Proto-oncogenes are the genes which are most frequently found amplified in human tumor cells. Acquisition of a drug-resistant phenotype by gene amplification is frequent for in-vitro cultured cells but is very rare in human tumors. Proto-oncogenes amplified in human tumors belong essentially to one of three families (erbB, ras, myc) or to the 11q13 locus. Amplification is always specific for the tumor cells and is not found in constitutional DNA of the patient, indicating that amplification of the gene is selected for during tumor growth. For genes of the first three families, amplification results in overexpression in most of the cases. These are strong arguments in favor of a role of this amplification in tumor progression. The gene whose overexpression is the driving force for the selection of the amplification of the 11q13 locus is not known. The prad1 gene is presently a good candidate. Amplification of one type of proto-oncogene is generally not restricted to one tumor type. However, the N-myc gene is amplified mainly in tumors of neuronal or neuroendocrine origin and L-myc amplification is restricted to lung carcinomas. To understand the role of proto-oncogene amplification and overexpression in tumor progression it is necessary to know the function of the corresponding protein in the cell. erbB proteins are transmembrane receptors for growth factors. ras genes encode small GTP-binding proteins which are possibly involved in signal transduction. The myc proteins are transcription factors. The expression of the c-myc gene is induced a few hours after cells of various types have been induced to proliferate. The genes of these three families therefore encode proteins which appear to be involved in signal transduction. It is possible that overexpression of one of them, as a result of gene amplification, makes the cell a better responder to low levels of growth stimuli. For several genes which are found amplified in human tumors, it was shown that overexpression of the normal protein could confer a transformed or tumorigenic phenotype to in-vitro cultured cells. In addition, several studies on animal and human tumor-derived cell lines with an amplified proto-oncogene have established a relationship between proto-oncogene amplification and the tumorigenic phenotype. In neuroblastomas, it was proposed that down-modulation of MHC Class I antigens is a consequence of N-myc amplification and that this could be important in the progression toward a metastatic phenotype. Several studies have concluded that proto-oncogene amplification is significantly more frequent in large tumors with a low level of differentiation and which have already metastasized. Such tumors may be considered in a more advanced stage of progression. This suggests that proto-oncogene amplification is a late event in tumor progression. Studies on the occurrence of proto-oncogene amplification is experimentally induced animal tumors reach the same conclusion. Why would proto-oncogene amplification be a late event in tumor progression? First, it is possible that amplification and overexpression of members of the erbB, ras and myc families or of the putative 11q13 gene is advantageous only for cells which have already progressed up to a point where they have escaped from several of the growth control mechanisms and are already in a proliferating stage. There is presently little experimental evidence to support this hypothesis. A second possibility is that only cells at a late stage of tumor progression are able to amplify DNA sequences. Indeed, several authors have shown that gene amplification cannot be detected in normal somatic cells and that the ability to amplify regions of the genome increases when cells acquire a transformed or tumorigenic phenotype. In addition, the acquisition of gene amplification by some cells is tightly coupled with the loss of wild-type p53 alleles which is a late event in tumor progression. © 1989.
