Loss of growth control by TGF-beta occurs at a late stage of mouse skin carcinogenesis and is independent of ras gene activation.

The relationship between the expression of a mutant ras gene in epithelial cells and loss of responsiveness to the negative effects of transforming growth factor beta (TGF-beta) is presently unclear. We have investigated this question using a series of cell lines derived from benign and malignant mouse skin tumours which express mutant forms of the H-ras gene. Immortalised, non-tumorigenic mouse epidermal cells respond to TGF-beta by cessation of growth, whereas in a series of malignant carcinoma lines the response was substantially reduced. Introduction of a mutant H-ras gene into the immortalised cells did not lead to any appreciable change in TGF-beta responsiveness, suggesting that initiation of carcinogenesis by ras mutation does not directly alter growth control by this pathway. Of two non-tumorigenic papilloma lines tested which had mutant H-ras genes, one retained complete sensitivity to TGF-beta, whereas the other showed a similar response to carcinomas. We conclude that growth control by TGF-beta is lost at a relatively late stage of carcinogenesis in this system, and is independent of ras gene activation.

Further evidence for the importance of parental source of the Xce allele in X chromosome inactivation.

Using mice that were mosaics for both Xce and phosphoglycerate kinase (Pgk-1) alleles, we present further evidence that the parental source of the X chromosome may affect the probability of that X chromosome remaining active. The reciprocal cross differences in PGK-1 activity described here are intermediate between those published previously for other alleles of Xce.

Discordant transforming growth factor beta 1 RNA and protein localization during chemical carcinogenesis of the skin.

Transforming growth factor beta (TGF-beta) inhibits proliferation of normal keratinocytes, and this response is retained, to variable extents, in benign tumors of the skin (S. Haddow, D. J. Fowlis, K. Parkinson, R. J. Akhurst, and A. Balmain, Oncogene, 6: 1465-1470, 1991). To investigate the profile of TGF-beta biosynthesis during various stages of chemical carcinogenesis of the skin, we used a combination of ribonuclease protection assay, in situ hybridization with gene-specific probes for TGF-beta 1, -beta 2, and -beta 3, and immunohistochemistry with isoform-specific antibodies against TGF-beta 1. Following 12-O-tetradecanoylphorbol-13-acetate treatment of adult mouse skin, there was a rapid induction of TGF-beta 1 protein. Intracellular TGF-beta 1 protein was localized to suprabasal keratinocytes, and the extracellular form was localized predominantly to the dermis. Despite ubiquitous induction of TGF-beta 1 protein by 12-O-tetradecanoylphorbol-13-acetate in various mouse strains, we noted strain-specific differences in the quantitative induction of TGF-beta 1 RNA. Papillomas and carcinomas induced in vivo had elevated levels of TGF-beta 1 RNA within the basal keratinocyte compartment but did not contain significant levels of TGF-beta 1 protein within the tumor. We postulate that the tumor evades TGF-beta 1-controlled negative growth regulation by altered translational and/or posttranslational processing mechanisms of this growth factor. Levels of TGF-beta 2 and -beta 3 RNA were not elevated at any stage of chemical carcinogenesis of the skin.

Altered epidermal cell growth control in vivo by inducible expression of transforming growth factor beta 1 in the skin of transgenic mice.

An inducible bovine KIV* keratin gene promoter was used to target expression of latent or activated transforming growth factor beta 1 (TGF beta 1) to keratinocytes in transgenic mice. This short (2.2-kb) keratin 6 (K6) promoter element was generally silent in untreated animals but was induced in keratinocytes when placed in culture or, in vivo, in response to hyperplasia that follows topical application of the tumor promoter, 12-O-tetradecanoylphorbol-13-acetate. All of the K6-TGF beta 1 transgenic lines studied showed attenuation of the basal keratinocyte proliferative response to 12-O-tetradecanoylphorbol-13-acetate as a consequence of inducible TGF beta 1 gene expression. One of the six lines studied showed constitutive transgene expression at low levels in the skin, and this line had a 2- to 3-fold increase in epidermal DNA labeling index over control mice. Although in vitro TGF beta 1 is known to be a potent negative regulator of epithelial cell proliferation, in vivo TGF beta 1 has complex biological activities and can act as either a positive or negative regulator of keratinocyte proliferation.

The role of TGF-beta s in mammalian development and neoplasia.

To date, three mammalian TGF-beta isoforms have been identified, each encoded by different genetic loci. Through each is very similar in primary amino acid structure, there are clear differences both in the mature bioactive peptide region and in the latency-associated peptide, which could potentially confer differential biological specificity. As one route to investigate differential biological function in vivo we have used gene specific probes for in situ hybridization studies to examine the distribution of RNA transcripts during mammalian embryogenesis. Mouse embryos from 6 to 14.5 gestational age and human embryos from 32 to 57 days post-fertilization have been probed. A general conclusion from these studies is that each TGF-beta gene has a distinct, through overlapping, pattern of transcript distribution and that this pattern, in most cases, is conserved between mouse and man. We have focused on the biological function the TGF-betas play in certain epithelia and in cardiogenesis, which will be discussed in this presentation.

Functional loss of tumour suppressor genes in multistage chemical carcinogenesis.

Studies of multistage carcinogenesis in mouse skin have provided many of the early concepts of tumour initiation, promotion and progression. Genetic approaches have led to the identification of a number of mutational alterations in proto-oncogenes and tumour suppressor genes which take place at specific stages of carcinogenesis in this particular system. Initiation involves, at least in a proportion of tumours, mutational activation of the cellular H-ras proto-oncogene. Trisomy of chromosome 7, which develops during the premalignant clonal expansion phase, possibly as a consequence of tumour promoter treatment, is followed by further alterations on chromosome 7 which lead to a relative increase in the expression of mutant ras alleles. The p53 tumour suppressor gene undergoes mutational alteration and loss of heterozygosity in a proportion of squamous carcinomas but this particular gene does not appear to be involved in the further transition of squamous carcinomas to highly undifferentiated spindle cell tumours. The latter transition appears to be a recessive event which can be complemented by fusion with cells at earlier stages of malignancy. Mouse skin carcinogenesis therefore continues to provide invaluable information on the nature of the genetic and biological transitions which occur during the step-wise progression of normal cells to malignancy.

Concerted action of TGF-beta 1 and its type II receptor in control of epidermal homeostasis in transgenic mice.

Transforming growth factor-beta 1 (TGF-beta 1) is a modulator of cellular proliferation, differentiation, and extracellular matrix deposition. It is a potent epithelial growth inhibitor and can alter the differentiative properties of keratinocytes, in vitro, but little is known about its normal physiological function in the epidermis in vivo. Transgenic mice were generated using a keratin 10 (K10) gene promoter to drive constitutive expression of TGF-beta 1 in the suprabasal keratinocyte compartment. Surprisingly, these mice showed a two- to threefold increase in epidermal DNA labeling index over control mice, in the absence of hyperplasia. The transgene, however, acted in the expected fashion, as a negative regulator of cell growth, when hyperplasia was induced by treatment by 12-tetradecanoyl-phorbol-13-acetate (TPA). Epidermal TGF-beta type I and II receptor (T beta RI and T beta RII) levels were examined in control and transgenic mice during induction of hyperplasia by TPA. Whereas T beta RI levels remained relatively constant, T beta RII expression was strongly induced in TPA-treated skins, prior to the induction of the growth inhibitory response to TGF-beta 1, and its level of expression correlated with growth sensitivity to TGF-beta 1 in vivo and in vitro. These results suggest that TGF-beta 1 and its type II receptor are part of the endogenous homeostatic regulatory machinery of the epidermis.

TGFbeta1 inhibits the formation of benign skin tumors, but enhances progression to invasive spindle carcinomas in transgenic mice.

TGFbeta1 has been implicated in cell cycle control and carcinogenesis. To address the exact function of TGFbeta1 in skin carcinogenesis in vivo, mice with TGFbeta1 expression targeted to keratinocytes were subjected to long-term chemical carcinogenesis treatment. TGFbeta1 showed biphasic action during multistage skin carcinogenesis, acting early as a tumor suppressor but later enhancing the malignant phenotype. The transgenics were more resistant to induction of benign skin tumors than controls, but the malignant conversion rate was vastly increased. There was also a higher incidence of spindle cell carcinomas, which expressed high levels of endogenous TGFbeta3, suggesting that TGFbeta1 elicits an epithelial-mesenchymal transition in vivo and that TGFbeta3 might be involved in maintenance of the spindle cell phenotype. The action of TGFbeta1 in enhancing malignant progression may mimic its proposed function in modulating epithelial cell plasticity during embryonic development.