National Academies Press: OpenBook

Cancer Today: Origins, Prevention, and Treatment (1984)

Chapter: Biology of Cancer

« Previous: Introduction
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 17
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 18
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 19
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 20
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 21
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 22
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 23
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 24
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 25
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 26
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 27
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 28
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 29
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 30
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 31
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 32
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 33
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 34
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 35
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 36
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 37
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 38
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 39
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 40
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 41
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 42
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 43
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 44
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 45
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 46
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 47
Suggested Citation:"Biology of Cancer." Institute of Medicine. 1984. Cancer Today: Origins, Prevention, and Treatment. Washington, DC: The National Academies Press. doi: 10.17226/18700.
×
Page 48

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 Genes Gone Awry Every minute 10 million cells divide in the human body. Usu- ally, they divide in the right way and at the right time, governed by a complex set of controls that have yet to be fully elucidated. When those controls fail, cancer may arise. The carefully ordered pattern of cell growth, division, and differentiation is lost. Instead, the cells begin to divide relentlessly, proliferating and massing into a tumor. Throughout the course of modern biology, researchers have been trying to determine what causes a cancer cell to run amok. As early as 1866, the French physician Pierre Paul Broca suspected a hereditary basis for the disease. That year he published the lineage of his wife's family, which revealed a clear predisposition to breast cancer. Subsequent investigators also pursued a more general ge- netic explanation for cancer, reasoning that it must arise from This chapter is based on the presentation given by J. Michael Bishop, University of California-San Francisco, at the 1983 annual meeting of the Institute of Med- icine. 17

CANCER TODAY damage to DNA, the genetic material, perhaps resulting in a break- down of the genetic regulation of the cell. The search apparently has reached its goal. Molecular biologists have found in the chromosomes of both animal and human cells a set of genes that can trigger cancer's unbridled growth. These oncogenes, as they are called, are normal genes that have gone awry. In their benign state, these genes perform some as yet un- known, but certainly essential, function in cell metabolism. Yet when they are inappropriately activated or altered in some way, these genes can start a cell on the path to cancer. Over two dozen oncogenes have been detected to date in a variety of different types of tumors. The exact way in which these oncogenes participate in the process of cancer remains obscure. Nor is it clear if they are involved in all or only some types of cancer. Investigators are beginning to understand some of the mechanisms that switch on these malignant genes. A far greater challenge will be to determine where and how these genes wreak their havoc in the cell—what proteins they make, and the nature of their function. The Legacy From Viruses The current insights into the genetics of cancer stem largely from research on viruses that cause tumors in certain animals. Tumor viruses have been studied sporadically since about 1910, when Francis Peyton Rous identified an infectious agent, which he assumed was a virus, that caused a particular type of cancer, known as a sarcoma, in chickens. Viruses are now known to cause a variety of animal cancers. The notion that viruses also cause human cancer has come in and out of favor since Rous' discovery. Recently, viruses have indeed been detected in some human tu- mors. This research is continuing, but at this time viruses do not appear to be a major cause of human cancer; thus other mechanisms must be sought. The most intriguing new findings in cancer biology have come from another line of research on cancer viruses. Rather than probe the causative role of tumor viruses in human cancer, some virol- ogists began in the 1960s to use them as experimental tools to 18

GENES GONE AWRY study what goes wrong in a cancer cell. Cancer seems to have many causes, such as radiation, chemicals, and perhaps viruses. Yet in each manifestation, the pattern is the same: the cell becomes disorganized and begins uncontrolled division. This suggests that cancer may proceed through the same molecular mechanisms no matter what its cause. Animal tumor viruses afford a powerful tool to ferret out these mechanisms, for by infecting normal cells with a tumor virus, investigators can observe the process of can- cerous transformation in the laboratory. Retroviruses Viruses are essentially a packet of genetic information encased in a protein sheath. Researchers have found that many animal tumor viruses, including the sarcoma virus detected by Rous, belong to a family of viruses known as retroviruses. These differ from other viruses and from higher organisms in that their genetic material is not DNA (deoxyribonucleic acid) but the related mol- ecule RNA (ribonucleic acid). Consequently, an extra step is in- volved in viral replication and gene expression. During the normal course of events in cell growth, the DNA molecules replicate themselves prior to cell division, producing one copy for each eventual daughter cell. The genetic information is expressed when DNA is transcribed into RNA, which then directs the synthesis of the protein encoded by the gene. Thus for retroviruses, RNA first must be reverse transcribed, or copied backward, into DNA, before the virus can replicate or infect a cell. Once this DNA copy has been made, it is inserted into the chromosomal DNA of the host cell that the virus infects. When the host cell DNA then replicates and undergoes transcription into RNA, it expresses the inserted viral genes along with its own genes. As a result, the host animal cell begins to make the proteins coded for by the viral genes, becoming, in essence, a virus factory. Investigators have found two distinct routes by which retro- viruses can induce cancer. One route is insertional mutagenesis: when a retrovirus inserts its DNA into the chromosomal DNA of the host cell, it mutates one or more genes of the host cell in the process. Some of these mutations can engender cancer. 19

CANCER TODAY The second route, which came to light in the early 1970s, is through the action of specific cancer-causing genes. One of the key discoveries was made by G. Steven Martin of the University of California at Berkeley, who was studying the Rous sarcoma virus. This is a minute virus, containing only four genes—by contrast, a higher organism has tens of thousands of genes. Through a series of experiments, Martin found that just one gene of the Rous sarcoma virus is responsible for the entire cancerous trans- formation of a cell. (The other three genes are involved in repli- cation of the virus.) Martin had detected an oncogene. That gene, now known as src for the sarcoma, or tumor, it induces, has since been more precisely identified by Peter H. Duesberg of Berkeley and Charles Weissmann, Martin Billeter, and John M. Coffin of the University of Zurich. Further research revealed that many other retroviruses also carry oncogenes, although some do not. It is this latter group of retro- viruses—those without oncogenes—that causes cancer through insertional mutagenesis. It has also recently been learned that these two forms of carcinogenesis are related in an unexpected way, which will be described later. About twenty retroviral oncogenes have been identified, each with a three-letter name such as src, myc, and ras. All of these genes are able to transform cells in culture, that is, to trigger cancerous growth. Oncogenes, like other genes, direct the synthesis of a specific protein, and it is these oncogene proteins that must be the actual culprits. At this stage, however, researchers have few clues to where and how these proteins act in the cell to induce cancer. Wayward Genes Soon after Martin's discovery, an intense research effort began on the oncogenes carried by retroviruses. The hope was that these genes would not turn out to be anomalies related only to a specific type of virally induced cancer, but that they might instead reveal something about the abnormalities of all cancer cells. In this vein, two researchers at the National Cancer Institute (NCI), Robert J. Huebner and George J. Todaro, soon developed the oncogene hypothesis. They postulated that once an oncogene is inserted into 20

GENES GONE AWRY an animal or human cell by a virus, it becomes a stable part of that cell's genetic complement, or genome, passed from one gen- eration to the next. It remains a harmless resident of the cell until a carcinogen, such as radiation or a chemical, spurs it to action. If so, then oncogenes should be present in normal animal cells. J. Michael Bishop, Harold E. Varmus, and Dominique Stehelin of the University of California at San Francisco (UCSF) began looking for them. They soon found the src gene in the cells of healthy chickens that had not been infected by a retrovirus. They went on to find a nearly identical gene in every animal tested, from rodents to human beings. It looked as if Huebner and Todaro might be right. Yet when the UCSF team examined the src gene from the healthy chicken cell in detail, they found it was not a viral gene at all, as Huebner and Todaro would have predicted. Instead, it bore the distinctive markings of a vertebrate gene; it was an integral part of the chicken's genome. Moreover, in the healthy cell the gene was active, so it obviously had a different function from its cancer-causing counterpart in the retrovirus. This finding added an unexpected twist to the oncogene hypothesis: it turns out that retroviral oncogenes are actually wayward cellular genes, picked up by the virus during its evolution. Since this initial discovery by Bishop and Varmus, close relatives of all the retroviral onco- genes have been found in normal cells. They are indeed harmless residents, and are called cellular oncogenes, or proto-oncogenes. They are designated c-src or c-myc, etc., to distinguish them from the viral oncogenes, v-src or v-myc. What these proto-oncogenes are doing in the cell is not known, but obviously, as Varmus has said, they were not put there to cause cancer. There are several clues, however. First, these proto- oncogenes have been conserved throughout vertebrate evolution. In other words, the same gene, such as the src proto-oncogene or the myc proto-oncogene, appears in almost identical form in all types of vertebrates. Some of these genes have even been found in invertebrates, such as fruit flies, and in yeast. That the proto- oncogenes have survived long periods of evolution unchanged indicates that they play an essential, if unknown, role in cell me- tabolism. And because cancer involves a disruption of normal 21

CANCER TODAY cellular growth, investigators suspect that proto-oncogenes are involved in the regulation of cell development and differentiation. The Enemy Within Retroviral oncogenes, then, appear to be slightly altered versions of normal genes; they are normal cellular genes gone awry. Within the cell, the proto-oncogene is benign. Yet when it is seized by a retrovirus and reintroduced into a cell, its malignant potential is unleashed. In some way as yet unknown, recombination with a retrovirus turns friend into foe. Experiments by George F. Vande Woude and Edward M. Scol- nick of the National Cancer Institute have provided evidence that the normal proto-oncogenes and viral oncogenes are one and the same. The unaltered proto-oncogenes do not transform cells in culture, as do the active, viral oncogenes. Yet when the researchers hooked up each of two proto-oncogenes (ras and mos) to a piece of viral DNA—a piece that dictates brisk expression of nearby genes—the proto-oncogene was able to convert normal cells to cancer cells. It is not clear how oncogenes disrupt normal cellular behavior. What is particularly puzzling is that the cancer-causing genes and their normal counterparts appear to be virtually the same. The proteins made by some of the oncogenes and their related proto- oncogenes have recently been identified, and those, too, appear to be quite similar. The question, then, is how to account for the dramatic difference in their functions. Two possible explanations have been proposed. One is that the gene is slightly mutated when it is captured by the virus, resulting in a slightly different protein having abnormal activity. Indeed, all retroviral oncogenes ex- amined to date are clearly damaged or mutated versions of their proto-oncogene precursors. The other explanation is that once the cellular gene is put under viral controls, it is expressed at the wrong time, or in the wrong amount, producing an overabundance of its protein product. Evidence exists for this hypothesis as well. Vande Woude and Scolnick found that when the ras and mos proto- oncogenes were put under viral control, their expression was en- hanced, which suggests that at least in some cases an increased 22

GENES GONE A WRY dosage of a normal protein may be involved in the initation of cancerous growth. Other researchers have found that those retroviruses that do not contain oncogenes—and instead induce cancer through insertional mutagenesis—also act on the normal proto-oncogenes contained in an animal cell. William S. Hayward and Benjamin G. Neel of the Memorial Sloan-Kettering Cancer Center and Susan M. Astrin of the Institute for Cancer Research in Fox Chase, Pennsylvania, have studied the site of insertion for one of these retroviruses, the chicken lymphoma virus. They have found that the viral DNA is almost always inserted into the chicken genome in the immediate vicinity of the myc proto-oncogene. Insertion seems to change c-myc, for its expression is greatly enhanced. The same has been found for other retroviruses that do not contain their own on- cogenes. It appears that these retroviruses exploit the cellular proto- oncogenes to induce tumors. More recently, activated oncogenes have also been detected in human tumors in which viruses are not involved. From these and other findings, a new cancer theory has emerged. Many biologists now believe that oncogenes lie at the heart of every cancer, no matter what its cause. They suspect that all agents of cancer— viruses, radiation, and chemicals—act upon the family of proto- oncogenes contained in each cell, somehow rendering them ma- lignant (Figure 2-1). There appear to be many ways to turn on an oncogene. Two have been identified so far in human cancers. Investigators recently found that the oncogene present in human bladder cancer is ac- tivated by a mutation in the gene's coding region, that is, where the instructions for making proteins lie. It is the smallest mutation possible (the substitution of a single nucleotide base), yet it is sufficient to instruct the gene to produce a slightly abnormal pro- tein, rather than an overabundance of a normal protein, as has been found in the viral oncogenes. In another human cancer, Bur- kitt lymphoma, the myc oncogene appears to be activated when a chromosome breaks and the gene is plucked from its usual site and inserted into a more active site on another chromosome. These modes of activation will be described in detail in the following two chapters. 23

CANCER TODAY NORMAL GROWTH AND DEVELOPMENT RETROVIRUS MUTAGENSOR CARCINOGENS (CHEMICALS. RADIATION, NONONCOGENE VIRUS. ETC.I CANCEROUS GROWTH FIGURE 2-1 Cancer-gene concept. (SOURCE"Oncogenes," J. Michael Bishop, Scientific American, March 1982, © 1982 by Scientific American, Inc. All rights reserved.) Cancer Proteins The proteins encoded by oncogenes perform some specific bio- chemical activity within the cell, and it is that activity that turns a normal cell cancerous. The key to understanding malignancy, then, appears to lie within those proteins. Since oncogenes were first detected, a number of investigators have been trying to iden- tify those proteins and determine where and how they act in the cell. Although most of the work to date has been performed on viral oncogenes, it provides insight into the biochemical activities of human oncogenes as well. The first success came from the laboratory of Raymond L. Erickson at the University of Colorado School of Medicine. In 1977, he and Joan S. Brugge reported the isolation of the protein encoded by the src gene of the Rous sarcoma virus, the first retro- viral oncogene detected. They named it pp60v-src. (The pp indi- cates that it is a phosphoprotein, the 60 refers to its molecular weight of 60,000 daltons, and the v-src indicates that its genetic origin is the viral src gene.) Isolation proved to be the easy part: they and others are still 24

GENES GONE AWRY trying to determine how the src protein induces cancerous growth. The first crucial step was to identify the protein by type. That was done by two groups, working independently: Erickson and his colleague Marc S. Collett, and Bishop, Varmus, and Arthur Levinson of the University of California School of Medicine in San Francisco. They found that the protein is a kinase, a type of enzyme that attaches phosphate ions to other proteins in a process known as protein phosphorylation. This discovery yielded one of the first clues to how the activity of a single protein could disrupt the normal behavior of a cell. Protein phosphorylation is thought to be one of the chief means by which the activities of growing cells are controlled. By phosphorylating cellular proteins, the src enzyme could conceivably change many aspects of cell structure and growth. Soon after, it was found that the src protein works in an un- conventional way. Most protein kinases attach proteins to one of two amino acids, threonine or serine. But Tony Hunter and Bar- tholomew M. Sefton of the Salk Institute for Biological Studies found that the src protein phosphorylates a different amino acid, tyrosine. Their finding spurred investigations to determine if ty- rosine phosphorylation occurs in normal cells. Now it is clear that it does, and that it plays a role in the regulation of cell growth. This is just one of many instances in which the study of oncogenes is providing insight into the normal processes of cell growth and development. As for the abnormal process, there is still only circumstantial evidence that protein phosphorylation is involved in the genesis of tumors. Hunter and other investigators are now locating and examining the targets of the src enzyme—the proteins it acts upon— to determine how their function is changed by phosphorylation. The src enzyme resides at the outskirts of the cell, bound to the plasma membrane. At least one of the proteins it attacks also resides in the membrane. Somehow this reaction at the cell's boundary influences events in the heart of the cell, the nucleus. Discovering how promises to be a lengthy process, for although one of the target proteins has been identified and dissected in fine detail, the investigators still know nothing of its normal function or how it is changed by phosphorylation. 25

CANCER TODAY A number of other oncogene proteins have also been charac- terized. Some, like the src enzyme, are protein kinases that phos- phorylate tyrosine. Others are glycoproteins (proteins attached to carbohydrates) or nuclear proteins. Some work in the outer mem- brane, others in the cytoplasm or the nucleus. With one exception, the normal physiologic role of these genes is not known, nor is it understood how their activation can trigger cancerous growth. The exception is the protein encoded by the oncogene of a monkey retrovirus, v-sis, which bears an uncanny resemblance to the growth factor found in blood platelets. This lends support to the hypoth- esis that the benign proto-oncogenes are part of the cell's normal regulatory network, and that activated oncogenes may be distorted versions of regulatory genes. Multistep Carcinogenesis Perhaps the greatest uncertainty at present revolves around fit- ting what is known about the activation of an oncogene into the overall scheme of carcinogenesis. In human beings, tumorigenesis is known to be a complex process, arising from a number of discrete steps occurring within a cell. While the activation of an oncogene may be a necessary part of the process, it is not sufficient in and of itself to induce cancerous growth; From the start, molecular biologists have suspected that several oncogenes may cooperate in transforming cells. That hunch has been borne out in recent experiments. Robert A. Weinberg of the Massachusetts Institute of Technology has found that some on- cogenes are unable to transform cells in culture, yet when another oncogene is added, cancerous growth ensues. Perhaps tumors arise from the concerted action of a number of oncogenes, each rep- resenting one of the multiple steps in tumorigenesis. Investigators can only speculate on whether each oncogene must be activated separately, or if perhaps the activation of one oncogene triggers the next, and then the next, resulting in a cascade of reactions. Perhaps the proto-oncogenes are part of a delicately balanced reg- ulatory network, and even a slight nudge is enough to tip the balance in favor of uncontrolled growth. The study of oncogenes is providing the first glimpses into some 26

GENES GONE AWRY of the genetic and biochemical changes that give rise to a cancer cell. It is too soon to predict exactly how this knowledge will be used to devise strategies to treat, cure, or prevent cancer. Inves- tigators speculate that once the molecular events that start the disease are understood, it may be possible to interrupt them. It may, perhaps, be possible to develop biological agents specifically targeted to kill cancer cells, or to somehow disrupt the action of an oncogene protein. Such strategies, however, await the further understanding of the mechanisms by which oncogenes derange the cell, as well as the interactions among oncogenes and the ex- ternal factors that trigger cancer. 27

3 Oncogenes in Human Cancer In the early 1970s, after a decades-long search for the roots of cancer, molecular biologists at last identified a suspect: the on- cogenes. By studying the retroviruses that cause cancer in animals, they had found a set of genes that acting alone or in concert could induce malignancy. Although these genes were first detected in viruses, it was soon learned that they were not bona fide viral genes at all. Instead, they are cellular genes that were picked up by the virus sometime during its evolution. These genes are found in benign form (in which case they are called proto-oncogenes) in the cells of higher animals, including human beings. Yet once they are captured by a virus and then reintroduced into an animal cell, they can trigger cancerous growth. Since this discovery, one of the most intriguing questions in cancer research has been what, if anything, these oncogenes can This chapter is based on the presentation given by Mariano Barbacid, National Cancer Institute-Frederick Cancer Research Facility, at the 1983 annual meeting of the Institute of Medicine. 29

CANCER TODAY reveal about human cancer. Do human cells also contain a set of genes that are responsible for the uncontrolled growth of tumor cells? If so, how are these putative human oncogenes activated? Probably not by viruses, or not often, because viruses may be only rarely involved in human cancer. Instead, most human can- cers are thought to arise from exposure to certain chemicals or to physical carcinogens, such as the various forms of radiation. The obvious question is whether these agents somehow activate human oncogenes. In the late 1970s, several groups of investigators began looking for oncogenes in human tumors. This was a separate endeavor from the work of other molecular biologists studying how retro- viral oncogenes induce cancer. Instead, these researchers studied malignant human cells, using gene transfer and other techniques to try to determine the molecular basis of human cancer. Before long, however, the efforts of these distinct groups converged. The Search for Human Oncogenes Some of the first attempts to look for human oncogenes took place in the laboratories of Robert A. Weinberg at the Massachu- setts Institute of Technology and Geoffrey M. Cooper of the Dana- Farber Cancer Institute at Harvard Medical School. In Weinberg's laboratory Chiaho Shih began looking for transforming genes in human and animal tumors. Using gene-transfer techniques, he isolated DNA from a variety of human tumor cell lines and in- troduced it into normal animal cells in culture. Those cells became cancerous. When the experiment was repeated using DNA from a normal human cell as a donor, there was no transformation. By transferring smaller and smaller pieces of DNA, the investigators were able to determine that the cancer-causing agent was a single gene, a human oncogene. The next task was to isolate these human oncogenes and clone additional copies for study. Weinberg, Cooper, and scientists in two other laboratories, those of Mariano Barbacid at the National Cancer Institute and Michael Wigler of Cold Spring Harbor, began trying to fish out the oncogene activated in human bladder cancer. The work was far trickier than isolating a viral oncogene because 30

ONCOGENES IN HUMAN CANCER the human cell contains tens of thousands of genes, as opposed to a handful in viruses. Once the oncogene had been isolated, the four groups began scanning its nucleotide sequences, the procedure used to identify a gene. They were surprised to find that the gene was an old acquaintance: the human bladder oncogene was remarkably similar to an oncogene that had first been detected in the Harvey sarcoma virus, which causes tumors in rats. Similar results were obtained with oncogenes found in human lung and colon tumors: they turned out to be closely related to the oncogene of another rat tumor virus, the Kirsten strain of the sarcoma virus. In turn, both of those rat oncogenes—and thus the human oncogenes as well— are also related, belonging to a gene family known as ras. With that discovery, the two groups of researchers—those in- vestigating viral oncogenes and those looking for human onco- genes—realized that at least in some cases they were studying the same genes. The ras genes from the rat sarcoma viruses have in- nocuous counterparts in the rat genome from which they are de- rived. Similarly, investigators found that the human oncogenes also have related proto-oncogenes in normal cells. Moreover, it became immediately clear that these proto-oncogenes could be activated to a malignant state by two distinct routes—recombi- nation with a virus or some nonviral mutation in a cell. Several other human oncogenes have since been isolated. They are found in a broad range of human cancers: carcinomas of the bladder, breast, lung, and colon, as well as in fibrosarcomas, neu- roblastomas, and leukemias. The same oncogene may be present in clinically unrelated cancers—there seems to be no correlation between a specific oncogene and a particular type of malignancy. Apparently, the nature of a tumor depends more on the type of tissue from which it derives than on which oncogene initiated the cancer process. One puzzling finding is that these transforming genes have so far been detected in only 15 to 20 percent of human cancers, although these represent the broad range of human malignancies. There are two hypotheses. One is that oncogenes participate in all human cancers but that current assays are not sensitive enough to detect them. There is some evidence to support this, as the myc 31

CANCER TODAY oncogene active in Burkitt lymphoma (discussed in Chapter 4) does not show up in some of these assays. The other possibility is that some human cancers arise by mechanisms that do not in- volve oncogenes. Most of the human oncogenes belong to the ras family of genes. Three distinct ras genes, H-ras, K-ras, and N-ras, have been isolated to date. Although they differ structurally, they all code for the same or similar protein, known as p21, which has a molecular weight of 21,000 daltons. The normal role of these ras genes is not known, but as with the other oncogenes, they are thought to be involved in the regulation of cell growth and differentiation. Most of the oncogenes detected to date in retroviruses and in mammalian cells also seem to fall into distinct groups or families. For molecular biologists, this is encouraging news. It suggests that the task of figuring out how these genes participate in cancer might not be as awesome as originally expected. Perhaps they code for a related group of proteins that perform a limited number of func- tions within the cell. The prevailing hypothesis is that the proto- oncogenes constitute part of a regulatory network, and when al- tered by a carcinogen they perform a slightly perverted version of this normal function. A caveat is necessary when describing the transforming abilities of human oncogenes. They have been shown to transform only cells in culture, not cells in the living tissue of whole organisms. The use of the cell culture assay is a simplification—and not a foolproof one—of what is actually an extremely complex process. In the standard assay for transforming properties, genes are intro- duced into a line of cells, known as NIH 3T3, derived from con- nective tissue cells of a mouse. These cells are normal in most respects, except that they are immortal—that is, they have adapted to grow and divide indefinitely in a culture medium. There is abundant evidence suggesting that as the cells adapt to culture, they become more susceptible to transformation by oncogenes. Many biologists suspect that immortal cells are already well on their way to transformation, and that the introduction of an on- cogene merely pushes them over the brink. Consequently, while this assay does detect malignant properties, it cannot be used as a 32

ONCOGENES IN HUMAN CANCER definitive determination of the ability of a gene to transform cells in vivo. Activation The key question is how a cancer gene differs from a normal gene—what change endows it with transforming properties? In 1982, Weinberg at MIT, Barbacid at the National Cancer Institute, and Wigler at Cold Spring Harbor found the answer for the human bladder oncogene. In repeated tests, the investigators had been unable to detect a structural difference between this oncogene and its normal counterpart, the proto-oncogene. Nonetheless, the on- cogene transformed cells in culture while the proto-oncogene did not. They realized that an extremely subtle change was involved— yet they never suspected just how subtle it would be. Working independently, all three groups found that the ras bladder oncogene differs from the normal gene in a single nucleotide substitution, the smallest genetic change possible. A short review of molecular genetics may help to put this ex- traordinary finding in perspective. Nucleotides are the subunits of the double-stranded DNA molecule. Each of the two DNA strands is composed of a string of four nucleotide bases—adenine, gua- nine, cytosine, and thymine—arranged in varying order. The two strands are held together by weak bonds between the nucleotide bases: adenine on one strand always pairs with thymine on the other, as does guanine with cytosine. A gene is a relatively short segment of DNA, composed of roughly 5,000 of these base pairs. Each gene contains the instruc- tions for the synthesis of a specific protein. (Actually, protein synthesis is a two-step process—DNA is transcribed to a related molecule, RNA, then RNA is translated into protein.) The genetic code for a protein is contained in the sequence of nucleotide bases. A codon, or series of three nucleotides, such as TAT (thymine- adenine-thymine) codes for a specific amino acid, the basic com- ponent of a protein. During protein synthesis, the genetic code is "read" and amino acids are inserted into the growing protein chain according to the message of each codon. 33

CANCER TODAY The research teams of Barbacid, Weinberg, and Wigler found a single mutation in one of the 5,000 nucleotides of the active human bladder oncogene. In one spot in the gene, the normal codon GGC is replaced by GTC—a thymine has been substituted for a guanine. This tiny substitution, known as a point mutation, results in a change in the protein encoded by the gene. At one end oftheoncogene's protein—specifically, at position 12—the amino 1 ONCOGENE ^BB£ PROTO-ONCOGENE ACTIVE V///////////////////////////////////////////A INACTIVE CLEAVE - "• :: J "a? :- ^ X. : -.-. : •': •': -1 V///////////////////777A V/////////////////A I LIGATE BECOMES INACTIVE '/////////////////A BECOMES ACTIVE BECOMES INACTIVE V////////, y////////////////A BECOMES ACTIVE 350-BASE SEGMENT — GLYCINE V//////////////77/ BECOMES INACTIVE BECOMES ACTIVE VALINE FIGURE 3-1 Point mutation in the EJ bladder carcinoma cell line. The proto-oncogene and the oncogene were cleaved at the same site and the resulting segments were joined. The resulting recombined genes were tested to see which had become active and which had lost activity. Successively smaller segments were similarly interchanged until a sequence only 350 bases in length was shown to be critical. This segment contains a single mutation: a guanine (G) in the proto-oncogene is converted to a thymine (T) in the oncogene, resulting in the specification of the amino acid valine rather than glycine. (SOURCE: "A Molecular Basis of Cancer," Robert A. Weinberg, Scientific American, November 1983, © 1983 by Scientific American, Inc. All rights reserved.) r 34

ONCOGENES IN HUMAN CANCER acid valine is inserted instead of the usual glycine. Somehow, this substitution changes the protein such that it can trigger cancerous growth (see Figure 3-1). The investigators had not expected to find a mutation in the protein-coding region of the gene. They had assumed instead that the regulatory region of the oncogene would be altered in some way. That seems to be the case in most of the retroviral oncogenes, at least. Researchers had already found that once those genes are put under viral control they are then turned on full blast. This excessive expression results in an overabundance of the protein product. In still other cases, investigators had found that oncogenes are present in multiple copies within a cell, also resulting in in- creased amounts of the protein. In both instances, too much of an otherwise normal protein seems to contribute to the malignant growth. For the human bladder oncogene, however, the change appears to be qualitative, not quantitative. The oncogene seems to be expressed in the correct amount, but in a slightly different form. Thus, an altered rather than overabundant protein seems to be at fault. It is not known what causes this mutation in the oncogene, or how the altered protein might act to induce cancer. Through com- puter analysis, investigators have recently found that the substi- tution of valine for glycine dramatically changes the structure of the protein. Presumably, this change in turn alters the protein's interactions with other molecules in the cell. Efforts are now under way to determine the biochemical activities of both the normal and transforming protein. Cause or Consequence? Although it is tempting to suppose that the altered oncogene initiates the tumor, it could nonetheless be a consequence, not a cause. The point mutation in the human bladder oncogene could be a product of the genetic disarray of a cancer cell. The case for the altered gene as cause has been strengthened, however, by a finding reported by Barbacid and his colleagues in early 1984. The NCI researchers, Barbacid, Eugenio Santos, Dionisio Martin-Zanca, and E. Premkumar Reddy, were aided by Marco A. Pierotti and 35

CANCER TODAY Guiseppe Delia Porta of the National Institute for the Study and Cure of Tumors in Milan, Italy. In cells of a squamous lung tumor removed from a 66-year-old man, the investigators found the active, mutated oncogene. Yet in normal bronchial tissue from the same patient, the mutation was not present. Thus, the mutated oncogene is not a part of the patient's normal genetic complement. Rather, it is clearly associated with the development of a cancer. As Barbacid says, it is hard to think that it is not responsible for that cancer. Perhaps, the researchers speculate, the gene was mu- tated by a chemical in cigarette smoke, for the patient was a heavy smoker. Barbacid and his colleagues then turned to animals to study the chemical initiation of cancer. They hoped to discern the method of attack of a chemical carcinogen—specifically, whether it acti- vated an oncogene. They induced breast cancer in female rats by injecting them with nitroso-methylurea, a potent chemical carcin- ogen. These induced tumors were found to contain transforming genes, while the breast cells from untreated rats did not. Again, the transforming gene turned out to be the now-familiar ras gene, activated by the same point mutation that occurs in human tumors. Of all the possible changes in the gene's 5,000 base pairs, the switch occurred in exactly the same one in all of the treated animals. Although more work must be done, this is the clearest indication yet to Barbacid and his colleagues that the chemical carcinogen, acting either directly or indirectly, mutates the proto-oncogene, and that such a mutation is responsible for the development of the tumor. Barbacid emphasizes that what is seen in the laboratory is just a small piece of the process of carcinogenesis—that the acti- vation of a single oncogene probably accounts for one of numerous biochemical changes necessary for a tumor to develop. That one change, however, appears to be critical, Barbacid says. By using induced tumors in rats to study chemical carcinogenesis, the NCI researchers hope to learn the exact role of oncogenes in human cancer. So far, their work has provided another piece of evidence, albeit indirect, that some human cancers arise when a chemical or other environmental insult activates an oncogene, creating a slightly altered protein, which then disrupts normal cell behavior. 36

ONCOGENES IN HUMAN CANCER To date, the point mutation has been found only in the ras gene. Another human oncogene, the myc gene, has recently been shown to be switched on by a different mechanism, chromosome re- arrangement. Additional mechanisms of activation probably re- main to be discovered. In short, there seem to be many ways to convert a normal gene to a cancer gene. 37

Broken Chromosomes When biologists first peered inside tumor cells, they found that many of the chromosomes, the rodlike structures that carry the genes, were strikingly abnormal. Many chromosomes had been broken apart; sometimes segments of DNA appeared in extra cop- ies. Sometimes pieces were either missing entirely or else had been flipped over or shuffled among chromosomes. In the past few years, geneticists have developed various tech- niques to keep track of the movements of chromosomes. Jorge Yunis of the University of Minnesota, for instance, stains tiny bands of chromosomes, often containing as few as 10 genes (an entire chromosome typically contains 1,000 genes). He can then follow that band should it break off of one chromosome and hook up with another or change position on its own chromosome. Using this technique, Yunis examined the malignant cells in This chapter is based on the presentation given by Philip Leder, Harvard Uni- versity School of Medicine, at the 1983 annual meeting of the Institute of Med- 39

CANCER TODAY tumors of 380 patients with various forms of cancer. Ninety-five percent of the cases showed some kind of chromosome defect— such as a translocation or a deletion. In 72 percent of the cases, the defect was specific to the type of cancer with which it was associated. Geneticists have long suspected that these chromosomal abnor- malities are somehow involved in the genesis of tumors. Yet until recently, they had no idea how. An answer is emerging in large part from another line of cancer research. As geneticists were cataloging the chromosome defects in cancer cells, molecular biologists were pursuing specific cancer- causing genes, the oncogenes. Part of the normal genetic com- plement of all cells, these oncogenes can be stirred to malignant action when they are disturbed in some way, perhaps by the action of a mutagen or a virus. There is now evidence that chromosome rearrangement may be one means by which an oncogene is acti- vated. In other words, cancer may arise when an oncogene on one chromosome is moved to a new position on another chromosome. Burkitt Lymphoma The evidence about chromosome rearrangement comes largely from the study of Burkitt lymphoma, a cancer that primarily afflicts children in Central Africa and New Guinea, and a related cancer that occurs in mice. Both this mouse plasmacytoma and Burkitt lymphoma are malignancies of the B-cells, a type of cell in the immune system that produces antibodies—the forces the body musters to ward off viruses and other foreign invaders. The B-cells of Burkitt victims show a specific chromosomal abnor- mality known as a translocation. In roughly 80 percent of the cases, a piece of chromosome 8 changes places with a piece of chromosome 14. In the remainder, the translocation occurs be- tween chromosome 8 and chromosome 2 or chromosome 8 and chromosome 22. A similar translocation occurs in mouse plas- macytoma. (The chromosomes are numbered according to size; in human beings, the largest is chromosome 1, the smallest is chromosome 22. The sex chromosomes are known as X and Y.) The first hint that these translocations might induce malignancy 40

BROKEN CHROMOSOMES came not from cancer research but from basic investigations into the genetics of the immune system. One of the most perplexing aspects of the immune system is its incredible diversity—how is the body able to produce millions of different antibodies to the substances it encounters? Each antibody protein is encoded by a specific gene, so the answer must lie there. Several researchers, including Philip Leder of Harvard University and Leroy Hood of the California Institute of Technology, have been studying these genes, which are located at particular sites known as the immu- noglobulin loci, on various chromosomes. The details have been worked out in the past few years. It turns out that the genes for specific antibodies are made when tiny bits of DNA—smaller than a gene—from a number of chromosomes join together. In a developing B-cell, chromosomes constantly break and then rejoin as bits of DNA are shuffled into new ar- rangements, each coding for a distinct protein, an antibody. These genes are then actively transcribed, that is, they are switched on and then direct synthesis of the antibody protein. At the sites of the antibody genes, then, chromosome re- arrangement is the norm. It was not long before several researchers noted the similarity between these normal events and the aberrant translocations that occur—with devastating effect—in Burkitt lymphoma. Suspicion that the normal process might somehow be involved in carcinogenesis increased when various investigators began mapping the antibody genes to their respective chromo- somes. They found that the antibody genes reside on three chro- mosomes, numbers 14, 2, and 22—precisely the same chromo- somes that exchange pieces with chromosome 8 in Burkitt lymphoma. The connection was not lost on George Klein, who was studying mouse plasmacytoma at the Karolinska Institute in Sweden. He proposed that chromosome 8 must harbor a proto-oncogene—a benign gene that has malignant potential. Burkitt lymphoma would arise, he reasoned, when that oncogene was moved from chro- mosome 8 into the antibody-producing region on chromosome 14, where it would be switched on and expressed along with the antibody genes. Following Klein's hunch, several other laboratories began look- 41

CANCER TODAY ing for a proto-oncogene on chromosome 8. Some of the key researchers in this effort include Leder, Hood, Carlo Croce at the Wistar Institute, Ricardo Dalla-Favera and Robert Gallo of the National Cancer Institute, Michael Cole of the St. Louis Univer- sity School of Medicine, and Kenneth Marcu of the State Uni- versity of New York at Stonybrook. The cellular myc gene, originally isolated from a chicken retro- virus, seemed a likely candidate for the suspected oncogene, as William S. Hayward of the Memorial Sloan-Kettering Cancer Center had found that this gene is activated in other forms of lymphoma. Hayward, Gallo, and Croce found that indeed, the myc gene is located on chromosome 8, on the segment that is swapped in Burkitt lymphoma. At the same time, Leder and his colleagues found exactly where on chromosome 14 the myc gene is inserted— into or very close to the site of an active antibody gene (see Figure 4-1). They have subsequently found that the oncogene is also FIGURE 4-1 Chromosome translocation in Burkitt lymphoma. (SOURCE. Bristol-Myers Company/University of Chicago Cancer Research Center.) 42

BROKEN CHROMOSOMES inserted into the antibody gene sites on chromosomes 2 and 22. Other researchers, including Marcu, Croce, and Hood, found that in mouse plasmacytoma the myc gene was also translocated into the region of the antibody gene. Uncertain Data In its new position, the once-benign myc gene appears to be suddenly malignant. Yet the exact way in which the translocation activates the oncogene remains obscure. Similarly, there are few clues to how the activated oncogene might cause Burkitt lym- phoma. There are two possibilities. The translocation could disturb the regulation of the gene, turning it on at the wrong time or in the wrong amount, as has been found for many retroviral oncogenes. Alternately, the reshuffling could mutate the coding region of the gene, resulting in the production of a slightly different protein. Another human oncogene, the ras gene, has recently been found to be switched on in that way. Some evidence exists to support each theory. Several investi- gators have found that in Burkitt cells the translocated myc gene is overexpressed—it is turned on too high and is making too much of its protein product. This finding fits in well with Klein's original hypothesis. He had postulated that when the gene was inserted into the antibody site it would fall under different control signals and be overexpressed. The excess protein would then spur malig- nant growth. This theory is buttressed by the finding that levels of myc expression are elevated in other, non-Burkitt tumors. Other research points to an altered protein caused by mutation or other damage to the gene. At St. Louis University, for instance, Michael Cole has evidence that a segment of the myc gene may be lost in the translocation. Sorting out these apparently conflicting data will take some time. Investigators are just beginning to characterize the normal myc gene—a crucial step in determining the difference between the normal gene and the transforming one. Comparisons of the level of expression of the two versions of the gene are speculative, as little is known about its expression in normal cells. Similarly, 43

CANCER TODAY not much is known about either the physical properties of the normal myc protein or its function within the cell. Once these and other results are in, it may turn out that the two theories—a regulatory disturbance or an altered gene product— are both correct. In Burkitt cells, the translocations do not always occur in exactly the same place; several distinct crossover points between chromosomes 8 and 14 have been found. In some cases, the myc gene is not inserted into the antibody gene site at all, although it is translocated to a nearby site on the same chromo- some. Perhaps the myc gene can be activated in a number of ways depending on where the break occurs. A Regulatory Disturbance The newest evidence suggests that a regulatory disturbance is the major culprit in Burkitt lymphoma. Much of this evidence has come from Philip Leder and his colleagues at Harvard University, Jim Battey, Christopher Moulding, William Murphy, Huntington Potter, Timothy Stewart, and Rebecca Taub, and Gilbert Lenoir of the World Health Organization's International Agency for Re- search on Cancer. They have found that the structure of the coding region of the translocated myc gene is identical to its normal, unrearranged coun- terpart. This strongly suggests that their protein products must also be identical. If so, then the transforming gene and the normal gene must differ chiefly in their expression, at least in the particular line of Burkitt cells that the Harvard team examined. The geneticists had expected to find overexpression of the trans- located myc gene in Burkitt cells. Yet their measurements revealed that expression of the translocated oncogene was only slightly, if at all, higher than that of a myc gene in a noncancerous B-cell in culture. Other, investigators have consistently found a slight ele- vation in expression of the translocated myc gene. In yet other cases, a sizable, 20- to 40-fold increase in expression has been detected In short, although the gene appears to be aberrantly ex- pressed in most cases, the level of expression varies dramatically. Investigators have been perplexed by the inconsistent findings. Given this wide variation, Leder began to think it unlikely that 44

BROKEN CHROMOSOMES the elevated expression of the myc gene was at fault in Burkitt lymphoma, or not so simply. In particular, he wondered why such a minimal increase in expression should confer transforming properties on a gene. He began to suspect that a more complex regulatory disturbance had occurred. In some instances, Leder's group has been able to study a normal and a rearranged myc gene within the same cell. (Genes are always present in two copies, known as alleles—one on the maternal chromosome and one on the paternal. In most Burkitt cells, only one of the chromosomes has been rearranged, the other is un- changed.) In comparing expression of the two versions of the myc gene, the Harvard researchers have detected subtle changes in expression, specifically, in the way in which the gene is tran- scribed. They have also noted another intriguing phenomenon: in Burkitt cells, the translocated gene is active—that is, it is pro- ducing protein—while the normal gene is silent. Leder and his colleagues have now found that although the protein-coding region of the translocated myc gene is intact, an- other part of the gene has been damaged by the translocation. They think that these mutations occur in the region of the gene that regulates its expression. Usually the damage is slight, a small alteration in the nucleotide sequences. But sometimes this entire control region is missing. Kathleen Kelly, another researcher in Leder's laboratory, has been examining expression of the myc gene in normal cells. The myc gene is thought to be involved in cell growth and division, perhaps in mitosis, or DNA replication, or transcription. In col- laboration with Brent Cochran and Charles Stiles of the Dana- Farber Cancer Center, Kelly is trying to determine at what stage in the cell cycle the gene is normally active. They have found that expression of the myc gene appears tightly controlled: it is switched on at a very specific time, as the cell is moving from the resting phase (through which a cell must pass in order to divide) to the growth phase. The rest of the time it is silent, presumably blocked by some kind of repressive control. Leder suspects that during translocation, the myc gene slips loose of the repressive controls. Then, because the control region of the translocated gene is damaged or missing, whatever usually shuts 45

CANCER TODAY off the gene can no longer affect it at its new site. This would explain why the translocated gene is active in a Burkitt cell and the normal allele is usually silent. The important consequence of this deregulation, Leder postulates, is not overproduction of a protein, although that may be involved, but rather a change in the time the gene is expressed. Because myc expression is tightly related to the cycle of cell division and growth, a change in its timing would disrupt the cycle's normal pattern. The action of the myc gene alone is not sufficient to induce Burkitt lymphoma. Leder and others suspect that the gene col- laborates with other, as yet unknown oncogenes to transform the cell. Several activated oncogenes have been detected in many other tumors. Robert A. Weinberg of MIT recently found evidence of actual oncogene collaboration. He has shown that in some non- Burkitt cell line systems, the myc gene is unable to transform cells alone; it requires the action of another oncogene. Indeed, since the myc gene was detected, another transforming gene, known as Blym, has recently turned up in Burkitt cells. Whether it collaborates with the myc gene, and if so in what way, remains to be deter- mined. Spontaneous Carcinogenesis Details remain to be worked out on exactly how the translo- cation activates the myc gene. Yet even at this preliminary stage, the research in Burkitt cells provides a model for spontaneous Carcinogenesis, or cancer that arises apparently without the action of a virus or a mutagen. In the past decade or so, biologists have become increasingly aware that genes are not permanently fixed in one place. They can change position on their chromosome and are sometimes shuffled between different chromosomes. The translocations in Burkitt cells suggest that some cancers arise when this normal process goes awry—when genes end up in the wrong place. It is intriguing to view Burkitt lymphoma as a perversion of the normal process of antibody production. In a normal B-cell, chromosome 14 ex- changes pieces with chromosomes 2 and 22. In Burkitt cells, the instructions seem to have been muddled, and instead, these three 46

BROKEN CHROMOSOMES chromosomes begin exhanging bits with chromosome 8, which bears the myc oncogene. Chromosomal abnormalities are not restricted to Burkitt cells; as mentioned earlier, some type of chromosome defect occurs in most cancer cells. Cells isolated from many types of leukemias also show chromosomal translocations, and investigators have re- cently detected oncogenes on the chromosomes involved in these switches. In other types of cancer, such as solid tumors, the most frequent abnormality is a deletion of part of a chromosome. Re- search is under way to determine if these other chromosomal defects also affect an oncogene. 47

Next: Diet and Cancer »
Cancer Today: Origins, Prevention, and Treatment Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!