The transforming growth factor-betas (TGF-betas) are multifunctional regulators of cellular growth and function and potent inhibitors of epithelial cell proliferation. The widespread expression of TGF-beta indicates a pivotal role in epithelial homeostasis. These features make the TGF-betas attractive candidates for new therapeutic intervention approaches to the prevention and treatment of cancer. TGF-beta plays a major role in adult physiology, as well as in the control of differentiation and morphogenesis in embryonic development. The tissue distribution pattern of the TGF-betas, which include TGF-betas 1, 2, and 3 in mammals, has possible significance for signaling roles in epithelial-mesenchymal interactions during embryogenesis, as well as in cancer and carcinogenesis. TGF-beta is secreted by a variety of normal and malignant cells. The TGF-betas function through a set of cell surface protein receptors that includes TGF-beta type I (RI) and type II (RII). TGF-beta RII can bind TGF-beta directly to form a complex, which then is able to bind TGF-beta RI, and TGF-beta RII is then able to phosphorylate TGF-beta RI, which is necessary for signal transduction. The TGF-beta signaling system has been implicated as a tumor suppressor pathway in several organ systems. Loss of functional TGF-beta RI or RII contributes to loss of TGF-beta responsiveness, resulting in tumor progression. Defects in responsiveness to TGF-beta have been implicated in the pathogenesis of several human epithelial cancers, suggesting that TGF-beta has tumor suppressor properties. But, many advanced tumors show increased expression of TGF-beta, and parallel poor prognosis, suggesting that TGF-beta also has properties of a tumor suppressor. The complex mechanisms of action of TGF-beta in its capacity as a tumor suppressor and a tumor promoter and the target genes that are regulated by TGF-beta in these different capacities must be clearly defined to be able to exploit this gene for clinical therapeutic intervention purposes. Our broad goal is to determine how TGF-beta signaling regulates the development and malignant transformation of epithelial cells, now with a primary emphasis on lung epithelial cells. Our approach is based on the hypotheses that (1) Signaling pathways for TGF-beta occur that are separate from the growth inhibitory pathway and these pathways may be operational in lung cancer cells that are resistant to growth inhibition by TGF-beta; (2) The integrity of the TGF-beta signaling pathways is important for the normal regulation of downstream target molecules of TGF-beta and dysregulation of the activity of these signaling pathways during progressive stages of lung tumorigenesis impacts on the regulation of these target genes; (3) The tumor suppressor and tumor promoter activities of TGF-beta differentially regulate target genes that contribute to these activities. Our research efforts are focused around the proposal that there is a delicate balance between the tumor suppressor and tumor promoter roles of TGF-beta in epithelial tissues. Our premise is that the ability of TGF-beta to act as a tumor suppressor is primary to that of a tumor promoter in normal epithelial cells, and that the ability of TGF-beta to act as a tumor promoter takes on added significance as cells that are sensitive to TGF-beta become transformed and eventually resistant to TGF-beta, expand clonally, and ultimately progress to malignancy. Awareness of the timing of the phenotypic switch from TGF-beta sensitivity to TGF-beta resistance that occurs during tumorigenesis is likely to be important in designing and applying strategies for tumor prevention and treatment. The most recent efforts have examined the role of TGF-beta 1 and TGF-beta RII as tumor suppressor genes in lung carcinogenesis. Because of the limitations imposed by the advanced stages of human lung biopsy specimens and cell culture systems, we have complemented and expanded our studies with whole animal model systems. We mated C57BL/6 mice heterozygous (HT) for deletion of the TGF-beta 1 gene with A/J mice to produce AJBL6 TGF-beta 1 HT progeny and their wildtype (WT) littermates. Immunohistochemical staining, in situ hybridization, and northern blot analyses showed lower staining and hybridization for TGF-beta 1 protein and mRNA, respectively, in the lungs of normal HT mice compared to WT mice. Competitive reverse transcription-polmerase chain reaction (CRT-PCR) amplification showed the level of TGF-beta 1 mRNA in the lungs of HT mice to be 4-fold lower than in WT lung. When challenged with ethyl carbamate, lung adenomas were detected in 55% of HT mice by 4 months, while in only 25% of WT littermates at this time. While all HT mice had adenomas by 6 months, it was not until 10 months before all WT mice had adenomas. After 12 months, the average number of adenomas was 5-fold higher in HT lungs compared to WT lungs. Most dramatic was the appearance of lung carcinomas in HT mice 8 months before they were visible in WT mice. Thus, the AJBL6 TGF-beta 1 HT mouse provides an excellent model system to examine carcinogen-induced lung tumorigenesis by increasing progressive lesion incidence and multiplicity relative to their wildtype littermates. Immunohistochemical staining showed expression of the TGF-beta RI at moderate to strong levels in lung adenomas and carcinomas in HT and WT mice. In contrast, while weak immunostaining for TGF-beta RII was detected in 67% of HT carcinomas at 12 months, only 22% of WT carcinomas showed weak staining for this protein. Individual lung carcinomas showing reduced TGF-beta RII expression and adjacent normal bronchioles were excised from HT lungs using laser capture microdissection, and CRT-PCR amplification of the extracted RNA showed 12-fold less TGF-beta RII mRNA in these carcinomas compared to bronchioles. Decreasing TGF-beta RII mRNA levels occurred with increasing tumorigenesis in lung hyperplasias, adenomas, and carcinomas, with carcinomas having 4- and 7-fold lower levels of TGF-beta RII mRNA than adenomas and hyperplasias, respectively. Our findings show enhanced ethyl carbamate-induced lung tumorigenesis in AJBL6 HT mice compared to WT mice, suggesting that both TGF-beta 1 alleles are necessary for tumor suppressor activity. Reduction of TGF-beta RII mRNA expression in progressive stages of lung tumorigenesis in HT mice suggests that loss of TGF-beta RII may play an important role in the promotion of lung carcinogenesis in mice with reduced TGF-beta 1 gene dosage upon challenge with carcinogen. This new mouse model will facilitate our studies by providing a means to generate substantial numbers of lung tumors in progressive stages of development with reduced latency so that we can begin to address the contribution of the tumor suppressor and tumor promoter activities of the TGF-beta signaling pathway to lung tumorigenesis and the ability of TGF-beta to regulate target genes that may be important in tumorigenesis. To complement our animal model systems, the TGF-beta 1 sensitive epithelial non-small cell lung cancer (NSCLC) cell line NCI-H727, whose growth can be inhibited by TGF-beta 1, was used as a model system to identify potential genes involved in TGF-beta 1 growth inhibition. Comparative cDNA expression patterns between NCI-H727 cells treated with TGF-beta 1 or vehicle alone were generated by differential mRNA display. Among the several cDNA fragments that represent genes that are potentially differentially regulated by TGF-beta 1 that were recovered, one 496-bp cDNA fragment that hybridized to a 2.7-kb mRNA species and whose expression was differentially increased 3-fold by TGF-beta 1 in NCI-H727 cells by northern blot analysis, revealed no significant match to any known gene sequence. The mRNA transcript of this novel gene, which we named Differentially Expressed Nucleolar TGF-beta 1 Target (DENTT), is induced by TGF-beta 1 in NSCLC cells that are sensitive to TGF-beta 1. In contrast, DENTT mRNA expression was not affected by TGF-beta 1 in normal human bronchial epithelial cells or in NSCLC cells resistant to TGF-beta 1 growth inhibition. Human brain cDNA library screening and 5' rapid amplification of cDNA ends yielded full-length DENTT cDNA containing an 899-bp open reading frame encoding a predicted 633-amino acid protein with 4 potential nuclear localization signals (NLS) and 2 coiled-coil regions. DENTT contains a conserved 191-residue domain that shows significant identity to, and defines, the TSPY/TSPY-like/SET/NAP-1 superfamily. DENTT transfected into COS-7 cells showed nucleolar and cytoplasmic locallization. Transfection of EGFP-tagged DENTT NLS deletion constructs lacking the bipartite NLS-1 were excluded from the nucleolus. While NLS-1 is necessary for localization of DENTT, it is not sufficient for sole nucleolar localization. These findings show that DENTT mRNA induction by TGF-beta 1 correlates with induction of TGF-beta 1 mRNA, induction of extracellular matrix gene expression, and inhibition of colony formation in soft agarose in TGF-beta 1 responsive NSCLC cells when exposed to TGF-beta 1. Our findings also suggest that this novel TGF-beta 1 target gene has distinct domains for direction to different subnuclear locations that may be important in its regulation and its ability to interact with other proteins in lung tumorigenesis. The significance of this project is to determine the role of TGF-beta in cancer and the manner in which the signaling pathway of TGF-beta and its target genes is altered from the normal condition. The identification of other important proteins with which TGF-beta 1 and its target genes interact in normal cells will be important in defining and determining how this interaction may be altered in cancer and carcinogenesis.
Showing the most recent 10 out of 13 publications
