  35. Nikitin AY, et al. Classification of proliferative pulmonary lesions of the mouse: recommendations of the mouse models of human cancers consortium. Cancer Res. (2004) 64:2307–2316.[Abstract/Free Full Text]
  36. Jackson E, et al. Analysis of lung tumor initiation and progression using conditional expression of oncogenic K-ras. Genes Dev. (2001) 15:3243–3248.[Abstract/Free Full Text]
  37. Chen C-R, et al. Inaugural article: defective repression of c-myc in breast cancer cells: a loss at the core of the transforming growth factor beta growth arrest program. Proc. Natl Acad. Sci. USA (2001) 98:992–999.[Abstract/Free Full Text]
  38. Ivkovic S, et al. Connective tissue growth factor coordinates chondrogenesis and angiogenesis during skeletal development. Development (2003) 130:2779–2791.[Abstract/Free Full Text]
  39. Massague J. Integration of Smad and MAPK pathways: a link and a linker revisited. Genes Dev. (2003) 17:2993–2997.[Free Full Text]
  40. Ramakrishna G, et al. Ki-ras and the characteristics of mouse lung tumors. Mol. Carcinog. (2000) 28:156–167.[CrossRef][Web of Science][Medline]
  41. Polly P, et al. High levels of c-myc gene expression precede point mutational activation of Ki-ras in mouse lung cancer. Cancer Lett. (1994) 76:87–92.[CrossRef][Web of Science][Medline]
  42. Franza BJ, et al. In vitro establishment is not a sufficient prerequisite for transformation by activated ras oncogenes. Cell (1986) 44:409–418.[CrossRef][Web of Science][Medline]
  43. Land H, et al. Tumorigenic conversion of primary embryo fibroblasts requires atleast two cooperating oncogenes. Nature (1983) 304:596–602.[CrossRef][Medline]
  44. Zhang Y, et al. Regulation of Smad degradation and activity by Smurf2, an E3 ubiquitin ligase. Proc. Natl Acad. Sci. USA (2001) 98:974–979.[Abstract/Free Full Text]
  45. Colasante A, et al. Gene expression of transforming growth factor beta receptors I and II in non-small-cell lung tumors. Cytokine (2003) 24:182–189.[CrossRef][Web of Science][Medline]
  46. Lehmann K, et al. Raf induces TGF beta production while blocking its apoptotic but not invasive responses: a mechanism leading to increased malignancy in epithelial cells. Genes Dev. (2000) 14:2610–2622.[Abstract/Free Full Text]
  47. Saha D, et al. Oncogenic Ras represses transforming growth factor-beta/Smad signaling by degrading tumor suppressor Smad4. J. Biol. Chem. (2001) 276:29531–29537.[Abstract/Free Full Text]
  48. Oft M, et al. TGF beta signaling is necessary for carcinoma cell invasiveness and metastasis. Curr. Biol. (1998) 8:1243–1252.[CrossRef][Web of Science][Medline]
  49. Lakos G, et al. Targeted disruption of TGF-beta/Smad3 signaling modulates skin fibrosis in a mouse model of scleroderma. Am. J. Pathol. (2004) 165:203–217.[Abstract/Free Full Text]
  50. Bonniaud P, et al. Smad3 null mice develop airspace enlargement and are resistant to TGF-beta-mediated pulmonary fibrosis. J. Immunol. (2004) 173:2099–2108.[Abstract/Free Full Text]
Received March 1, 2007; revised June 6, 2007; accepted June 10, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cancer Res.Home page
K. A. Miller, N. Yeager, K. Baker, X.-H. Liao, S. Refetoff, and A. Di Cristofano
Oncogenic Kras Requires Simultaneous PI3K Signaling to Induce ERK Activation and Transform Thyroid Epithelial Cells In vivo
Cancer Res., April 15, 2009; 69(8): 3689 - 3694.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/12/2589    most recent
bgm136v1
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 ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Pandey, J.
Right arrow Articles by Jakowlew, S. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pandey, J.
Right arrow Articles by Jakowlew, S. B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?