Since TGF-α as well as EGF itself may induce phenotypic transformation in normal cells in vitro, it is reasonable to suggest that these agents may act as promoters in vivo. Furthermore, there is now ample evidence that TGF-α is expressed in a variety of malignant neoplasms, and many of these neoplasms exhibit an enhanced level of the EGF receptor, many times owing to gene rearrangements or amplification of the EGF receptor gene as a component of the stage of tumor progression (cf. Salomon et al., 1990).
Various types of “dysregulation” of growth factor–receptor pathways in transformed cells have been described (cf. Hatakeyama and Taniguchi, 1988). A number of examples of changes in the epidermal growth factor–receptor pathway in transformed cells have been described (To- daro et al., 1976; Cherington et al., 1979; Hollenberg et al., 1979). Several of these reports (e.g., Hollenberg et al., 1979; Cherington et al., 1979) have demonstrated a decreased requirement for epidermal growth factor in transformed cells. The role that autocrine stimulation of such cells played in these findings was not clear at that time. In addition, in Syrian hamster embryo cells transformed in vitro, alterations and the responsiveness of these cells to epidermal growth factor, platelet-derived growth factor, and transforming growth factor β were significantly altered (Is- fort et al., 1994).
Figure 16.14 Diagram of the types of cellular secretion. The black dots represent the secretory product or ligand and the heavy black half circles on the cell membrane the specific receptor for the secreted ligand. (Adapted from Todaro et al., 1981, with permission of the authors and publisher.)
The response of transformed and neoplastic cells to TGF-β tends to be lost, primarily dur- ing the stage of progression. Since TGF-β acts primarily to inhibit the growth of epithelial and many mesenchymal cells, such a loss of receptor activity and function is perhaps to be expected. There is now substantial evidence that much of this TGF-β resistance is the consequence of in- activating mutations either in the type 1 or type 2 TGF-β receptors or one of the several Smad proteins (Reiss, 1997). In fact, Smad2 has been implicated as a tumor suppressor gene involved in the control of cell invasion (Prunier et al., 1999). In contrast, Oft and associates (1998) have argued that TGF-β signaling may be rather important for the invasiveness and metastasis of cer- tain carcinoma cells both in vivo and in vitro. Krieg et al. (1991) reported that constitutive over- expression of TGF-β1 mRNA was observed in malignant carcinomas but not in benign premalignant lesions. Furthermore, 100% of pancreatic and 83% of colonic neoplasms have a mutation affecting at least one component of the TGF-β pathway (cf. Blobe et al., 2000). How- ever, the activity of the entire pathway was not investigated. An autocrine type of secretion for TGF-β has been described in transformed fibroblasts, and in many instances this is associated with aberrant EGF receptor regulation (Newman, 1993). Repression or inactivation of TGF-β receptors has also been described in the genetically based retinoblastoma and in hereditary non- polyposis colorectal cancer (cf. Ruscetti et al., 1998).
Thus, it appears that growth factor signaling pathways are largely aberrant in transformed cells in vivo as they are in many neoplastic cells in vitro. Furthermore, a number of pathways in other growth factors are also abnormal in transformed cells, as noted in Table 16.10. The EGF/TGF-α autocrine pathway is not functional in all neoplasms, although all cells transformed by chemicals appear to exhibit an enhancement in TGF-α expression (cf. Lee et al., 1993). It is likely that such increased expression of these growth factors makes up the basis for the de- creased requirement for EGF in chemically transformed mouse cells (Cherington et al., 1979). Furthermore, some neoplastic cells may begin to express growth factor receptors and thus re- spond to growth factors to which their normal counterparts would not (cf. Descamps et al.,
1998). A similar effect had earlier been reported with the platelet-derived growth factor and its receptor (Bowen-Pope et al., 1984). Insulin-like growth factors I and II may also be expressed at higher levels in a variety of neoplasms in vivo and in vitro, often with their receptor exhibiting expression as well, leading to autocrine growth potential (Werner and LeRoith, 1996). Mutations in the IGF-II/mannose-6-phosphate receptor, which prevent such autocrine effects or growth fac- tor effects on the neoplastic cells themselves, have been reported, as discussed in Chapter 15. Interestingly, hepatocyte growth factor may actually inhibit proliferative activity of a variety of neoplastic cell lines (Tajima et al., 1991). In another cell culture system, Seslar et al. (1995) demonstrated an interaction between TGF-β production and hepatocyte growth factor expres- sion, with a clear interaction between various cell types in the culture.
Since the responses to growth factors in the neoplastic cell may be altered as a result of the development of autocrine stimulation, mutations in signaling pathways, interactions between growth factors and adjacent cell types (paracrine secretion), or altered imprinting, it is very likely that growth factors do play a role in the enhanced cell replication during the stage of pro- gression. It is very likely that many of the characteristics described in this section are the result of evolving karyotypic instability in this stage, and thus, while many interesting different pat- terns of responses may be seen in transformed and neoplastic cells, no single alteration may be seen ubiquitously in all transformed and neoplastic cells.