Insulin-like growth factors: the unrecognised oncogenes.

The identification of genes involved in the transformation of cells proceeded rapidly during the 1970s and 1980s from the analysis of oncogenic retroviruses and the identification of oncogenes using DNA transfection techniques. Although many important genes were isolated, it was unlkely that these methods alone would identify all transforming genes. This commentary briefly reviews recent studies, inchluding one published in this issue of the Britih Journal of Cancer from the laboratory of Dr CF Graham (Oxford, UK; see Bates et al., 1995), which implicate components of the insulin-like growth factor (IGF) signal transduction pathway in the malignant transformation of cells. Insulin-like growth factors (IGF-I and IGF-H) are involved in controlling normal growth and development and IGF-I in particula is involved in mediating the effects of growth hormone. They are thought to control cell growth and cell division pincipally through the type I IGF receptor, a heterotetrameic receptor located on the cell surface that has a similar structure to the insulin receptor. IGFs, but not insulin, also bind to a family of proteins called the IGF-binding proteins (IGFBPs) and the majority of IGFs in the circulation are present as a complex with IGFBP-3. Six IGFBPs have been described to date and although their function is not totally clear, they are generally thought to modulate the biological activity of IGFs. There has been increasing recognition during the past 5 years of the role of IGFs, type I IGF receptor and IGFBPs in the control of the proliferation of cancer cells. IGFs are potent mitogens for a wide variety of tumour cell types and have been identified as major autocrine or paracrne growth factors in a number of cancrs (Daughaday, 1990; Macauley, 1994). IGFs have also been implicated as mediators of the effects of steroids on the proliferation of hormone-responsive cancers, such as breast cancer (Westley and May, 1994). In addition, there has been some suggestion of the importance of circulating IGF levels on tumour growth and reports that therapeutic strategies may influence tumour cell growth by lowering circulating IGF levels (Pollak et al., 1990). Most experments to date have focused on the importance of IGFs in controlling the proliferation of previously transformed cells, however, evidence is now starting to accumulate that components of the IGF signal transduction system may play a role in the transformation process itself. The article by Bates et al. (1995) in this issue of the British Journal of …

The identification of genes involved in the transformation of cells proceeded rapidly during the 1970s and 1980s from the analysis of oncogenic retroviruses and the identification of oncogenes using DNA transfection techniques. Although many important genes were isolated, it was unlkely that these methods alone would identify all transforming genes. This commentary briefly reviews recent studies, inchluding one published in this issue of the Britih Journal of Cancer from the laboratory of Dr CF Graham (Oxford, UK; see Bates et al., 1995), which implicate components of the insulin-like growth factor (IGF) signal transduction pathway in the malignant transformation of cells.
Insulin-like growth factors (IGF-I and IGF-H) are involved in controlling normal growth and development and IGF-I in particula is involved in mediating the effects of growth hormone. They are thought to control cell growth and cell division pincipally through the type I IGF receptor, a heterotetrameic receptor located on the cell surface that has a similar structure to the insulin receptor. IGFs, but not insulin, also bind to a family of proteins called the IGFbinding proteins (IGFBPs) and the majority of IGFs in the circulation are present as a complex with IGFBP-3. Six IGFBPs have been described to date and although their function is not totally clear, they are generally thought to modulate the biological activity of IGFs.
There has been increasing recognition during the past 5 years of the role of IGFs, type I IGF receptor and IGFBPs in the control of the proliferation of cancer cells. IGFs are potent mitogens for a wide variety of tumour cell types and have been identified as major autocrine or paracrne growth factors in a number of cancrs (Daughaday, 1990;Macauley, 1994). IGFs have also been implicated as mediators of the effects of steroids on the proliferation of hormone-responsive cancers, such as breast cancer (Westley and May, 1994). In addition, there has been some suggestion of the importance of circulating IGF levels on tumour growth and reports that therapeutic strategies may influence tumour cell growth by lowering circulating IGF levels (Pollak et al., 1990).
Most experments to date have focused on the importance of IGFs in controlling the proliferation of previously transformed cells, however, evidence is now starting to accumulate that components of the IGF signal transduction system may play a role in the transformation process itself. The article by Bates et al. (1995) in this issue of the British Journal of Cancer contributes significntly to this debate by demonstrating that transgenc mie, in which expresson of insunlike growth factor II (IGF-II) was targeted to the mammary gland by plaing it under the control of the sheep Plactoglobulin promoter, develop an excess of mammary tumours.
The same group (Ward et al., 1994) had previously made IGF-H tansgenic mice, in which the IGF-H gene had been placed under the control of a keratin promoter that resut in elevated expression in skin, alimentary canal and uterus. Growth effects were observed in these organs, but no tumours formed in animals up to 9 months of age-Rogler et al. (1994) made transgenic mice in which IGF-ll expression was targeted to the liver under the control of the major urinary protein promoter. In this case, mice developed diverse tumours with a preponderance of heptocellular carcinoma. The studies of Bates et al. (1995) andRogler et al. (1994) are therefore in broad agreement and are the first in vivo experiments to suggest a direct role for IGFs in the malignant transformation of cells.
Although neither the IGFs nor the type I IGF receptor had been identified as oncogenes from the analysis of oncogenic viruses or by cell transformation assays, earlier stuies (Kaleko et al., 1990) had suggested that the type I IGF receptor could act as a transforming gene when overexpressed in NIH 3T3 cells. Interetingly, these eexperiments used the normal receptor and transformation was ligand dependent Tlhis contrasts with tansformation by other cellular oncogenes, for example the erbB/EGF receptor in which the ligand binding domain of the oncogenic form is deleted.
Evidence for the involvement of the IGF signal transduction pathway in cell transformation has also come from other avenues of research.
Prager et al. (1994) transfected cells with wild-type and a truncated a-subunit mutant of the type I IGF receptor (truncated at amino acid 952 to abolish tyrosine kinase activity). Cells overexpressing wild-type receptor showed increased ligand-dependent transformation. In contrast, cells transfected with truncated receptor were completely nonresponsive to IGF-1, were unable to sustain anchorageindependent growth and did not form tumours in nude mice these latter two features being characteristics of transformed cells. The truncated receptor therefore appeared to behave as a dominant negative inhibitor of endogenous type I IGF receptor and again emphasised the importance of this receptor in maintaining the transformed phenotype.
Two studies have shown that the IGF signal transduction pathway is involved in the transformation of cells by other oncogenes.
Studies on the mechnism by which the src oncogene transforms cells have identified the type I IGF receptor as a functionally significant substrate for pp60' (Peterson et al., 1994). Tyrosine phosphorylation of the type I IGF receptor correlates with cell transformation using a panel of partially defective src mutants. Phosphorylation of type I IGF receptor by src increases receptor tyrosine kinase activity both towards itself and exogenous substrates and these experiments are therefore consistent with the type I IGF receptor acting as an intermediary in the transformation of cells by src. , Christofori et al. (1994), studied -the induction of pancreatic tumours in transgenic mice expresng the sinuan virus-4 large T antigen under the control of the insulin gene regulatory region. In this system a high proportion of the islets become hyperplastic and vascularisation kads to tumours in 1-2% of islets. A survey of the expression of growth factors, receptors and oncogenes showed that IGF-fl expression is focally activated in a subset of islets and is further up-regulated in all tumours. This study concluded that IGF-H provides an important second signal in eliciting the hyperproliferation, which eventually leads to tumour for- the type I IGF receptor can act as a ligand-dependent oncogene and that expression of IGFs is important in tumorigenesis. Is there any epidemiological evidence or clinical studies that might suggest that individuals with local or systemic elevated IGF levels are at an increased risk of cancer? Stoll (1993) has recently reviewed the evidence that circulating levels of insulin and insulin-like growth factors are risk markers for breast cancer. Case-control studies have reported increased serum insulin (Bruning et al., 1992) and plasma IGF-I (Peyrat et al.. 1993) in women presenting with breast cancer. Increased circulating levels of insulin and IGF-I may be linked to other recognised risk markers for breast cancer including early onset of menarche. relative tallness and upper body type of obesity. Earlier onset of pubertal hyperinsulinaemia may be involved in earlier onset of ovulatory cycles because IGFs increase the effect of FSH in stimulating ovarian steroid synthesis (Garzo and Dorrington. 1984). Prospective studies (e.g. De Waard, 1975;London et al., 1989;Tretll. 1989) have reported an association between tallness and breast cancer risk. There is also an increased risk of breast cancer in women with upper (male type) obesity associated with high insulin levels (Schapira et al., 1990;Conover et al.. 1992).
The analysis of patients with acromegaly. a condition in which IGF-I levels are elevated as a result of increased secretion of growth hormone, may provide evidence for an increased risk of cancer from elevated IGF levels. Interestingly, Klein et al. (1982) showed a signficantly increased frequency of colonic polyps (frequently regarded as premalignant lesions) in patients with acromegaly and has suggested that this group of patients may have increased rates of colon cancer.
Is there any risk that the population at large is being systematically exposed to elevated levels of IGFs or that IGFs could be environmental carcinogens? Diet has been implicated in cancer risk and the association between tallness and breast cancer risk for example could reflect the increases in IGF-I levels in well-nourished individuals. Particular diets, however. may be rich in IGFs. IGFs are present in milk. being highest immediately post-partum and then decreasing gradually thereafter. Dairy products are therefore a potential dietary source of IGFs and these levels are higher in milk from cows treated with bovine somatomammotrophin (BST) to increase milk yield. Reassuringly. however. IGFs appear to be destroyed in the gastrointestinal tract and there is no evidence to suggest that the peculiarly human habit of ingesting milk in adulthood results in elevated systemic IGF levels (Juskevich and Guyer. 1990). It is however a formal possibility that ingested IGFs in dairy products could have a luminal site of action and the relationship between a diet rich in dairy products and cancers of the gastrointestinal tract could be of interest.
Although the evidence that components of the IGF signal transduction system can be involved in cell transformation is compelling. the mechanisms involved are not known. IGFs have been shown to inhibit the effect of myc on apoptosis (Harrington et al.. 1994) and Bates et al. (1995) suggest that the principle function of IGF-II is to increase cell survival thereby allowing somatic mutations to accumulate in other oncogenes. The studies of Prager et al. (1994) and Peterson et al. (1994) however are consistent with a more active role and suggest that the IGF growth factor signal transduction system itself could be responsible for increased proliferation and the malignant transformation of cells.