NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Cooper GM. The Cell: A Molecular Approach. 2nd edition. Sunderland (MA): Sinauer Associates; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of The Cell

The Cell: A Molecular Approach. 2nd edition.

Show details


Cancer results from alterations in critical regulatory genes that control cell proliferation, differentiation, and survival. Studies of tumor viruses revealed that specific genes (called oncogenes) are capable of inducing cell transformation, thereby providing the first insights into the molecular basis of cancer. However, the majority (approximately 80%) of human cancers are not induced by viruses and apparently arise from other causes, such as radiation and chemical carcinogens. Therefore, in terms of our overall understanding of cancer, it has been critically important that studies of viral oncogenes also led to the identification of cellular oncogenes, which are involved in the development of non-virus-induced cancers. The key link between viral and cellular oncogenes was provided by studies of the highly oncogenic retroviruses.

Retroviral Oncogenes

Viral oncogenes were first defined in RSV, which transforms chicken embryo fibroblasts in culture and induces large sarcomas within 1 to 2 weeks after inoculation into chickens (Figure 15.18). In contrast, the closely related avian leukosis virus (ALV) replicates in the same cells as RSV without inducing transformation. This difference in transforming potential suggested the possibility that RSV contains specific genetic information responsible for transformation of infected cells. A direct comparison of the genomes of RSV and ALV was consistent with this hypothesis: The genomic RNA of RSV is about 10 kb, whereas that of ALV is smaller, about 8.5 kb.

Figure 15.18. Cell transformation by RSV and ALV.

Figure 15.18

Cell transformation by RSV and ALV. Both RSV and ALV infect and replicate in chicken embryo fibroblasts, but only RSV induces cell transformation.

In the early 1970s, Peter Vogt and Steven Martin isolated both deletion mutants and temperature-sensitive mutants of RSV that were unable to induce transformation. Importantly, these mutants still replicated normally in infected cells, indicating that RSV contains genetic information that is required for transformation but not for virus replication. Further analysis demonstrated that both the deletion and temperature-sensitive RSV mutants define a single gene responsible for the ability of RSV to induce tumors in birds and transform fibroblasts in culture. Because RSV causes sarcomas, its oncogene was called src. The src gene is an addition to the genome of RSV; it is not present in ALV (Figure 15.19). It encodes a 60-kd protein that was the first protein-tyrosine kinase to be identified (see the Key Experiment in Chapter 13).

Figure 15.19. The RSV genome.

Figure 15.19

The RSV genome. RSV contains an additional gene, src, that is not present in ALV and encodes the Src protein-tyrosine kinase.

More than 40 different highly oncogenic retroviruses have been isolated from a variety of animals, including chickens, turkeys, mice, rats, cats, and monkeys. All of these viruses, like RSV, contain at least one oncogene (in some cases two) that is not required for virus replication but is responsible for cell transformation. In some cases, different viruses contain the same oncogenes, but more than two dozen distinct oncogenes have been identified among this group of viruses (Table 15.3). Like src, many of these genes (such as ras and raf) encode proteins that are now recognized as key components of signaling pathways that stimulate cell proliferation (see Figure 13.32).

Table 15.3. Retroviral Oncogenes.

Table 15.3

Retroviral Oncogenes.


An unexpected feature of retroviral oncogenes is their lack of involvement in virus replication. Since most viruses are streamlined to replicate as efficiently as possible, the existence of viral oncogenes that are not an integral part of the virus life cycle seems paradoxical. Scientists were thus led to question where the retroviral oncogenes had originated and how they had become incorporated into viral genomes—a line of investigation that ultimately led to the identification of cellular oncogenes in human cancers.

The first clue to the origin of oncogenes came from the way in which the highly oncogenic retroviruses were isolated. The isolation of Abelson leukemia virus is a typical example (Figure 15.20). More than 150 mice were inoculated with a nontransforming virus containing only the gag, pol, and env genes required for virus replication. One of these mice developed a lymphoma from which a new, highly oncogenic virus (Abelson leukemia virus), which now contained an oncogene (abl), was isolated. The scenario suggested the hypothesis that the retroviral oncogenes are derived from genes of the host cell, and that occasionally such a host cell gene becomes incorporated into a viral genome, yielding a new, highly oncogenic virus as the product of a virus-host recombination event.

Figure 15.20. Isolation of Abelson leukemia virus.

Figure 15.20

Isolation of Abelson leukemia virus. The highly oncogenic virus Ab-MuLV was isolated from a rare tumor that developed in a mouse that had been inoculated with a nontransforming virus (Moloney murine leukemia virus, or MuLV). MuLV contains only the gag (more...)

The critical prediction of this hypothesis was that normal cells contain genes that are closely related to the retroviral oncogenes. This was definitively demonstrated in 1976 by Harold Varmus, J. Michael Bishop, and their colleagues, who showed that a cDNA probe for the src oncogene of RSV hybridized to closely related sequences in the DNA of normal chicken cells. Moreover, src-related sequences were also found in normal DNAs of a wide range of other vertebrates (including humans) and thus appeared to be highly conserved in evolution. Similar experiments with probes for the oncogenes of other highly oncogenic retroviruses have yielded comparable results, and it is now firmly established that the retroviral oncogenes were derived from closely related genes of normal cells.

The normal-cell genes from which the retroviral oncogenes originated are called proto-oncogenes. They are important cell regulatory genes, in many cases encoding proteins that function in the signal transduction pathways controlling normal cell proliferation (e.g., src, ras, and raf). The oncogenes are abnormally expressed or mutated forms of the corresponding proto-oncogenes. As a consequence of such alterations, the oncogenes induce abnormal cell proliferation and tumor development.

An oncogene incorporated into a retroviral genome differs in several respects from the corresponding proto-oncogene. First, the viral oncogene is transcribed under the control of viral promoter and enhancer sequences, rather than being controlled by the normal transcriptional regulatory sequences of the proto-oncogene. Consequently, oncogenes are usually expressed at much higher levels than the proto-oncogenes and are sometimes transcribed in inappropriate cell types. In some cases, such abnormalities of gene expression are sufficient to convert a normally functioning proto-oncogene into an oncogene that drives cell transformation.

In addition to such alterations in gene expression, oncogenes frequently encode proteins that differ in structure and function from those encoded by their normal homologs. Many oncogenes, such as raf, are expressed as fusion proteins with viral sequences at the amino terminus (Figure 15.21). Recombination events leading to the generation of such fusion proteins often occur during the capture of proto-oncogenes by retroviruses, and sequences from both the amino and carboxy termini of proto-oncogenes are frequently deleted during the process. Such deletions may result in the loss of regulatory domains that control the activity of the proto-oncogene proteins, thereby generating oncogene proteins that function in an unregulated manner. For example, the viral raf oncogene encodes a fusion protein in which amino-terminal sequences of the normal Raf protein have been deleted. These amino-terminal sequences are critical to the normal regulation of Raf protein kinase activity, and their deletion results in unregulated constitutive activity of the oncogene-encoded Raf protein. This unregulated Raf activity drives cell proliferation, resulting in transformation.

Figure 15.21. The Raf oncogene protein.

Figure 15.21

The Raf oncogene protein. The Raf proto-oncogene protein consists of an amino-terminal regulatory domain and a carboxy-terminal protein kinase domain. In the viral Raf oncogene protein, the regulatory domain has been deleted and replaced by viral Gag (more...)

Many other oncogenes differ from the corresponding proto-oncogenes by point mutations, resulting in single amino acid substitutions in the oncogene products. In some cases, such amino acid substitutions (like the deletions already discussed) lead to unregulated activity of the oncogene proteins. An important example of such point mutations is provided by the ras oncogenes, which are discussed in the next section in terms of their role in human cancers.

Oncogenes in Human Cancer

Understanding the origin of retroviral oncogenes raised the question as to whether non-virus-induced tumors contain cellular oncogenes that were generated from proto-oncogenes by mutations or DNA rearrangements during tumor development. Direct evidence for the involvement of cellular oncogenes in human tumors was first obtained by gene transfer experiments in the laboratories of Robert Weinberg and of the author in 1981. DNA of a human bladder carcinoma was found to efficiently induce transformation of recipient mouse cells in culture, indicating that the human tumor contained a biologically active cellular oncogene (Figure 15.22). Both gene transfer assays and alternative experimental approaches have since led to the detection of active cellular oncogenes in human tumors of many different types (Table 15.4).

Figure 15.22. Detection of a human tumor oncogene by gene transfer.

Figure 15.22

Detection of a human tumor oncogene by gene transfer. DNA extracted from a human bladder carcinoma induced transformation of recipient mouse cells in culture. Transformation resulted from integration and expression of an oncogene derived from the human (more...)

Table 15.4. Representative Oncogenes of Human Tumors.

Table 15.4

Representative Oncogenes of Human Tumors.

Some of the oncogenes identified in human tumors are cellular homologs of oncogenes that were previously characterized in retroviruses, whereas others are new oncogenes first discovered in human cancers. The first human oncogene identified in gene transfer assays was subsequently identified as the human homolog of the rasH oncogene of Harvey sarcoma virus (see Table 15.3). Three closely related members of the ras gene family (rasH, rasK, and rasN) are the oncogenes most frequently encountered in human tumors. These genes are involved in approximately 20% of all human malignancies, including about 50% of colon and 25% of lung carcinomas.

The ras oncogenes are not present in normal cells; rather, they are generated in tumor cells as a consequence of mutations that occur during tumor development. The ras oncogenes differ from their proto-oncogenes by point mutations resulting in single amino acid substitutions at critical positions. The first such mutation discovered was the substitution of valine for glycine at position 12 (Figure 15.23). Other amino acid substitutions at position 12, as well as at positions 13 and 61, are also frequently encountered in ras oncogenes in human tumors. In animal models, it has been shown that mutations that convert ras proto-oncogenes to oncogenes are caused by chemical carcinogens, providing a direct link between the mutagenic action of carcinogens and cell transformation.

Figure 15.23. Point mutations in ras oncogenes.

Figure 15.23

Point mutations in ras oncogenes. A single nucleotide change, which alters codon 12 from GGC (Gly) to GTC (Val), is responsible for the transforming activity of the rasH oncogene detected in bladder carcinoma DNA.

As discussed in Chapter 13, the ras genes encode guanine-nucleotide binding proteins that function in transduction of mitogenic signals from a variety of growth factor receptors. The activity of the Ras proteins is controlled by GTP or GDP binding, such that they alternate between active (GTP-bound) and inactive (GDP-bound) states (see Figure 13.33). The mutations characteristic of ras oncogenes have the effect of maintaining the Ras proteins constitutively in the active GTP-bound conformation. In large part, this effect is a result of nullifying the response of oncogenic Ras proteins to GAP (GTPase-activating protein), which stimulates hydrolysis of bound GTP by normal Ras. Because of the resulting decrease in their intracellular GTPase activity, the oncogenic Ras proteins remain in the active GTP-bound state and drive unregulated cell proliferation.

Point mutations are only one of the ways in which proto-oncogenes are converted to oncogenes in human tumors. Many cancer cells display abnormalities in chromosome structure, including translocations, duplications, and deletions. The gene rearrangements resulting from chromosome translocations frequently lead to the generation of oncogenes. In some cases, analysis of these rearrangements has implicated already known oncogenes in tumor development. In other cases, novel oncogenes have been discovered by molecular cloning and analysis of rearranged DNA sequences.

The first characterized example of oncogene activation by chromosome translocation was the involvement of the c-myc oncogene in human Burkitt's lymphomas and mouse plasmacytomas, which are malignancies of antibody-producing B lymphocytes (Figure 15.24). Both of these tumors are characterized by chromosome translocations involving the genes that encode immunoglobulins. For example, virtually all Burkitt's lymphomas have translocations of a fragment of chromosome 8 to one of the immunoglobulin gene loci, which reside on chromosomes 2 ( κ light chain), 14 (heavy chain), and 22 ( λ light chain). The fact that the immunoglobulin genes are actively expressed in these tumors suggested that the translocations activate a proto-oncogene from chromosome 8 by inserting it into the immunoglobulin loci. This possibility was investigated by analysis of tumor DNAs with probes for known oncogenes, leading to the finding that the c-myc proto-oncogene was the chromosome 8 translocation break point in Burkitt's lymphomas. These translocations inserted c-myc into an immunoglobulin locus, where it was expressed in an unregulated manner. Such uncontrolled expression of the c-myc gene, which encodes a transcription factor normally induced in response to growth factor stimulation, is sufficient to drive cell proliferation and contribute to tumor development.

Figure 15.24. Translocation of c-myc.

Figure 15.24

Translocation of c-myc. The c-myc proto-oncogene is translocated from chromosome 8 to the immunoglobulin heavy-chain locus (IgH) on chromosome 14 in Burkitt's lymphomas, re-sulting in abnormal c-myc expression.

Translocations of other proto-oncogenes frequently result in rearrangements of coding sequences, leading to the formation of abnormal gene products. The prototype is translocation of the abl proto-oncogene from chromosome 9 to chromosome 22 in chronic myelogenous leukemia (Figure 15.25). This translocation leads to fusion of abl with its translocation partner, a gene called bcr, on chromosome 22. The result is production of a Bcr/Abl fusion protein in which the normal amino terminus of the Abl proto-oncogene protein has been replaced by Bcr amino acid sequences. The fusion of Bcr sequences results in unregulated activity of the Abl protein-tyrosine kinase, leading to cell transformation.

Figure 15.25. Translocation of abl.

Figure 15.25

Translocation of abl. The abl oncogene is translocated from chromosome 9 to chromosome 22, forming the Philadelphia chromosome in chronic myelogenous leukemias. The abl proto-oncogene, which contains two alternative first exons (1A and 1B), is joined (more...)

A distinct mechanism by which oncogenes are activated in human tumors is gene amplification, which results in elevated gene expression. Gene amplification (see Figure 5.54) is common in tumor cells, occurring more than a thousand times more frequently than in normal cells, and amplification of oncogenes may play a role in the progression of many tumors to more rapid growth and increasing malignancy. Indeed, novel oncogenes have been identified by molecular cloning and characterization of DNA sequences that are amplified in tumors.

A prominent example of oncogene amplification is the involvement of the N-myc gene, which is related to c-myc, in neuroblastoma (a childhood tumor of embryonal neuronal cells). Amplified copies of N-myc are frequently present in rapidly growing aggressive tumors, indicating that N-myc amplification is associated with the progression of neuroblastomas to increasing malignancy. Amplification of another oncogene, erbB-2, which encodes a receptor protein-tyrosine kinase, is similarly related to progression of breast and ovarian carcinomas.

Functions of Oncogene Products

The viral and cellular oncogenes have defined a large group of genes (about 100 in total) that can contribute to the abnormal behavior of malignant cells. As already noted, many of the proteins encoded by proto-oncogenes regulate normal cell proliferation; in these cases, the elevated expression or activity of the corresponding oncogene proteins drives the uncontrolled proliferation of cancer cells. Other oncogene products contribute to other aspects of the behavior of cancer cells, such as defective differentiation and failure to undergo programmed cell death.

The majority of oncogene proteins function as elements of the signaling pathways that regulate cell proliferation and survival in response to growth factor stimulation. These oncogene proteins include polypeptide growth factors, growth factor receptors, elements of intracellular signaling pathways, and transcription factors (Figure 15.26).

Figure 15.26. Oncogenes and signal transduction.

Figure 15.26

Oncogenes and signal transduction. Oncogene proteins act as growth factors (e.g., EGF), growth factor receptors (e.g., ErbB), and intracellular signaling molecules (Ras and Raf). Ras and Raf activate the ERK MAP kinase pathway (see Figures 13.32 and 13.35), (more...)

The action of growth factors as oncogene proteins results from their abnormal expression, leading to a situation in which a tumor cell produces a growth factor to which it also responds. The result is autocrine stimulation of the growth factor-producing cell (see Figure 15.9), which drives abnormal cell proliferation and contributes to the development of a wide variety of human tumors.

A large group of oncogenes encode growth factor receptors, most of which are protein-tyrosine kinases. These receptors are frequently converted to oncogene proteins by alterations of their amino-terminal domains, which would normally bind extracellular growth factors. For example, the receptor for platelet-derived growth factor (PDGF) is converted to an oncogene in some human leukemias by a chromosome translocation in which the normal amino terminus of the PDGF receptor is replaced by the amino terminal sequences of a transcription factor called Tel (Figure 15.27). The Tel sequences of the resulting Tel/PDGFR fusion protein dimerize in the absence of growth factor binding, resulting in constitutive activity of the intracellular kinase domain and unregulated production of a proliferative signal from the oncogene protein. Alternatively, genes that encode some receptor protein-tyrosine kinases, such as erbB-2, are activated by gene amplification. Other oncogenes (including src and abl) encode nonreceptor protein-tyrosine kinases that are constitutively activated by deletions or mutations of regulatory sequences.

Figure 15.27. Mechanism of Tel/PDGFR oncogene activation.

Figure 15.27

Mechanism of Tel/PDGFR oncogene activation. The normal PDGF receptor (PDGFR) is activated by dimerization induced by PDGF binding. The Tel/PDGFR oncogene encodes a fusion protein in which the normal extracellular domain of the PDGF receptor is replaced (more...)

The Ras proteins play a key role in mitogenic signaling by coupling growth factor receptors to activation of the Raf protein-serine/threonine kinase, which initiates a protein kinase cascade leading to activation of the ERK MAP kinase (see Figure 13.32). As discussed earlier, the mutations that convert ras proto-oncogenes to oncogenes result in constitutive Ras activity, which leads to activation of the MAP kinase pathway. The raf gene can similarly be converted to an oncogene by deletions that result in loss of the amino-terminal regulatory domain of the Raf protein (see Figure 15.21). These deletions result in unregulated activity of the Raf protein kinase, which also leads to constitutive MAP kinase activation.

The MAP kinase pathway ultimately leads to the phosphorylation of transcription factors and alterations in gene expression. As might therefore be expected, many oncogenes encode transcriptional regulatory proteins that are normally induced in response to growth factor stimulation. For example, transcription of the fos proto-oncogene is induced as a result of phosphorylation of Elk-1 by the ERK MAP kinase (see Figure 15.26). Fos and the product of another proto-oncogene, Jun, are components of the AP-1 transcription factor, which activates transcription of a number of target genes in growth factor-stimulated cells (Figure 15.28). Constitutive activity of AP-1, resulting from unregulated expression of either the Fos or Jun oncogene proteins, is sufficient to drive abnormal cell proliferation, leading to cell transformation. The Myc proteins similarly function as transcription factors regulated by mitogenic stimuli, and abnormal expression of myc oncogenes contributes to the development of a variety of human tumors. Other transcription factors are frequently activated as oncogenes by chromosome translocations in human leukemias and lymphomas.

Figure 15.28. The AP-1 transcription factor.

Figure 15.28

The AP-1 transcription factor. Fos and Jun dimerize to form AP-1, which activates transcription of a variety of growth factor-inducible genes.

G protein-coupled receptors and G proteins also act as oncogenes in some human tumors (Figure 15.29). For example, mutations of the gene encoding the thyrotropin receptor convert it to an oncogene in thyroid tumors. Thyrotropin is a pituitary hormone that stimulates proliferation of thyroid cells through a G protein-coupled receptor that activates adenylyl cyclase. Mutations of the thyrotropin receptor in thyroid tumors result in constitutive activity of the receptor, which then drives cell proliferation via activation of the cAMP signaling pathway. Likewise, the genes encoding G proteins can act as oncogenes in some cell types. The gsp oncogene, which encodes the α subunit of Gs, is generated by point mutations similar to those found in ras. These mutations result in constitutive activity of the Gs α subunit, leading to unregulated stimulation of adenylyl cyclase. As might be expected, the gsp oncogene is involved in thyroid and pituitary tumors, where cAMP stimulates cell proliferation. Similar mutations convert the genes encoding other G protein α subunits to oncogenes in other cell types, including adrenal and ovarian tumors.

Figure 15.29. Oncogenic activity of G protein-coupled receptors and G proteins.

Figure 15.29

Oncogenic activity of G protein-coupled receptors and G proteins. The thyrotropin receptor is coupled to adenylyl cyclase by Gs. The genes encoding both the receptor and the Gs α subunit (Gsα) can act as oncogenes by stimulating thyroid (more...)

The intracellular signaling pathways activated by growth factor stimulation ultimately regulate components of the cell cycle machinery that promote progression through the restriction point in G1. The D-type cyclins are induced in response to growth factor stimulation and play a key role in coupling growth factor signaling to cell cycle progression (see Figure 14.19). Perhaps not surprisingly, the gene encoding cyclin D1 is a proto-oncogene, which can be activated as an oncogene (called D1) by chromosome translocation or gene amplification. These alterations lead to constitutive expression of cyclin D1, which then drives cell proliferation in the absence of normal growth factor stimulation.

Although most oncogenes stimulate cell proliferation, the oncogenic activity of some transcription factors instead results from inhibition of cell differentiation. As noted in Chapter 13, thyroid hormone and retinoic acid induce differentiation of a variety of cell types. These hormones diffuse through the plasma membrane and bind to intracellular receptors that act as transcriptional regulatory molecules. Mutated forms of both the thyroid hormone receptor (ErbA) and the retinoic acid receptor (PML/RARα) act as oncogene proteins in chicken erythroleukemia and human acute promyelocytic leukemia, respectively. In both cases, the mutated oncogene receptors appear to interfere with the action of their normal homologs, thereby blocking cell differentiation and maintaining the leukemic cells in an actively proliferating state (Figure 15.30). In the case of acute promyelocytic leukemia, high doses of retinoic acid can overcome the effect of the PML/RARα oncogene protein and induce differentiation of the leukemic cells. This biological observation has a direct clinical correlate: Patients with acute promyelocytic leukemia can be treated effectively by administration of retinoic acid, which induces differentiation and blocks continued cell proliferation.

Figure 15.30. Action of the PML/RAR α oncogene protein.

Figure 15.30

Action of the PML/RAR α oncogene protein. The PML/RARα fusion protein blocks the differentiation of promyelocytes to granulocytes.

As discussed earlier in this chapter, the failure of cancer cells to undergo programmed cell death, or apoptosis, is a critical factor in tumor development, and several oncogenes encode proteins that act to promote cell survival (Figure 15.31). The survival of most animal cells is dependent on growth factor stimulation, so those oncogenes that encode growth factors, growth factor receptors, and signaling proteins such as Ras act not only to stimulate cell proliferation, but also to prevent cell death. As discussed in Chapter 13, the PI 3-kinase/Akt signaling pathway plays a key role in preventing apoptosis of many growth factor-dependent cells, and the genes encoding PI 3-kinase and Akt act as oncogenes in retroviruses and some human tumors. The downstream targets of PI 3-kinase/Akt signaling include a member of the Bcl-2 family (Bad), and it is notable that Bcl-2 was first discovered as an oncogene in human lymphomas. The bcl-2 oncogene is generated by a chromosome translocation that results in elevated expression of Bcl-2, which blocks apoptosis and maintains cell survival under conditions that normally induce cell death. The identification of bcl-2 as an oncogene not only provided the first demonstration of the significance of programmed cell death in the development of cancer, but also led to the discovery of the role of Bcl-2 and related genes as central regulators of apoptosis in organisms ranging from C. elegans to humans.

Figure 15.31. Oncogenes and cell survival.

Figure 15.31

Oncogenes and cell survival. The oncogene proteins that signal cell survival include growth factors, growth factor receptors, PI 3-kinase, and Akt. Signaling by the PI 3-kinase/Akt pathway regulates members of the Bcl-2 family, which promote cell survival (more...)

Box Icon


Key Experiment: The Discovery of Proto-Oncogenes.

Image ch13f32
Image ch13f33
Image ch5f54
Image ch13f35
Image ch15f9
Image ch6f27
Image ch14f19

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, Geoffrey M Cooper.
Bookshelf ID: NBK9840


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...