5. Insertion, translocation, or amplification may activate proto-oncogenes

28.4 Ras proto-oncogenes can be activated by mutation

Figure 28.9 The transfection assay allows (some) oncogenes to be isolated directly by assaying DNA of tumor cells for the ability to transform normal cells into tumorigenic cells.

Some oncogenes can be detected by using a direct assay for transformation in which "normal" recipient cells are transfected with DNA obtained from animal tumors. The procedure is illustrated in Figure 28.9. (Actually the established mouse NIH 3T3 fibroblast line usually is used as recipient. Historically these experiments started by using DNA extracted en masse, but now of course they are usually performed with a purified oncogene.) The ability of any individual gene to convert wild-type cells into the transformed state constitutes one form of proof that it is an oncogene. Another assay that can be used is to inject cells into "nude" mice (which lack the ability to reject such transplants immunologically). The ability to form tumors can then be measured directly in the animal.

When a cell is transformed in a 3T3 culture (or some other "normal" culture), its descendants pile up into a focus. The appearance of foci is used as a measure of the transforming ability of a DNA preparation. Starting with a preparation of DNA isolated from tumor cells, the efficiency of focus formation is low. However, once the transforming gene has been isolated and cloned, greater efficiencies can be obtained. In fact, the transforming "strength" of a gene can be characterized by the efficiency of focus formation by the cloned sequence.

DNA with transforming activity can be isolated only from tumorigenic cells; it is not present in normal DNA. The transforming genes isolated by this assay have two revealing properties:

• They have closely related sequences in the DNA of normal cells. This argues that transformation was caused by mutation of a normal cellular gene (a proto-oncogene) to generate an oncogene. The change may take the form of a point mutation or more extensive reorganization of DNA around the c-onc gene.
  
• They may have counterparts in the oncogenes carried by known transforming viruses. This suggests that the repertoire of proto-oncogenes is limited, and probably the same genes are targets for mutations to generate oncogenes in the cellular genome or to become viral oncogenes.

Oncogenes derived from the c-ras family are often detected in the transfection assay. The family consists of several active genes in both man and rat, dispersed in the genome. (There are also some pseudogenes.) The individual genes, N-ras, H-ras, and K-ras, are closely related, and code for protein products ~21 kD and known collectively as p21^ras.

Figure 28.8 Each transforming retrovirus carries an oncogene derived from a cellular gene. Viruses have names and abbreviations reflecting the history of their isolation and the types of tumor they cause. This list shows some representative examples of the retroviral oncogenes

The H-ras and K-ras genes have v-ras counterparts, carried by the Harvey and Kirsten strains of murine sarcoma virus, respectively (see Figure 28.8). Each v-ras gene is closely related to the corresponding c-ras gene, with only a few individual amino acid substitutions. The Harvey and Kirsten virus strains must have originated in independent recombination events in which a progenitor virus gained the corresponding c-ras sequence.

Oncogenic variants of the c-ras genes are found in transforming DNA preparations obtained from various primary tumors and tumor cell lines. Each of the c-ras proto-oncogenes can give rise to a transforming oncogene by a single base mutation. The mutations in several independent human tumors cause substitution of a single amino acid, most commonly at position 12 or 61, in one of the Ras proteins.

Position 12 is one of the residues that is mutated in the v-H-ras and v-K-ras genes. So mutations occur at the same positions in v-ras genes in retroviruses and in mutant c-ras genes in multiple rat and human tumors. This suggests that the normal c-Ras protein can be converted into a tumorigenic form by a mutation in one of a few codons in rat or man (and perhaps any mammal).

The general principle established by this work is that substitution in the coding sequence can convert a cellular proto-oncogene into an oncogene. Such an oncogene can be associated with the appearance of a spontaneous tumor in the organism. It may also be carried by a retrovirus, in which case a tumor is induced by viral infection.

The ras genes appear to be finely balanced at the edge of oncogenesis. Almost any mutation at either position 12 or 61 can convert a c-ras proto-oncogene into an active oncogene. All three c-ras genes have glycine at position 12. If it is replaced in vitro by any other of the 19 amino acids except proline, the mutated c-ras gene can transform cultured cells. The particular substitution influences the strength of the transforming ability.

Position 61 is occupied by glutamine in wild-type c-ras genes. Its change to another amino acid usually creates a gene with transforming potential. Some substitutions are less effective than others; proline and glutamic acid are the only substitutions that have no effect.

When the expression of a normal c-ras gene is increased, either by placing it under control of a more active promoter or by introducing multiple copies into transfected cells, recipient cells are transformed. Some mutant c-ras genes that have changes in the protein sequence also have a mutation in an intron that increases the level of expression (by increasing processing of mRNA ~10 ). Also, some tumor lines have amplified ras genes. A 20-fold increase in the level of a nontransforming Ras protein is sufficient to allow the transformation of some cells. The effect has not been fully quantitated, but it suggests the general conclusion that oncogenesis depends on over-activity of Ras protein, and is caused either by increasing the amount of protein or (more efficiently) by mutations that increase the activity of the protein (for review see Barbacid, 1987).

Transfection by DNA can be used to transform only certain cell types. Limitations of the assay explain why relatively few oncogenes have been detected by transfection. This system has been most effective with ras genes, where there is extensive correlation between mutations that activate c-ras genes in transfection and the occurrence of tumors.

Figure 28.10 Pathways that rely on Ras could function by controlling either GNRF or GAP. Oncogenic Ras mutants are refractory to control, because Ras remains in the active form.

Ras is a monomeric guanine nucleotide-binding protein that is active when bound to GTP and inactive when bound to GDP. It has an intrinsic GTPase activity. Figure 28.10 reviews the discussion of 26 Signal transduction in which we saw that the conversion between the two forms of Ras is catalyzed by other proteins. GAP proteins stimulate the ability of Ras to hydrolyze GTP, thus converting active Ras into inactive Ras. GEF proteins stimulate the replacement of GDP by GTP, thus reactivating the protein.

Constitutive activation of Ras could be caused by mutations that allow the GDP-bound form of Ras to be active or that prevent hydrolysis of GTP. What are the effects of the mutations that create oncogenic ras genes? Many mutations that confer transforming activity inhibit the GTPase activity. GAP cannot increase the GTPase activity of Ras proteins that have been activated by oncogenic mutations. In other words, Ras has become refractory to the interaction with GAP that turns off its activity. Inability to hydrolyze GTP causes Ras to remain in a permanently activated form; its continued action upon its target protein is responsible for its oncogenic activity (for review see Lowy, 1993).

This establishes an important principle: constitutive activation of a cellular protein may be oncogenic. In the case of Ras, its effects result from activating the ERK MAP kinase pathway and (possibly) other pathways. The level of expression is finely balanced, since overstimulation of Ras by either increase in expression or mutation of the protein has oncogenic consequences (although mutation is required for a full effect).

Reviews

Barbacid, M. (1987). Ras genes. Ann. Rev. Biochem 56, 779-827.

Lowy, D. R. (1993). Function and regulation of Ras. Ann. Rev. Biochem 62, 851-891.