Molecular Biology of the Cell - Volume 3 - Alberts B., Bray D., Lewis J., Raff M., Roberts K., Watson J. 1994

From Cells to Multicellular Organisms
Cancer
Molecular Genetics of Cancer

Since cancer arises from a series of random genetic events, it is highly unlikely that any two tumors, even of the same type, would be genetically identical. Nevertheless, one would expect any form of cancer to disrupt normal constraints on Cell proliferation, and for any given cell type, There is a finite number of potential ways this disruption can manifest. Moreover, certain elements of The Cell-division machinery appear to be identical across many or even all cell types, rendering them uniformly vulnerable. In fact, the primary contribution to deregulated Cell Division in cancer comes from a relatively small number of genes. Identifying and characterizing many of these genes stands as one of the greatest achievements of molecular biology over the past decade. Cell proliferation can be regulated directly—via mechanisms that trigger a cell to enter a new division cycle (see Section 13.3.2)—or indirectly, for instance, by regulating a cell's entry into terminal differentiation (see Section 17.4.1). In both cases, normal regulatory genes fall into two categories: those whose products promote cell proliferation, and those whose products inhibit it. Accordingly, There are two Selection/21.html">Types of Mutations that lead to uncontrolled proliferation, the hallmark of cancer Cells. Mutations of the first type result in the hyperactivity of a "stimulatory" Gene; these are dominant (a mutation in just one of the two cellular copies of such a gene is sufficient to manifest the effect), and the altered gene is termed an oncogene (its normal allele being a proto-oncogene). Mutations of the second type lead to the inactivation of an "inhibitory" gene; these mutations are recessive—both cellular copies of the gene must be inactivated or deleted to release the cell from inhibitory control. This lost gene is sometimes referred to as a tumor suppressor gene. Beyond standard mutations, Other types of genetic alterations can also drive cancer development: the cell-division control system can be subverted by foreign DNA introduced into the cell by a virus. Indeed, The Study of cancer Molecular Genetics began with the discovery of such tumor-causing Viruses, paving the way for the subsequent identification of oncogenes and Proto-oncogenes. Progress in solving the other—and more formidable—challenge of identifying and cloning tumor suppressor genes has been achieved relatively recently (which will be discussed at the end of this section).

21.2.1. Tumors Can Be Caused by Both DNA and RNA Viruses [15, 16]

Both DNA and RNA viruses (particularly Retroviruses) can participate in the malignant Transformation of a normal cell. This can be demonstrated experimentally both in laboratory animals (in which certain viruses can induce cancer) and in cell culture, where the same viruses alter The behavior of infected cells. These cells acquire the capacity to divide under conditions where normal cells cannot (see Section 13.4.1 for a list of The properties of neoplastically transformed cells in culture). Two intensively studied Examples of such viruses are SV40 (a DNA virus isolated from monkey cells) and Rous Sarcoma virus, an avian retrovirus. The Role of viruses in human cancer is more complex; among the multitude of causes leading to virtually all known human cancers, viruses do not figure prominently. It is possible that we are protected from many virus-induced tumors by The Immune System, which destroys virus-infected cells that might otherwise serve as a source of tumors. Nevertheless, compelling evidence now indicates that certain types of human cancer are caused by viruses (Table 21-3). They may exert either an indirect promoting effect or actively contribute to the neoplastic transformation of infected cells.

In a stably virus-transformed cell, a balance must be established: the virus must not kill the cell, and the cell must retain the viral genes during reproduction (passing them from generation to generation). This is usually accomplished by integrating these genes into one or more cellular Chromosomes. However, viral genes sometimes persist in The Cell as Plasmids that replicate synchronously with the chromosomes. For example, SV40 and retroviruses integrate into the chromosomes of the cells they transform; papillomaviruses—a Class of DNA viruses that cause warts in humans (and likely contribute to Cervical cancer)—exist as plasmids under certain conditions and as integrated elements under others. In both cases, they "bestow" a hereditarily altered genome upon the cell. However, DNA viruses and retroviruses differ in The Nature of the genes that cause neoplastic transformation and in the stage of the normal viral life cycle at which transformation takes place.

Table 21-3. Viruses Associated with Human Cancers

For all of these viruses, the number of infected individuals far exceeds the number of people who develop cancer, indicating that viruses must act in conjunction with other factors. Moreover, THE CONTRIBUTION OF certain viruses to cancer development is merely indirect: the AIDS virus, for example, impairs cellular immune defenses. As a result, endothelial cells that have undergone transformation by other agents escape destruction by the immune system and form a tumor.____________________________________

21.2.2. Disruption of Cell Division Control by DNA Tumor Viruses as Part of Their Survival Strategy [15, 17]

As noted in Chapter 5 (Section 5.5.3), the propagation of tumor-causing DNA viruses, such as SV40, does not naturally lead to cancer under normal conditions. Upon entering a host cell, SV40 typically does not integrate firmly into the cellular genome. Instead, a protein (or group of Proteins) encoded by the viral genome rapidly activates the cellular METABOLISM/36.html">DNA Replication machinery, which the virus then hijacks to replicate its own DNA. This viral DNA, in turn, serves as a template for synthesizing other viral components at the expense of the host cell. This production of Viral Particles continues until the cell dies, releasing a multitude of progeny viruses. Much less frequently, the virus ends up in a cell unsuited for its replication, where it may persist indefinitely as a result of stable integration into one or more host chromosomes. In this scenario, the viral gene responsible for activating cellular DNA replication can also be transcribed, thereby pushing the cell to enter the S phase of the Cell Cycle AND driving it through endless rounds of division. The viral gene begins to function as an oncogene, inducing tumor transformation. However, such an oncogene differs fundamentally from the classes of oncogenes we will examine later—it has no homolog in the normal cellular genome.

Fig. 21-20. Putative mechanism by which certain papillomaviruses can induce cervical cancer. These viruses possess a circular double-stranded DNA genome of approximately 8,000 Base Pairs. In warts and other benign growths, the viral chromosomes stably persist as independently replicating plasmids. Through a rare accidental event, the viral genome may integrate into a host chromosome, altering the local genomic environment of the viral genes and disrupting their expression control. Unregulated Synthesis of the viral replication protein "pushes" the cell into the S phase of the cell cycle, thereby promoting tumorigenesis.

DNA viruses are a remarkably diverse group, but the General Principles outlined above apply, with minor variations, to most of those involved in cancer Pathogenesis. An example of one such variation is found in the papillomaviruses, where a persistent association with the host Organism's cells is an integral part of their life cycle. Papillomaviruses, like SV40, belong to the papovavirus family, but they appear capable of switching between "non-productive" (lysogenic) and "productive" (lytic) infections, and vice versa. In the former case, the virus replicates synchronously with the cell without causing harm; in the latter, it rapidly multiplies and kills (lyses) the cell, releasing a flood of new viral particles capable of infecting other cells. Like SV40, these viruses can "commandeer" the cellular DNA Synthesis machinery, and the viral genes executing this function can act as oncogenes. Figure 21-21 illustrates how papillomaviruses most likely participate in human cervical carcinogenesis.

21.2.3. Retroviruses Can Accurately Capture Oncogenes [15, 18]

Unlike DNA viruses, most retroviruses (see Sections 5.5.8 and 13.4.2) are relatively harmless to the host cell. An infected cell continuously sheds new viral particles that bud off from The Plasma Membrane without undergoing neoplastic transformation. On rare occasions, however, a retrovirus may accidentally "capture" or acquire a regulatory cellular gene (or a mutated copy/fragment thereof) that is non-essential for the virus's own life cycle but can profoundly influence The Fate of the host cell. In particular, as we saw in Chapter 13 (Section 13.4.2), a retrovirus like Rous sarcoma virus, having acquired a cellular oncogene (Fig. 21-21), is easily detected by its dominant transforming effect on infected host cells, which begin to proliferate vigorously. The oncogene can be identified and isolated using molecular genetics Methods and used to synthesize DNA probes for detecting homologous genes in normal cells. This approach has led to the isolation of more than 20 oncogenes homologous to proto-oncogenes (Table 21-4) found in The Genome of every normal vertebrate cell (see Section 13.4.2). These oncogenes belong to several distinct families (see Section 13.4.4), the largest of which is the protein kinase family, which includes the Rous sarcoma virus src oncogene (Fig. 21-22).

Fig. 21-21. Structure OF THE Rous sarcoma virus. A. Features of Genome Organization compared with a more typical retrovirus (murine leukemia virus). Rous sarcoma virus differs from other oncogenic retroviruses in that it has retained all three genes required for its life cycle: gag, which encodes a polyprotein that is cleaved to form capsid proteins; pol, which encodes Reverse Transcriptase and the enzyme mediating viral chromosomal integration into the host genome; and env, which encodes the envelope glycoprotein. In other oncogenic retroviruses, one or more of these viral genes are absent because the virus has acquired a transforming oncogene instead (see Fig. 21-22). Consequently, such transforming viruses can only produce infectious particles if the cell is simultaneously co-infected with a non-defective, non-transforming helper virus that supplies the missing Functions. B. Relationship between the v-src oncogene and the cellular proto-oncogene c-src from which it originated. The introns present in the c-src gene are absent from v-src; furthermore, v-src carries mutations that alter the Amino Acid Sequence of the v-src protein, converting it into a hyperactive, unregulated Tyrosine-specific protein kinase (see Section 13.4.5). Cancer researchers have subjected Rous sarcoma virus to rigorous selection for its cell-transforming capacity. Modern strains accomplish this with extraordinary speed and efficiency.

There are two mechanisms by which a proto-oncogene is converted into an oncogene upon incorporation into a retrovirus: sequence alteration or fragmentation of the gene, resulting in the synthesis of a protein with aberrant activity; or placement of the proto-oncogene under the control of potent viral promoters and enhancers, leading to the overexpression of its product or inappropriate timing of its function—frequently, both occur simultaneously. Retroviruses can exert a similar oncogenic effect through another route, without capturing cellular genes and transferring them between cells: DNA copies of viral RNA can simply integrate into the cellular genome adjacent to or even within proto-oncogenes. This phenomenon is known as insertional mutagenesis, and the gene modified in this manner is inherited by all progeny of that cell. Indeed, the random integration of viral RNA DNA copies into the host genome is part of the normal retroviral life cycle, and if it occurs within roughly 10,000 base pairs of a proto-oncogene, it can trigger aberrant activation of the disrupted gene. Insertional mutagenesis provides a means to identify proto-oncogenes by virtue of their proximity to the integrated retrovirus. Proto-oncogenes identified in this manner have largely matched those discovered through other methods, though novel ones have also emerged (Table 21-5), such as the int-1 gene activated in mice infected with mouse mammary tumor virus (Fig. 21-23). Cloning and sequencing of this gene revealed a striking similarity to the Drosophila wingless gene (see Section 16.5.9), which is likely involved in the Intercellular signaling pathways that regulate fly body segmentation.

Table 21-4. Oncogenes Initially Identified Within Transforming Retroviruses

Table 21-5. Selected Oncogenes First Identified by Means Other Than Presence in Transforming Retroviruses

Fig. 21-22. Many proteins encoded by transforming virus oncogenes are protein Kinases; their primary structures are compared schematically. Note that most oncogenes carry fragments of normal viral genes and therefore produce "fusion" proteins containing the N-terminus of the viral gag protein. (T. Hunter, Sci. Am., 251(2): 70–79, 1984.)

Fig. 21-23. Insertional mutagenesis. Integration activates the int-1 gene, resulting in mammary tumors in mice infected with mouse mammary tumor virus (MMTV). Viral integration sites identified in 19 distinct tumors are indicated by arrows. Note that insertions can activate Transcription of the int-1 gene from distances exceeding 10 kb on either side of the gene. This effect is driven by the presence of a potent enhancer within the MMTV long terminal repeats.

Fig. 21-24. Conversion of the c-abl proto-oncogene into an oncogene in patients with chronic myelogenous leukemia. As a result of chromosomal translocation, the bcr gene on chromosome 22 fuses with the c-abl gene from chromosome 9 (see Fig. 21-4). The resulting chimeric protein carries the N-terminal sequence of the bcr protein and the C-terminal sequence of the abl tyrosine protein kinase. The abl kinase domain acquires abnormal activity, which triggers the excessive proliferation of the hematopoietic cell clone in the Bone Marrow.

21.2.4. Investigation of the genetic nature of cancer points to the same small group of proto-oncogenes [15, 19]

While some researchers pursued a line of investigation that could be described as "from retroviruses to oncogenes," others adopted a more direct approach, searching for specific DNA sequences in human tumor cells capable of inducing uncontrolled proliferation when introduced into normal cells. The method developed for this purpose involved transfecting NIH 3T3 cells (a non-tumorigenic mouse 3T3 cell line) with DNA extracted from human tumor cells (see Section 13.4.3 and Fig. 13-33). The result was dramatic: oncogenes were detected in numerous human cancer cell lines, and in several cases, these oncogenes proved to be mutant alleles of the very same proto-oncogenes identified through retroviral research. For instance, a mutant representative of the ras gene family (oncogenes that cause rat sarcoma) was found in approximately every fourth human tumor. Thus, two independent lines of research led to the exact same group of genes.

Another approach, based on the karyotyping of tumor cells, yielded a similar result. As we mentioned earlier (Section 21.1.2), nearly all patients with chronic myelogenous leukemia exhibit the same chromosomal translocation in their cells (between chromosomes 9 and 22), while cells from Burkitt's lymphoma typically show a translocation between chromosome 8 and one of three chromosomes carrying immunoglobulin genes. It was discovered that in both types of malignancies, the translocation breakpoint (the site where a piece of one chromosome joins another) coincides with the Location of proto-oncogenes already known from retroviral studies: c-abl in chronic myelogenous leukemia and c-myc in Burkitt's lymphoma. Similar chromosomal translocations are associated with several other types of cancer. In some cases, the translocation converts a proto-oncogene into an oncogene by fusing it with another gene, resulting in the synthesis of an altered protein (Fig. 21-24); in others, the translocation places the proto-oncogene in an "unnatural" chromosomal environment, which activates its transcription and leads to the overproduction of the normal protein.

21.2.5. There are many ways to convert a proto-oncogene into an oncogene [14, 15, 20]

To date, about 60 proto-oncogenes have been identified (see Tables 21-4 and 21-5), each of which can be converted into an oncogene that plays a decisive role in The Development of a particular type of cancer. Most of these genes have been repeatedly discovered in A wide variety of tumors; it is therefore likely that the majority of mammalian proto-oncogenes have now been identified. Various genetic mechanisms can be responsible for converting a proto-oncogene into an oncogene (Fig. 21-25). A gene can be altered as a result of a point mutation, a chromosomal translocation, or the insertion of a mobile genetic element, such as a retrovirus. These changes may occur within the protein-coding Regions of the gene (resulting in a hyperactive product) or in the adjacent regulatory regions (which can lead to simple gene overexpression). Alternatively, overexpression can result from gene amplification (likely modeled on The process of anomalous chromosomal replication) (Fig. 21-26). Certain Genes frequently undergo very specific types of alterations (which may even serve as their signature); likewise, certain carcinogens cause specific, recognizable damage. For example, GENES OF THE myc family are often overexpressed or amplified, and in 90% of mouse Skin cancers induced by the tumor initiator dimethylbenz[a]anthracene, an A-to-T substitution is observed in the exact same region of the mutant H-ras gene.

Fig. 21-25. Three pathways for converting a proto-oncogene into an oncogene: deletion or point mutation (left), gene amplification (middle), and chromosomal rearrangement (right). The fourth mechanism (not shown) involves recombination between a proto-oncogene and retroviral DNA. The ultimate effect is similar to that of chromosomal reorganization: the proto-oncogene comes under the control of a viral enhancer and/or fuses with an actively transcribed viral gene.

Fig. 21-26. A possible mechanism of gene amplification leading to protein overproduction. The process begins with a duplication event, which appears to be driven by Illegitimate Recombination. The scheme illustrated suggests that illegitimate recombination may result from the destabilizing effects of excessive DNA replication. Once Gene Duplication has occurred, unequal sister chromatid exchange resulting from recombination between identical gene copies during DNA replication can further increase the gene copy number (see Section 10.5.2); as a result, their count within the chromosome can reach tens or hundreds. Numerous DNA repeats make the segment containing them visible, appearing on the chromosome as a homogeneously staining region. The amplified segment may also be excised from its locus (presumably again via some recombination mechanism) to give rise to autonomous double-minute chromosomes (see Section 21.1.13). The total length of a DNA segment amplified by such a mechanism typically exceeds 100,000 nucleotide pairs.

Fig. 21-27. Interrelationships among Major Classes of proto-oncogenes within the intracellular regulatory network responsible for the reception, Transduction, and execution of proliferative signals. One representative member is indicated for each class. An arrow from A to B means that in a normal cell, the Activation of a leads to the activation of B, either directly or indirectly. The scheme is based on data obtained from studies of molecules in cell-free systems and, in part, from studies of cells in which specific components were activated by oncogene Introduction or inactivated (via microinjection of corresponding Antibodies). Crucially, each class of regulatory molecules is represented by many members, so that every arrow in the diagram "conceals" a multitude of parallel arrows connecting individual members of one class to individual members of another. Moreover, members of a given class can often interact with one another (e.g., through mutual phosphorylation) as well as with members of other classes (both upstream and downstream). The presence of multiple parallel signaling pathways apparently enhances the cell's resilience to damage, meaning that a single oncogenic mutation is normally insufficient to render a cell tumorigenic; however, The complexity of this system and the presence of numerous feedback loops make its experimental study a formidable challenge. (For details, see Sections 10.2.8, 12.3.11-12.3.14, and 13.4.4-13.4.6.)

The molecular functions of proto-oncogenes mutated through such diverse mechanisms are now coming to light; furthermore, the number of proto-oncogenes directly involved in cell signaling pathways (or related to them, see Chapters 12 and 13, Sections 12.3.10 and 13.4.4) is constantly growing. The products of many genes interact with one another as components of a complex regulatory network (Fig. 21-27). Some proto-oncogenes encode growth factors, others encode their receptors or specific protein kinases, while still others encode ras-family G proteins or nuclear regulatory proteins. The MOLECULAR MECHANISMS OF carcinogenesis are thus taking on physical form: mutations in the c-sis gene, which encodes platelet-derived growth factor (PDGF), lead to its overexpression, endowing cells with The ability to stimulate themselves continuously; mutations in the c-erbB gene, which encode the epidermal growth factor (EGF) receptor, alter its properties as if it were perpetually activated by its growth factor; mutations that enhance the expression of the c-myc gene (whose product is believed to participate in preparing The Nucleus for division) grant the cell the capacity for unlimited proliferation. Nevertheless, far too much remains unclear. Despite the wealth of information regarding the DNA sequences and molecular structures of proto-oncogenes, oncogenes, and their products, we still have only a vague understanding of the role most of them play in cell physiology.

Fig. 21-28. Dynamics of tumor development in Three types of Transgenic Mice: those carrying the myc oncogene, those carrying the ras oncogene, and those carrying both oncogenes. Two lines of transgenic mice were used in the experiment: one carried an inserted copy of an oncogene generated by fusing the c-myc proto-oncogene with the MMTV promoter/enhancer (this promoter drives the overexpression of the gene under its control—in this case, c-myc—in specific Tissues, such as the mammary gland); the other carried an inserted v-H-ras oncogene under the control of the same promoter/enhancer. Tumors arose much more frequently in mice of both lines than in normal mice (typically in the mammary and Salivary Glands). Mice carrying both oncogenes were obtained by crossing these two strains. The tumor incidence in such hybrids was significantly higher than the combined incidence for carriers of the individual oncogenes. Despite this, tumors began to appear only after some time and exclusively in a small fraction of the cells within the tissue where both genes were expressed: apparently, for cancer to develop, an additional (stochastic) event was required beyond the presence of the two oncogenes. (Based on E. Sinn et al., Cell, 40, 465-475, 1987.)

21.2.6. Transgenic mice provide a suitable test system for studying oncogene action [21]

The oncogene concept contains a paradox: on the one hand, the first part of this chapter provided ample evidence that a single mutation is insufficient to cause cancer; on the other hand, an oncogene is a dominant gene capable of inducing neoplastic cell transformation in culture. This apparent paradox reflects the gap between the simplified cancer models most widely used in molecular biology research and the complexity of real human disease. The standard method for identifying oncogenes detects their effects not on normal human somatic cells, but on the mouse 3T3 cell line; these cells have already undergone mutations during adaptation to culture, which makes their transformation surprisingly easy upon a single additional genetic alteration. Moreover, as noted in Section 21.1.4, mice are less susceptible to cancer risk (having a shorter lifespan and a smaller Total Cell Count) than humans, and consequently their cells may be less robustly defended against the consequences of carcinogenic mutations compared to human cells.

However, even in the mouse, a single oncogene is usually insufficient to convert a normal cell into a cancerous one. This has been clearly demonstrated in transgenic mice. One can "stitch" an oncogene, taken as a DNA fragment from a virus or a tumor cell, to a suitable promoter DNA sequence and introduce the resulting molecule into the nucleus of a mouse egg cell. Such a recombinant DNA molecule often integrates into one of the chromosomes, yielding a line of transgenic mice that carry the oncogene in all their cells. This oncogene may be expressed in many tissues or restricted to a few "chosen" ones, depending on the tissue Specificity of the linked promoter. Typically, in mice engineered with myc or H-ras oncogenes in this way, the tissues expressing them enlarge dramatically, proliferating to excessive proportions, while individual cells eventually undergo further changes and give rise to tumors. Yet the vast majority of cells in a transgenic mouse expressing myc or H-ras oncogenes do not form tumors, indicating that the presence of a single oncogene is inadequate for tumorigenic transformation.

This approach was employed for further investigations. Transgenic mice carrying myc were crossed with transgenic mice carrying H-ras, yielding offspring in which both oncogenes were present simultaneously. Such progeny developed cancer much more frequently (and much faster) than either parental line (Fig. 21-28), but once again, the tumors arose as isolated growths scattered among non-cancerous cells. Thus, even a cell expressing two oncogenes must undergo further random changes before it can transform into a tumor cell.

Although this result is typical, exceptions have been discovered where cells of a single tissue in a transgenic mouse carrying a single oncogene simultaneously transformed into cancerous ones. However, in these cases, the oncogene was apparently "double-activated"—through a mutation in the coding region and/or fusion with an abnormal regulatory sequence, leading to its overexpression or inappropriate expression in the affected tissue. All this implies that even if a single oncogene is capable in some instances of driving The conversion of a normal cell into a tumor cell, a single genetic lesion is most likely unable to achieve this on its own.

21.2.7. Alongside the acquisition of oncogenes, cancer involves the loss of tumor suppressor genes [22]

All the currently available evidence creates the impression that the control and regulatory system of cell proliferation must possess a "safety margin" in the sense that no single genetic alteration is fatal to its normal functioning: cell division continues to remain within safe boundaries even if any single component of the system has failed. As a rule, a single oncogene is able to exert its dominant transforming effect on cell behavior only when the control system is already severely compromised. Success in identifying oncogenes has largely depended on the choice of a suitable test cell line.

Identifying and isolating tumor suppressor genes—that is, genes that normally inhibit excessive cell proliferation—has proven significantly more difficult. Nevertheless, there is sufficient evidence that the loss of such genes plays a key role in the pathogenesis of cancer. Specifically, in Section 13.4.3 we discussed how when cultured transformed cells are fused with untransformed ones, the resulting hybrids frequently display an untransformed phenotype. If the hybrid subsequently loses a specific chromosome inherited from the "normal" parent, its behavior once again takes on cancerous traits. It is logical to assume that in such cases, the lost chromosome harbored the sole cellular copy of a specific tumor suppressor gene.

In a normal diploid cell, both copies of a tumor suppressor gene must be inactivated or lost before growth and proliferation control is compromised: a mutation that contributes to the development of cancer through the inactivation or deletion of certain genes must be recessive. Conversely, a mutation that generates an oncogene can influence cell behavior even in the presence of a copy of the corresponding proto-oncogene. Owing to these properties, defective tumor suppressor genes can be recognized and distinguished from oncogenes using simple genetic criteria. A prime illustration is retinoblastoma, a rare childhood cancer that develops from neuronal cell precursors in the immature retina. Two forms of this tumor are known: hereditary and non-hereditary. The hereditary form typically presents with multiple tumors affecting both eyes and sometimes the bones, whereas in non-hereditary retinoblastoma, a single tumor affects one eye. Furthermore, some patients with the hereditary form exhibit a karyotypic anomaly—the disappearance of a specific band on chromosome 13. Deletions of this same locus have been documented in tumor cells from patients with the non-hereditary form of the disease. These facts indicate that the ROOT cause of tumor growth lies in the loss of a suppressor gene rather than the activation of an oncogene. Available data lead to the Conclusion that cancer predisposition in patients with the hereditary form is due to the absence in their cells of one of the two normal copies of a tumor suppressor gene known as the retinoblastoma gene, or RB1. Therefore, a single somatic mutation that ruins the remaining "good" copy of this gene in just one out of a million or more Cells of the growing retina is now sufficient to trigger cancer. In children without a hereditary predisposition, the onset of retinoblastoma is exceedingly rare, meaning it requires the coincidence in a single retinal cell of two somatic mutations that destroy both copies of the RB1 gene (Fig. 21-29). This gene has now been cloned; it has been shown to encode a protein that is normally expressed both in the retina and in many other tissues. It contains so-called "zinc fingers" (Structural motifs found in many DNA-binding proteins, see Section 9.1.9) and also binds to proteins that are products of oncogenes from certain DNA viruses. Using DNA probes derived from cloned RB1, researchers confirmed that, unlike their untransformed neighboring counterparts, tumor cells harbor deletions or inactivating lesions of the RB1 gene in both maternal and paternal chromosome sets; the loss of the normal gene can occur via various pathways (Fig. 21-30). Similar mechanisms, involving other genes, are responsible for certain other early childhood malignancies (such as Wilms Tumor). Although such variants of cancer are rare and, in a sense, exceptional, accumulating evidence strongly suggests that the loss or inactivation of tumor suppressor genes plays a vital role in the development of many cancers in adults.

Fig. 21-29. Genetic mechanisms underlying the onset of retinoblastoma. In the inherited form of the disease, all body cells lack one of the two normally present functional copies of a tumor suppressor gene. If the remaining copy is lost or inactivated through a somatic mutation in any cell, that cell can give rise to a tumor. In the non-inherited form, all cells initially contain both functional copies of the gene, and a tumor develops only if both copies are lost or inactivated—a process that requires the coincidence of two somatic mutations in a single cell.

Fig. 21-30. Potential mechanisms by which a cell with a defect in one of the two copies of a tumor suppressor gene may lose the second, functionally active copy, thereby taking a step toward malignant transformation. Cloned DNA probes combined with restriction fragment length polymorphism (RFLP) analysis are highly effective for analyzing tumor DNA and determining which specific genetic event has occurred in a given patient. It should be noted that most of these mechanisms result in the complete loss by the cell of a functional copy (either paternal or maternal) of the tumor suppressor gene in question. The remaining homolog is intrinsically defective and may be present as one or more genocopies. Such reduction to homozygosity or hemizygosity is a hallmark of cancers arising from the loss of a tumor suppressor gene. (Based on W. K. Cavenee et al., Nature 305, 779-784, 1983.)

21.2.8. Molecular biological analysis of Lung Tumors highlights the complexity and Variability of human cancers [23]

Each year, approximately 140,000 new cases of Lung Cancer are diagnosed in the United States. To date, no effective treatments exist for this form of cancer, and more than 90% of affected patients die within a year of Diagnosis. To understand the molecular Biological features of this malignancy, detailed studies were conducted on one of its forms—small-cell lung carcinoma (SCLC)—which accounts for 20% of all lung cancer cases. It is believed that small-cell carcinoma arises from neuroendocrine cells of the lung that secrete gastrin-releasing peptide (GRP; also known as bombesin due to its similarity to an amphibian neuropeptide of the same name) and other local chemical mediators. More than a hundred cultured cell lines have been isolated and analyzed from tumors of various patients with this type of cancer. These cells exhibit a wide spectrum of abnormalities, which also vary from line to line. Alterations have been identified in at least ten known oncogenes and tumor suppressor genes, alongside numerous chromosomal deletions and translocations, as well as disturbances in the secretion of signaling molecules that influence cell proliferation.

For instance, many of these cell lines are characterized by overexpression of the myc proto-oncogene or its relatives, N-myc and L-myc. This is frequently associated with the amplification of the corresponding chromosomal region, manifested in the karyotype as double minutes or homogeneously staining regions (Fig. 21-31, see also Section 21.1.13). In these same cells, genes of the ras family often undergo specific point mutations. Amplification of the myc gene consistently correlates with an especially rapid growth of tumor cells in culture and the complete suppression of differentiation. Patients whose tumors exhibit amplification of this gene face a particularly poor prognosis.

Alongside other chromosomal abnormalities, some tumor cell lines in small-cell lung carcinoma exhibit major chromosomal defects in the RB1 gene region: more than half of the lines lack detectable mRNA for this gene, even though it is actively synthesized in normal lung cells and most cancers of other origins. Furthermore, many tumor cell lines display distinct deletions involving a specific band on chromosome 3, suggesting that the loss of this gene is crucial for cancer development. Using cloned DNA probes complementary to this region on chromosome 3, it has been demonstrated that in nearly every case (of small-cell carcinoma and other major forms of lung cancer), one of the two parental copies of this chromosomal segment is lost. This deletion is absent in normal cells from the same patients, as well as in other tumor types from other individuals. Thus, this provides compelling evidence that the critical step in lung cancer development is the elimination of a specific tumor suppressor gene located on chromosome 3 via one of the mechanisms described for retinoblastoma in Fig. 21-30. It is hypothesized that in a typical lung cancer cell, one of the two original copies of the lung cancer suppressor gene is lost through the deletion of the chromosome segment containing it, while the other is inactivated by a more subtle (most likely point) mutation. By analogy with retinoblastoma, it is also suggested that individuals carrying a heritable recessive defect at this locus may exist, predisposing them to lung cancer. Because the onset of lung cancer presumably requires alterations in additional genes (alongside its marked dependence on environmental factors), hereditary predisposition in this case may be less pronounced than in retinoblastoma (although there are several reports of genetic variability in the susceptibility of smokers to lung cancer). Recently, similar data have been obtained regarding the loss of a tumor suppressor gene from a different locus in another common malignancy—colorectal carcinoma; it has been shown that a rare inherited form exists in this case, although its molecular relationship to the sporadic form remains unclear. It is expected that molecular biological studies similar to those just described will help identify individuals at heightened risk for specific types of cancer, allowing targeted preventive measures to achieve maximum efficacy.

Fig. 21-31. Karyotype of a tumor cell in small-cell lung carcinoma. Arrows indicate double minute chromosomes containing the amplified myc gene. (Courtesy of F. Whang-Peng and F. D. Minna.)

Another clinically significant finding from the molecular biological study of small-cell lung carcinoma concerns bombesin (GRP). This peptide acts as a growth factor, stimulating the proliferation of tumor cell lines. Certain cell lines (though not all) have proven capable not only of responding to this factor but also of secreting it, thereby establishing an autocrine loop through which the cells stimulate their own division. In culture, the proliferation of such cells can be arrested using antibodies that bind to the peptide and thereby block this stimulation. It is hoped that in the future, it will become possible to halt tumor growth in patients with this form of lung cancer using similar approaches.

21.2.9. Each case of cancer represents an independent natural experiment in cellular evolution

Any success in Treatment depends primarily on an accurate diagnosis. If we cannot precisely identify the disease, we can neither uncover its causes, nor predict its outcome, nor select the appropriate treatment for a given patient. As we have seen with small-cell lung carcinoma, the traditional Classification of oncological diseases is imprecise; upon close inspection, one of its standard categories dissolves into a heterogeneous group of diseases, each harboring its own distinct set of genetic lesions. Molecular biology is beginning to forge the tools required to precisely determine which genes are amplified, deleted, or mutated in the tumor cells of any individual patient. Information of this kind is just as vital for the treatment and Prevention of cancer as identifying the causative agent in infectious diseases.

The discovery of oncogenes and, more recently, tumor suppressor genes marked the end of an era of groping in the dark for the BIOCHEMICAL BASIS OF cancer. Nevertheless, a long path still lies ahead in bridging the gap between the oversimplified laboratory models that made these triumphant advances possible and an understanding of the true complexity of human cancers. We are still far from success, and progress in developing effective rational therapies has been frustratingly modest. We know the DNA sequences of many oncogenes and proto-oncogenes, yet at the same time, their exact physiological functions are known for only a few. A deeper understanding is urgently needed of how these and other molecules interact to govern the behavior of individual cells, alongside a firm grasp of the population-level mechanisms that drive The Emergence of cancer cells through mutation and natural selection.

Every cancer patient is the victim of an unsuccessful experiment in cellular evolution staged by nature itself. The complexity (and intricacy) of the cancer phenomenon stems from the vast richness of evolutionary pathways, yet it is precisely these evolutionary principles that hold the key to solving The problem of cancer. Thus, we return to the guiding idea introduced at the outset of this book: nothing in biology makes sense except in the light of evolution.

Conclusion

Normal cell proliferation is regulated by inhibitory and stimulatory molecules, which are the respective products of tumor suppressor genes and proto-oncogenes. The manifestation of cancerous traits in a cell can result either from the loss or inactivation of both cellular copies of a tumor suppressor gene, or from the amplification or hyperactivation of one of the two copies of a proto-oncogene. Heritable disruptions in proliferative control can also be triggered by the introduction of foreign viral genetic material into the cell. Retroviruses can become oncogenic themselves by "capturing" a copy of a host cell's proto-oncogene and converting it into an oncogene; they can also generate an oncogene within a cell by acting as insertional mutagens, integrating into the host genome adjacent to a proto-oncogene. Although most human cancers are thought not to be virally induced, mutations detected in tumor DNA frequently target the same proto-oncogenes identified through the study of retroviruses. Mechanisms of converting proto-oncogenes into oncogenes in human tumors include point mutations, gene amplification, and chromosomal translocations, which can lead to dysregulated expression of the proto-oncogene or its fusion with another gene, resulting in the synthesis of a novel chimeric protein. Similarly, tumor suppressor genes can be functionally lost through a wide variety of mutational events; individuals who inherit a deletion or a defective copy of one such gene from their parents may exhibit a marked predisposition to specific types of cancer, as demonstrated by retinoblastoma. Molecular biological analyses of tumor cells from patients suffering from some of the most common cancers have revealed a complex and heterogeneous spectrum of genetic lesions, encompassing both the activation of oncogenes and the loss of tumor suppressor genes. These findings reflect the stochastic Nature of the evolutionary process through which cancer arises, indicating that every malignant tumor is molecularly unique.

References

General

Cairns J. Cancer: Science and Society. San Francisco, Freeman, 1978.

Cancer Biology: Readings from Scientific American. New York, W. H. Freeman, 1986.

Darnell J.E., Lodish H.F., Baltimore D. Molecular Cell Biology, pp. 1035-1080. New York, W. H. Freeman/Scientific American Books, 1986.

De Vita V.T., Hellman S., Rosenberg S.A., eds. Cancer: Principles and Practice of Oncology, 2nd ed. Philadelphia, Lippincott, 1985.

Farmer P. B., Walker J. M., eds. The Molecular Basis of Cancer. New York. Wiley, 1985.

Franks L. M., Teich N., eds. Introduction to the CELLULAR AND MOLECULAR Biology of Cancer. Oxford, U.K., Oxford University Press, 1986.

Prescott D.M., Flexer A. S. Cancer: The Misguided Cell, 2nd ed. Sunderland, M. A., Sinauer, 1986.

Watson A. M., Hopkins N. H., Roberts J. W., Steitz J. A., Weiner A. M. Molecular Biology of the Gene, 4th ed., pp. 962-1096. Menlo Park, CA, Benjamin-Cummings, 1987.

Cited references

1. Cairns J. Mutation selection, and the natural history of cancer. Nature, 255, 197-200, 1975.

Nowell P. C. The clonal evolution of tumor cell populations. Science, 194, 23-28, 1976.

2. Doll R., Peto R. Epidemiology of cancer. In Oxford Textbook of Medicine (D.J. Weatherall, J.G.G. Ledingham, D.A. Warrell, eds.), 2nd ed., pp.

4.95-4.123. Oxford U.K., Oxford University Press, 1987.

Parkin D. M., Laara E., Muir C. S. Estimates of the worldwide frequency of sixteen major cancers in 1980. Int. J. Cancer, 41, 184-197, 1988. Ritchie A. C. The classification, Morphology, and behaviour of tumours. In General Pathology (H. Florey, ed.), 4th ed., pp. 668-719. London, Lloyd-Luke, 1970.

Robbins S. L., Cotran R. S., Kumar V. Pathologic Basis of Disease, 3rd ed., pp. 214-272. Philadelphia, Saunders, 1984.

3. Fearon E. R., Hamilton R., Vogelstein B. Clonal analysis of human colorectal tumors. Science, 238, 193-197, 1987.

Fialcow P. J. Clonal origin of human tumors. Biochem. Biophys. Acta, 458, 283-321, 1976.

Groffen J.J., et al. Philadelphia chromosomal breakpoints are clustered within a limited region, bcr, on chromosome 22. Cell, 36, 93-99, 1984. Nowell P. C., Hungerforcl D. A. A minute chromosome in human granulocytic leukemia. Science, 132, 1497, 1960.

4. Ames В., Durston W.E., Yamasaki E., Lee F.D. Carcinogens are mutagens: a simple test system combining liver homogenates for activation and

Bacteria for detection. Proc. Natl. Acad. Sci. USA, 70, 2281-2285, 1073.

Ashby J., Tennant R. W. Chemical structure, Salmonella mutagenicity and extent of carcinogenecity as indicators of genotoxic carcinogenesis among 222 chemicals tested in rodents by the U.S. NCI/NTP. Mutation Res., 204, 17-115, 1988.

Case R. M. P. Tumours of the Urinary Tract as an occupational disease in several industries. Ann. R. Coll. Surg. Engl., 39, 213-235, 1966.

Doll R. Strategy for detection of cancer hazards to man. Nature, 265, 589-597, 1977.

Hay A. How to identify a carcinogen. Nature, 332, 782-783, 1988.

Walcer J. M. Testing for carcinogens. In The Molecular Basis of Cancer (P. B. Farmer, J. M. Walcer, eds.), pp. 181-200. New York, Wiley, 1985.

5. Armitage P., Doll R. The age distribution of cancer and a multi-stage theory of carcinogenesis. Br. J. Cancer, 8, 1-12, 1954.

Peto R. Cancer epidemiology, multistage models and shortterm mutagenicity tests. In Origins of Human Cancer (H. H. Hiatt, J. D. Watson, J. A. Winsten, eds.), pp. 1403-1428. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1977.

6. Cairns J. Cancer. Science and Society, pp. 144 156. San Francisco, Freeman, 1978.

Campion M.J., McCance D.J., Cuzick ., Singer A. Progressive potential of mild cervical atypia: persrective, cytological, colposcopic, and virological study. Lancet, 2, 237-240, 1986.

Farber E., Cameron C. The sequential analysis of cancer development. Adv. Cancer Res., 32, 125-225, 1980.

Foulds L. The natural history of cancer. J. Chronic. Dis., 8, 2-37, 1958.

Mclndoe W.A., McLean M. R., Jones R. W., Millins P.R. The invasive potential of carcinoma in situ of the cervix. Obstet. Gynecol., 64, 451464, 1984.

7. Albanes D., Winick M. Are cell number and cell proliferation risk factors for cancer? J. Natl. Cancer. Inst., 80, 772 775, 1988.

Schinipper L. E. Clinical implications of tumor-cell heterogeneity. N. Engl. J. Med., 314, 1423-1431, 1986.

8. Berenblum I. A speculative review: the probable nature of promoting action and its significance in the understanding of The Mechanism of carcinogenesis. Cancer Res., 14, 471-477, 1954.

Berenblum I., Shibik P. The role of croton oil Applications associated with a single painting of a carcinogen in tumor induction in the mouse's skin. Br. J. Cancer, 1, 379-383, 1947.

Dolberg D.S., Holligsworth R., Hertle M., BisselM.J. Wounding and its role in RSV-mediated tumor formation. Science, 230, 676 678, 1985. Lamph W. W., Wamsley P., Sassone-Corsi P., Verma I. M. Induction of proto-oncogene JUN/AP-1 by serum and TPA. Nature 334, 629-631, 1988.

Reddy A.L., Fialcow P.J. Influence of dose of initiator on two-stage skin carcinogenesis in BALB/c mice with cellular mosaicism. Carcinogenesis, 9, 751-754, 1988.

9. Cairns J. The treatment of diseases and the war against cancer. Sci. Am., 253(5), 51-59, 1985.

Cohen L.A. Diet and cancer. Sci. Am., 257(5), 42-48, 1987. Doll R., Peto R. The Causes of Cancer. New York, Oxford University Press, 1981. Enstrom J. E. Cancer and mortality among active Mormones. Cancer, 42, 1943-1951, 1978.

Pike M.C., Ross R.K. Breast cancer. Br. Med. Bull., 40, 351-354, 1984.

10. Fentiman I. The local treatment of cancer. In Introduction to the Cellular and Molecular Biology of Cancer (L. M. Franks, N. Teich, eds.), pp. 350-362. Oxford, U.K., Oxford University Press, 1986.

Malpas J. Chemotherapy. In Introduction to the Cellular and Molecular Biology of Cancer (L. M. Franks, N. Teich, ed.), pp. 363-377. Oxford, U. K., Oxford University Press, 1986.

11. Sachs L. Growth, Differentiation, and the reversal of malignancy. Sci. Am., 254(1), 40-47, 1986.

12. Fidler L, Hart I. Biological diversity in metastatic neoplasms: origins and implications. Science, 217, 998-1003, 1982.

Liotta L. A. Tumor invasion and metastases - role of the Extracellular matrix: Rhoads Memorial Award lecture. Cancer. Res., 46, 1-7, 1986. McCarthy J.B., Skubitz A.P.N., Palm S.L., Furcht L.T. Metastasis inhibition of different tumor types by purified Laminin fragments and a heparin-binding fragment of Fibronectin. J. Natl. Cancer Inst., 80, 108-116, 1988.

Nicholson G. Cancer metastasis. Sci. Am., 240(3), 66-76, 1979.

Schirrmacher V. Cancer metastasis: experimental approaches, theoretical concepts, and impacts for treatment strategies. Adv. Cancer Res., 43, 173, 1985.

13. German J., ed. Chromosome Mutation and Neoplasia. New York, Liss, 1983.

Hanawalt P., Sarasin A. Cancer-prone Hereditary diseases with DNA Processing abnormalities. Trends Genet., 2, 124-129, 1986.

Lindahl T. Regulation and deficiencies in DNA Repair. Br. J. Cancer, 56, 91095, 1987.

Nowell P. C. Mechanisms of tumor progression. Cancer Res., 46, 2203-2207, 1986.

14. Alt F. W., Kellems R. E., Bertino J. R., Schimke R. T. Selective multiplication of Dihydrofolate Reductase genes in methotrexate-resistant variants of cultured murine cells. J. Biol. Chem., 253, 1357-1370, 1978.

Gerlach J. H., et al. Homology between P-glycoprotein and a bacterial haemolysin transport protein suggests a model for multidrug resistance. Nature, 324, 485-489, 1986.

Roninson l. В., Abelson H. Т., Houseman D. E., Howell N., Varshavsky A. Amplification of specific DNA sequences correlates with multi-drug resistance in Chinese hamster cells. Nature, 309, 626-628, 1984.

Schimke R. T. Gene amplification in cultured cells. J. Biol. Chem., 263, 5989-5992, 1988.

15. Klein G. The approaching era of the tumor suppressor genes., Science, 238, 1539-1545, 1987.

Watson J.D., Hopkins N.H., Roberts J. W., Steitz J.A., Weiner A.M. Molecular Biology of the Gene, 4th ed., Chapters 26-27. Menlo Park, CA, Benjamin-Cummings, 1987.

Wyke J. A. Viruses and Cancer. In Introduction to the Cellular and Molecular Biology of Cancer (L. M. Franks, N. Teich, eds.), pp. 176-199. Oxford, U.K., Oxford University Press, 1986.

16. Kripke M.L. Immunoregulation of carcinogenesis: past, present and future. J. Natl. Cancer Inst., 80, 722-727, 1988.

17. Laso P. A. Human papillomaviruses in oncogenesis. Bioessays, 9, 158-162, 1988. Steinherg B. M., Brandsma J. L., Taichman L. B. Cancer Cells 5/Papillomaviruses. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1987.

18. Bishop J.M. Oncogenes. Sci. Am., 246(3), 80-92, 1982.

Rijsewijk F., et al. The Drosophila homolog of the mouse mammary oncogene int-1 is identical to the segment polarity gene wingless. Cell, 50, 649-657, 1987.

Varmus H. E. The molecular genetics of cellular oncogenes. Annu. Rev. Genet., 18, 553-612, 1984.

19. Bishop J.M. The molecular Genetics of Cancer. Science, 235, 305-311, 1987.

Croce C. M., Klein G. Chromosome translocations and human cancer. Sci. Am., 252(3), 44-50, 1985. Weinberg R.A. A molecular basis of cancer. Sci. Am., 249(5), 126-142, 1983.

20. Bradshaw R. A., Prentis S. Oncogenes and Growth Factors. Amsterdam, Elsevier, 1987. (A collection of articles from Trends Biochem. Sci., Trends Genet., and Immunol. Today.)

Hunter T. The proteins of oncogenes. Sci. Am., 251(2), 70-79, 1984.

Kahn P., Graf Т., eds. Oncogenes and Growth Control. Berlin, Springer, 1986. (An excellent collection of short reviews.)

Quintanilla M., Brown К., Ramsden M., Balmain A. Carcinogen-specific mutation and amplification of Ha-ras during mouse skin carcinogenesis. Nature, 322, 78-80, 1986.

Smith M. R., DeGudicibus S.J., Stacey D. W. Requirement for c-ras proteins during viral oncogene transformation. Nature, 320, 540 543, 1986. (The article illustrates the potential of using antibodies to analyze the interplay of proto-oncogenes in the Regulation of cellular functions.)

21. Hanahan D. Dissecting multistep tumorigenesis in mice. Annu. Rev. Genet. 22 [in press], 1988.

Land H., Parada L.F., Weinberg R.A. Tumorigenic conversion of primary embryo fibroblasts requires at least two cooperating oncogenes. Nature, 304, 596-602, 1983.

Quaife C.J., Pinkert C.A., Ornitz D.N., Palmiter R.D., Brinster R.L. Pancreatic neoplasia induced by ras expression in acinar cells of transgenic mice. Cell, 48, 1023-1034, 1987.

Sinn E., et al. Coexpression of MMTV/v-Ha-ras and MMTV/c-mvc genes transgenic mice: synergistic action of oncogenes in vivo., Cell, 49, 465-475, 1987.

22. Hansen M. F., Cavenee W. K. Retinoblastoma and the progression of tumor genetics. Trends Genet., 4, 125-129, 1988.

Harris H. The genetic analysis of malignancy. J. Cell Sci. (suppl.), 4, 431-444, 1986.

Knudson A. G. Hereditary cancer, oncogenes, and antioncogenes. Cancer Res., 45, 1437-1443, 1985. Lee E. Y.-H. P., et al. Inactivation of the retinoblastoma susceptibility gene in human breast cancer. Science, 241, 218-221, 1988.

Weinberg R.A. Finding the anti-oncogene. Sci. Am., 259(3), 34 41, 1988.

23. Bodmer W. F., et al. Localization of the gene for familial adenomatous polyposis on chromosome 5. Nature, 328, 614-627, 1987.

Harbour J. W., et al. Abnormalities in structure and expression of the human retinoblastoma gene in SCLC. Science, 241, 353-357, 1988. (SCLC – small cell lung cancer.)

Kok K., et al. Deletion of a DNA sequence at the chromosomal region 3p21 in all Major Types of lung cancer. Nature, 330, 561-578, 1987. Leppert M., et al. The gene for familial polyposis coli maps to the long arm of chromosome 5. Science, 258, 1411-1413, 1987.

Minna J. D., et al. Molecular genetic analysis reveals chromosomal deletion, gene amplification, and autocrine growth factor production in the pathogenesis of human lung cancer. Cold Spring Harbor Symp. Quant. Biol., 51, 843-853, 1986.