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

Griffiths AJF, Miller JH, Suzuki DT, et al. An Introduction to Genetic Analysis. 7th edition. New York: W. H. Freeman; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of An Introduction to Genetic Analysis

An Introduction to Genetic Analysis. 7th edition.

Show details

Cancer: the genetics of aberrant cell control

A basic article of faith in genetic analysis is that we learn a great deal about normal biology and about the disease state by studying the properties of mutations that disrupt normal processes. This has certainly been true in regard to cancer. It has become clear that virtually all cancers of somatic cells are due to a series of special mutations that accumulate in a cell. We are seeing that these mutations fall into a few major categories: increasing the ability of a cell to proliferate, decreasing the susceptibility of a cell to apoptosis, or increasing the general mutation rate of the cell so that proliferation or apoptotic mutation is more likely to occur. We can hope that these insights into the basic events in cancer biology will translate into improved diagnosis, treatment, and control of this major group of diseases.

How cancer cells differ from normal cells

Malignant tumors, or cancers, are clonal. Cancers are aggregates of cells, all derived from an initial aberrant founder cell that, although surrounded by normal tissue, is no longer integrated into that environment. Cancer cells often differ from their normal neighbors by a host of specific phenotypic changes, such as rapid division rate, invasion of new cellular territories, high metabolic rate, and abnormal shape. For example, when cells from normal epithelial cell sheets are placed in cell culture, they can grow only when anchored to the culture dish itself. In addition, normal epithelial cells in culture divide until they form a continuous monolayer. Then, they somehow recognize that they have formed a single epithelial sheet, and stop dividing. In contrast, malignant cells derived from epithelial tissue continue to proliferate, piling up on one another (Figure 22-17). Clearly, the factors regulating normal cell differentiation have been altered. What, then, is the underlying cause of cancer? Many different cell types can be converted into a malignant state. Is there a common theme to the ontogeny of these different types of cancer or do they each arise in quite different ways? Indeed, we can think about cancer in a general way: as occurring by the production of multiple mutations in a single cell that cause it to proliferate out of control. Some of those mutations may be transmitted from the parents through the germ line. Others arise de novo in the somatic cell lineage of a particular cell.

Figure 22-17. Scanning electron micrographs of (a) normal cells and (b) cells transformed with Rous sarcoma virus, a virus that infects cells with the src oncogene.

Figure 22-17

Scanning electron micrographs of (a) normal cells and (b) cells transformed with Rous sarcoma virus, a virus that infects cells with the src oncogene. (a) A normal cell line called 3T3. Note the organized monolayer structure of the cells. (b) A transformed (more...)

Evidence for the genetic origin of cancers

Several lines of evidence have pointed to a genetic origin for the transformation of cells from the benign into the cancerous state. Most carcinogenic agents (chemicals and radiation) are also mutagenic. There are occasional instances in which certain cancers are inherited as highly penetrant single Mendelian factors; an example is familial retinoblastoma. Perhaps representing the more general case are less penetrant susceptibility alleles that increase the probability of developing a particular type of cancer. In the past few years, several susceptibility genes have been recombinationally mapped and molecularly cloned and localized with the use of RFLP mapping or related techniques. Oncogenes, dominant mutant genes that contribute to cancer in animals, have been isolated from tumor viruses—viruses that can transform normal cells in certain animals into tumorforming cells. Such dominant oncogenes can also be isolated from tumor cells by using cell-culture assays that can distinguish between some types of benign and malignant cells. Tumors do not arise as a result of single genetic events but rather are the result of multiple-hit processes, in which several mutations must arise within a single cell for it to become cancerous. In some of the best-studied cases, the progression of colon cancer and astrocytoma (a brain cancer) has been shown to entail the sequential accumulation of several different mutations in the malignant cells (Figure 22-18). In the next sections, we shall further consider the genetic origin of cancers and the nature of the proteins that are altered by cancer-producing mutations. We shall see that many of these proteins take part in intercellular communication and the regulation of the cell cycle.

Figure 22-18. The multistep progression to malignancy in cancers of the colon and brain.

Figure 22-18

The multistep progression to malignancy in cancers of the colon and brain. Several histologically distinct stages can be distinguished in the progression of these tissues from the normal state to benign tumors to a malignant cancer. (a) A common sequence (more...)


Tumors arise through a series of sequential mutational events that lead to a state of uncontrolled proliferation.

Mutations in cancer cells

Two general kinds of mutations are associated with tumors: oncogene mutations and mutations in tumor suppressor genes. Oncogenes are mutated in such a way that the proteins that they encode are activated in tumor cells carrying the dominant mutant allele. A tumor cell will typically be heterozygous for an oncogene mutation and its normal allelic counterpart. Tumor-promoting mutant alleles of tumor-suppressor genes mutationally inactivate the proteins that they encode. For such mutations, the tumor cell will lack any copy of the corresponding wild-type allele; in essence, tumor-suppressor mutations that are found in a tumor cell are recessive.

How have tumor-promoting mutations been identified? Several approaches have been used. It is well known that certain types of cancer can “run in families.” With modern pedigree analysis techniques, familial tendencies toward certain kinds of cancer can be mapped relative to molecular markers such as microsatellites, and, in several cases, this mapping has led to the successful identification of the mutated genes. Cytogenetic analysis of tumor cells themselves also has proved invaluable. Many types of tumors are typified by characteristic chromosomal translocations or by deletions of particular chromosomal regions. In some cases, these chromosomal rearrangements are so reliably a part of a particular cancer that they can be used for diagnostic purposes. For example, 95 percent of patients with chronic myelogenous leukemia (CML) have a characteristic translocation between chromosomes 9 and 22. This translocation, called the Philadelphia chromosome after the city where this translocation was first described, is a critical part of the CML diagnosis. The Philadelphia chromosome will be considered in more detail later in this chapter. Other translocations characterize other sorts of tumors; diagnostic translocations are most often found associated with cancers of the white blood cells—leukemias and lymphomas. Not all tumor-promoting mutations are specific to a given type of cancer, however. Rather, the same mutations seem to be tumor promoting for a variety of cell types and thus are seen in many different cancers.


Tumor-promoting mutations can be identified in a variety of ways. When located, they can be cloned and studied to learn how they contribute to the malignant state.

It is obvious why mutations that increase the rate of cell proliferation cause tumors. It is not so immediately obvious why mutations that decrease the chances that a cell will undergo apoptosis cause them. The reason seems to be twofold: (1) a cell that cannot undergo apoptosis has a much longer lifetime within which to accumulate proliferation-promoting mutations and (2) the sorts of damage and unusual physiological changes that occur inside a tumor cell will ordinarily induce the self-destruction pathway to engage.

Whether an element of the cell cycle or the apoptosis pathway is due to a dominant oncogene mutation or to a recessive tumor-suppressor gene mutation is a function of how that normal protein contributes to the regulation of cell proliferation or programmed cell death (Table 22-1). Genes encoding proteins that positively control the cell cycle or block apoptosis can typically be mutated to become oncogenes; these tumor-promoting alleles are gain-of-function mutations. On the other hand, genes encoding proteins that negatively regulate the cell cycle or positively regulate apoptosis are found in the tumor-suppressor class; in these cases, the tumor-promoting alleles are loss-of-function mutations.

Table 22-1. Relations Between Function of the Wild-Type Protein Product and the Types of Tumor-Promoting Mutations That Can Arise in the Genes Encoding Those Products.

Table 22-1

Relations Between Function of the Wild-Type Protein Product and the Types of Tumor-Promoting Mutations That Can Arise in the Genes Encoding Those Products.

Classes of oncogenes

Roughly 100 different oncogenes have been identified (examples are given in Table 22-2). How do their normal counterparts, proto-oncogenes, function? Proto-oncogenes generally encode a class of proteins that are selectively active only when the proper regulatory signals allow them to be activated. As mentioned, many proto-oncogene products are elements of cell cycle positive control pathways, including growth-factor receptors, signal transduction proteins, and transcriptional regulators. Other proto-oncogene products function to negatively regulate the apoptotic pathway. However, in an oncogene mutation, the activity of the mutant oncoprotein has been uncoupled from the regulatory pathway that ought to be controlling its activation, leading to continuous unregulated expression of the oncoprotein (Figure 22-19). Several categories of oncogenes depict different ways in which the regulatory functions have been uncoupled. We will look at examples of some of them.

Table 22-2. Some Well-Characterized Oncogenes and the Proteins That They Encode.

Table 22-2

Some Well-Characterized Oncogenes and the Proteins That They Encode.

Figure 22-19. The Ras oncoprotein.

Figure 22-19

The Ras oncoprotein. (a) The ras oncogene differs from the wild type by a single base pair, producing a Ras oncoprotein that differs from the wild type in one amino acid, at position 12 in the ras open reading frame. (b) The effect of this missense mutation (more...)


Oncogenes encode oncoprotein-deregulated forms of proteins whose wild-type function is to participate in the positive control of the cell cycle or in the negative control of apoptosis.

Types of oncogene mutations

Point mutations.  

The change from normal protein to oncoprotein often includes structural modifications to the protein itself, such as those caused by simple point mutation. A single base-pair substitution that converts glycine into valine at amino acid number 12 of the Ras protein, for example, creates the oncoprotein found in human bladder cancer (Figure 22-19a). Recall that the normal Ras protein is a G-protein subunit that takes part in signal transduction and, as described earlier in this chapter, normally functions by cycling between the active GTP-bound state and the inactive GDP-bound state (see Figure 22-13). The amino acid change caused by the ras oncogene missense mutation produces an oncoprotein that always binds GTP (Figure 22-19b), even in the absence of the normal signals such as phosphorylation of Ras, required for such binding by a wild-type Ras protein. In this way, the Ras oncoprotein continually propagates a signal that promotes cell proliferation.

Loss of protein domains.  

Structural alterations can also be due to the deletion of parts of a protein. The v-erbB oncogene encodes a mutated form of an RTK known as the EGFR, a receptor for the epidermal growth factor (EGF) ligand (Figure 22-20). The mutant form of the EGFR lacks the extracellular, ligand-binding domain as well as some regulatory components of the cytoplasmic domain. The result of these deletions is that the truncated v-erbB-encoded EGFR oncoprotein is able to dimerize even in the absence of the EGF ligand. The constitutive EGFR oncoprotein dimer is always autophosphorylated through its tyrosine kinase activity and thus continuously initiates a signal transduction cascade.

Figure 22-20. An oncogenic mutation affecting signaling between cells, EGFR, the normal receptor for epidermal growth factor (EGF), has a ligand-binding domain outside the cell, a transmembrane (TM) domain that allows the protein to span the plasma membrane, and an intracellular domain that has tyrosine-specific protein kinase activity.

Figure 22-20

An oncogenic mutation affecting signaling between cells, EGFR, the normal receptor for epidermal growth factor (EGF), has a ligand-binding domain outside the cell, a transmembrane (TM) domain that allows the protein to span the plasma membrane, and an (more...)

Gene fusions

Perhaps the most remarkable type of structurally altered oncoprotein is one caused by a gene fusion. The classic example of fused genes emerged from studies of the Philadelphia chromosome, which, as already mentioned, is a translocation between chromosomes 9 and 22 that is a diagnostic feature of chronic myelogenous leukemia (CML). Recombinant DNA methods have shown that the breakpoints of the Philadelphia chromosome translocation in different CML patients are quite similar and cause the fusion of two genes, bcr1 and abl (Figure 22-21). The abl proto-oncogene encodes a cytoplasmic tyrosine-specific protein kinase. The Brc1-Abl fusion oncoprotein has an activated protein kinase activity that is responsible for its oncogenic state.

Figure 22-21. The chromosome rearrangement in chronic myelogenous leukemia.

Figure 22-21

The chromosome rearrangement in chronic myelogenous leukemia. The Philadelphia chromosome, which is diagnostic of CML, is a translocation between chromosomes 9 and 22. The translocation breakpoints are in the middle of the c-abl gene, which encodes a (more...)

Some oncogenes produce an oncoprotein that is identical in structure with the normal protein. In these cases, the oncogene mutation induces misexpression of the protein—that is, it is expressed in cell types from which it is ordinarily absent. Several oncogenes that cause misexpression are also associated with chromosomal translocations diagnostic of various B-lymphocyte tumors. B lymphocytes and their descendants, plasma cells, are the cells that synthesize antibodies, or immunoglobulins. In these B-cell oncogene translocations, no protein fusion is produced; rather, the chromosomal rearrangement causes a gene near one breakpoint to be turned on in the wrong tissue. In follicular lymphoma, 85 percent of patients have a translocation between chromosomes 14 and 18 (Figure 22-22). Near the chromosome 14 breakpoint is located a transcriptional enhancer from one of the immunoglobulin genes. This translocated enhancer element is fused to the bcl2 gene, which is a negative regulator of apoptosis. This enhancer–bcl2 fusion causes large amounts of Bcl2 to be expressed in B lymphocytes. These large amounts of Bcl2 essentially block apoptosis in these mutant B lymphocytes and provide them with an unusually long lifetime in which to accumulate cell proliferation-promoting mutations. There are strong parallels between this sort of dominant oncogene mutation and the dominant gain-of-function phenotypes caused by the fusion of the enhancer of one gene to the transcription unit of another in producing the Tab allele of the Abd-B gene (see Chapter 23). In each case, the introduction of an enhancer causes a dominant gain-of-function phenotype by misregulation of the transcription unit. Mutations such as Tab arise in the germ line and are transmitted from one generation to the next, whereas most oncogene mutations arise in somatic cells and are not inherited by offspring.

Figure 22-22. The chromosomal rearrangement in follicular lymphoma.

Figure 22-22

The chromosomal rearrangement in follicular lymphoma. The translocation fuses the transcriptional enhancer element of a gene, on chromosome 14, that makes one protein subunit of an antibody (the IgH subunit, also called the immunoglobulin heavy chain) (more...)


Dominant oncogenes contribute to the oncogenic state by causing a protein to be expressed in an activated form or in the wrong cells.

Classes of tumor-suppressor genes

The normal functions of tumor-suppressor genes fall into categories complementary to those of proto-oncogenes (see Table 22-1). Some tumor-suppressor genes encode negative regulators of the cell cycle, such as the Rb protein or elements of the TGF-β signaling pathway. Others encode positive regulators of apoptosis (at least part of the function of p53 falls into this category). Still others act indirectly, through a general elevation in the mutation rate. We shall consider two examples here.

Inheritance of the tumor phenotype

In retinoblastoma, the gene encoding the Rb protein, considered in the regulation of the cell cycle, is mutated. In retinoblastoma, a cancer typically affecting young children, retinal cells lacking a functional RB gene proliferate out of control. These rb null cells are either homozygous for a single mutant rb allele or are heterozygous for two different rb mutations. Most patients have one or a few tumors localized to one site in one eye, and the condition is sporadic—in other words, there is no history of retinoblastoma in the family and the affected person does not transmit it to his or her offspring. Retinoblastoma is not transmitted in this case, because the rb mutation or mutations that inactivate both alleles of this autosomal gene arise in a somatic cell whose descendants populate the retina (Figure 22-23). Presumably, the mutations arise by chance at different times in development in the same cell lineage.

Figure 22-23. (a) Retinoblastoma, a cancer of the retina.

Figure 22-23

(a) Retinoblastoma, a cancer of the retina. (b) The mutational origin of retinal tumors in hereditary and sporadic retinoblastoma. Recessive rb alleles of the RB gene lead to tumor development. (Part a from Custom Medical Stock.)

A few patients, however, have an inherited form of the disease, called hereditary binocular retinoblastoma (HBR). Such patients have many tumors, and the retinas of both eyes are affected. Paradoxically, even though rb is a recessive trait at the cellular level, the transmission of HBR is as an autosomal dominant (Figure 22-23). How do we resolve this paradox? In the presence of a germ-line mutation that knocks out one of the two copies of the RB gene, the mutation rate for RB makes it virtually certain that at least some of the retinal cells of patients with HBR will have acquired an rb mutation in the single remaining normal RB gene, thereby producing cells with no functional Rb protein.

Why does the absence of RB promote tumor growth? Recall from our consideration of the cell cycle that Rb protein functions by binding the E2F transcription factor. Bound Rb prevents E2F from promoting the transcription of genes whose products are needed for S-phase functions such as DNA replication. An inactive Rb is unable to bind E2F, and so E2F can promote the transcription of S-phase genes. In homozygous null rb cells, Rb protein is permanently inactive. Thus, E2F is always able to promote S phase, and the arrest of normal cells in late G1 does not occur in retinoblastoma cells.

p53 tumor-suppressor gene: a link between the cell cycle and apoptosis

Another very important recessive tumor-promoting mutation has identified the p53 gene as a tumor-suppressor gene. Mutations in p53 are associated with many types of tumors, and estimates are that 50 percent of human tumors lack a functional p53 gene. The active p53 protein is a transcriptional regulator that is activated in response to DNA damage. Activated wild-type p53 serves double duty, preventing progression of the cell cycle until the DNA damage is repaired and, under some circumstances, inducing apoptosis. In the absence of a functional p53 gene, the p53 apoptosis pathway does not become activated, and the cell cycle progresses even in the absence of DNA repair. This progression elevates the overall frequency of mutations, chromosomal rearrangements, and aneuploidy and thus increases the chances that other mutations promoting cell proliferation or blocking apoptosis will arise. Other recessive tumor-promoting genes that have been identified also are implicated in the repair of DNA damage. Research suggests that genes that, when inactivated, produce the phenotype of elevated mutation rates are very important contributors to the progression of tumors in humans. Such recessive tumor-suppressor mutations that interfere with DNA repair promote tumor growth indirectly, because their elevated mutation rates make it much more likely that a series of oncogene and tumor-suppressor gene mutations will arise, corrupting the normal regulation of the cell cycle and programmed cell death.


Mutations in tumor-suppressor genes, like mutations in oncogenes, act directly or indirectly to promote the cell cycle or block apoptosis.

Complexities of cancer

As discussed in this chapter, numerous mutations that promote tumor growth can arise. These mutations are thematically related and can be understood in relation to the ways in which they alter the normal processes that govern proliferation and apoptosis (Figure 22-24). In some instances, such as colon cancer (Figure 22-18), we are even able to identify a series of independent mutations that contribute to the progression of a cell from a normal state through various stages of a benign tumor to a truly malignant state. The story does not stop there, however. Even among malignant tumors, their rates of proliferation and their abilities to invade other tissues, or metastasize, are quite different. Undoubtedly, even after a malignant state is achieved, more mutations accumulate in the tumor cell that further promote its proliferation and invasiveness. Thus, there is a considerable way to go before we have a truly comprehensive view of how tumors arise and progress.

Figure 22-24. The major pathways that are mutated to contribute to cancer formation and progression.

Figure 22-24

The major pathways that are mutated to contribute to cancer formation and progression. The main events that contribute to tumor formation are increased cell proliferation and cell survival (decreased apoptosis). The pathways in red are susceptible to (more...)

Image ch22f13
Image ch22f14

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

Copyright © 2000, W. H. Freeman and Company.
Bookshelf ID: NBK21896


  • Cite this Page
  • Disable Glossary Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...