Tumors often possess cytogenetically different clones that arise from the initial transformed cell through secondary or tertiary genetic alterations. This heterogeneity contributes to differences in clinical behavior and responses to treatment of tumors of the same diagnostic type. Apart from the initial clone and subclones, tumors can also contain progenitor cancer cells, all of which constitute a spectrum of cells with different genetic alterations and states of differentiation. These populations can differ in sensitivity to chemotherapy, radiotherapy, and other treatments, making clinical management difficult. For these reasons, the initiating steps in the development of cancer are of considerable clinical importance and are a priority in the development of rational cancer treatment.
An example of this concept is chronic myelogenous leukemia, which is initiated by a reciprocal t(9;22) chromosomal translocation that fuses the ABL proto-oncogene to the BCR gene.1,2 The fusion gene encodes an oncogenic ABL fusion protein with enhanced tyrosine kinase activity. All leukemic cells carry this chromosomal alteration, which is why inhibition of the excessive tyrosine kinase activity of the fusion protein by imatinib induces complete remission in most patients3,4; when relapse occurs, the leukemic cells usually carry mutations in ABL that render them resistant to the drug.5
Evidence of Somatic Genetic Change
The first evidence that cancer arises from somatic genetic alterations came from studies of Burkitt's lymphoma, in which one of three different translocations juxtaposes an oncogene, MYC, on chromosome 8q24 to one of the loci for immunoglobulin genes. Chromosomes 14q, 22q, and 2p — the translocation partners — each carries enhancer elements in the immunoglobulin loci, thereby activating the juxtaposed MYC oncogene (see Fig. 1 in the Supplementary Appendix, available with the full text of this article at www.nejm.org).6,7,8,9,10,11 Since every malignant lymphocyte carries the MYC translocation, deregulation of the MYC oncogene is probably the initiating event.
Second, transfection experiments have shown that mouse fibroblasts, when transfected in vitro with DNA from human cancer cells, acquire some of the properties of malignant cells (i.e., transformation). The transforming activity of the DNA was traced to a human homologue of the retroviral RAS oncogene. This oncogene bears mutations that activate the transforming property of the RAS oncogenic protein.12,13
Third, the cloning and characterization of the chromosomal breakpoints that are characteristic of follicular lymphomas and some diffuse large B-cell lymphomas14 have shown a juxtaposition of the BCL2 oncogene to enhancer elements in the immunoglobulin heavy-chain locus, resulting in deregulation of BCL214,15 (see Fig. 2 in the Supplementary Appendix).
Fourth, in transgenic mice that carry an activated oncogene from a human tumor, cancers develop that resemble the human tumor.16,17 That these cancers appear only after a latent period suggests that alterations in other genes must occur before progression to frank neoplasia can occur — activation of a particular oncogene seems to be necessary but not sufficient for the development of overt cancer.
Functional Properties of Oncogenes
Historically, transformation events in cancer have been defined as initiation events (contributing to the early stages of neoplastic transition) or progression events (referring to subsequent transformative processes). Oncogenes encode proteins that control cell proliferation, apoptosis, or both. They can be activated by structural alterations resulting from mutation or gene fusion,18 by juxtaposition to enhancer elements,19 or by amplification. Translocations and mutations can occur as initiating events20 or during tumor progression, whereas amplification usually occurs during progression. (Table 1 in the Supplementary Appendix lists oncogenes in tumors of different species, the methods used to identify them, their mechanisms of activation, and the functions of their encoded products; Table 2 in the Supplementary Appendix lists the molecularly characterized chromosomal rearrangements in human cancers.) The products of oncogenes can be classified into six broad groups: transcription factors, chromatin remodelers, growth factors, growth factor receptors, signal transducers, and apoptosis regulators.
Products of Oncogenes
Transcription Factors
Transcription factors are often members of multigene families that share common structural domains. To act, many transcription factors require interaction with other proteins. In some tumors, for example, the Fos transcription protein dimerizes with the Jun transcription factor to form the AP1 transcription factor, and this complex increases the expression of several genes that control cell division.21,22
Chromosomal translocations often activate transcription-factor genes in lymphoid cancers23 and sometimes do so in solid tumors (e.g., prostate cancer24; see Table 2 in the Supplementary Appendix). In certain sarcomas, chromosomal translocations that result in fused proteins occur consistently; in Ewing's sarcoma, for example, the EWS gene is fused with one of a number of partner genes, resulting in aberrant transcriptional activity of the fused proteins (see Table 2 in the Supplementary Appendix). The EWS protein is an RNA-binding molecule with a domain that, when fused to a heterologous DNA-binding domain, can greatly stimulate gene transcription. Prostate carcinomas carry translocations of the TMPR552 gene that fuse with and activate ERG1 or ETV1. These genes are members of the ETS family of transcription regulators, which can activate or repress genes involved in cellular proliferation, differentiation, and apoptosis. The fusion of TMPR552, which has androgen-responsive promoter elements, with an ETS-related gene creates a fusion protein that increases proliferation and inhibits apoptosis of cells in the prostate gland, thereby facilitating their transformation into cancer cells.24
Chromatin Remodelers
Modifications in the degree of compaction of chromatin play a critical role in the control of gene expression, replication, and repair and of chromosome segregation. Two kinds of enzymes remodel chromatin: ATP-dependent enzymes25 that move the positions of nucleosomes, the repeating subunits of the histones in chromatin around which DNA winds, and enzymes that modify the N-terminal tails of histones.26 The pattern of histone modification constitutes an epigenetic code that determines the interaction between nucleosomes and chromatin-associated proteins.27 These interactions, in turn, determine the structure of chromatin and its transcriptional capacity.
In acute lymphocytic leukemia and acute myelogenous leukemia, the ALL1 (also named MLL) gene can fuse with 1 of more than 50 genes. ALL1 is part of a very large, stable multiprotein complex. Most of the proteins in the complex are components of transcription complexes28; others are involved in histone methylation and RNA processing. The entire complex remodels, acetylates, deacetylates, and methylates nucleosomes and free histones.28 The fusion of ALL1 with 1 of more than 50 proteins results in the formation of the chimeric proteins that underlie acute lymphoblastic leukemia and acute myelogenous leukemia. ALL1 (MLL) fusion proteins deregulate homeobox genes (which encode transcriptions factors), the EPHA7 gene (which encodes a receptor tyrosine kinase), and microRNA genes such as miR191.
Growth Factors
Constitutive activation of a growth factor gene can contribute to malignant transformation. Platelet-derived growth factor (PDGF) consists of 
and β chains and is released from platelets during coagulation.29 It can induce the proliferation of various cell types and stimulate fibroblasts to participate in wound healing. The sis oncogene of simian sarcoma virus is structurally similar to the gene for the β chain of PDGF.29 Overexpression of PDGF induces the in vitro transformation of fibroblasts containing PDGF receptors; it does not influence fibroblasts lacking these receptors. This autocrine loop entails overexpression of PDGF-β, the expression of the PDGF-β receptor, and unregulated cell growth. An antibody against PDGF-β, an antibody against its receptor, or small molecules that block the receptor inhibit growth of the transformed fibroblasts.
The WNT family of secreted glycoproteins inhibits phosphorylation of β-catenin, which is involved in cell–cell adhesion and the activation of several signal-transduction pathways30 (Figure 1). The APC protein controls the activity of β-catenin. In familial adenomatous polyposis, inactivating mutations of APC block the degradation of β-catenin by inhibiting its phosphorylation. As a result, free β-catenin in the cytoplasm translocates to the nucleus, where it activates genes involved in cell proliferation and invasion (Figure 1).
|
Growth factor receptors are altered in many cancers (Figure 2).31 In many tumors, a deletion of the ligand-binding domain of epidermal growth factor receptor (EGFR), a transmembrane protein with tyrosine kinase activity, causes constitutive activation of the receptor in the absence of ligand binding.32 The activated receptor phosphorylates tyrosines in the intracellular domain of the receptor, providing interaction sites for cytoplasmic proteins containing the SRC homology domain and other binding domains. These interactions deregulate signaling in several pathways. Activating mutations occur in three other members of the EGFR family — ERBB2, ERBB3, and ERBB4 — and within the kinase domains of the HER2/neu and KIT signaling receptors. Such mutations occur in lung and breast cancer and gastrointestinal stromal tumors. Two classes of clinically active anti-EGFR agents have been developed: a monoclonal antibody against the extracellular domain of the receptor (cetuximab) and competitive inhibitors of the tyrosine kinase activity of the receptor (e.g., erlotinib and gefitinib).
|
|
Binding of receptor tyrosine kinases to the appropriate ligand causes reorganization of the receptors and autophosphorylation of tyrosines in the intracellular portion of the molecules35 (Figure 2). Autophosphorylation enhances the kinase activity of the receptor or promotes the interaction of the receptor with domains of cytoplasmic proteins (e.g., the SRC homology 2 domain) that are effectors and regulators of intracellular signaling.36 In humans, there are approximately 120 SRC homology 2 domains in 100 different proteins that mediate responses to signals initiated by phosphorylated tyrosines. Some of these proteins share domains with enzymatic activity, whereas others link activated receptors to downstream targets.
Many oncogenes encode members of signal-transduction pathways. They fall into two main groups: nonreceptor protein kinases and guanosine-triphosphate–binding proteins.37,38 The nonreceptor protein kinases are of two types: tyrosine kinases (e.g., ABL, LCK, and SRC) and serine and threonine kinases (e.g., AKT, RAF1, MOS, and PIM1). Proteins involved in signal transduction become oncogenic if they bear activating mutations. An important example is PI3K and some of its downstream targets, such as AKT and SGK, which are critical to tyrosine kinase signaling and can be mutated in cancer cells.
Apoptosis Regulators
The BCL2 gene, which is involved in the initiation of almost all follicular lymphomas and some diffuse large B-cell lymphomas (see Fig. 2 in the Supplementary Appendix),14,15 encodes a cytoplasmic protein39,40 that localizes to mitochondria and increases cell survival by inhibiting apoptosis.41 BCL2 is also important in chronic lymphocytic leukemia and lung cancer. The BCL2 family members BCL-XL and BCL2 inhibit apoptosis and are up-regulated in many cancers.
Two main pathways lead to apoptosis: the stress pathway and the death-receptor pathway (Figure 4). The stress pathway is triggered by proteins that contain the BCL2 homology 3 domain; this domain inactivates BCL2 and BCL-XL (which normally inhibit apoptosis) and thereby activates the caspases that induce apoptosis (Figure 4). Drugs that mimic the BCL2 homology 3 domain and can bind to BCL-XL or BCL2 (peptides or small organic molecules that bind in a groove of these proteins) are under development. This approach has attracted considerable attention because many tumors overexpress BCL2 or related proteins. The death-receptor pathway is activated by the binding of Fas ligand, TRAIL, and tumor necrosis factor , to their corresponding (death) receptors on the cell surface. Activation of death receptors activates caspases that cause cell death (Figure 4).
|
Activation of oncogenes by chromosomal rearrangements, mutations, and gene amplification confers a growth advantage or increased survival of cells carrying such alterations. All three mechanisms cause either an alteration in the oncogene structure or an increase in or deregulation of its expression.42
Chromosomal Rearrangements
Chromosome inversions and translocations are common cytogenetic abnormalities in cancer cells. In hematopoietic cancers and solid tumors, the translocations and inversions increase or deregulate transcription of the oncogene. In prostate cancer, gene fusion occurs between a gene that carries a promoter that is very active in the target cells, and another that carries the oncogenic activity (e.g., ERG1).24 In cancers of B and T cells, the most common mechanism of activation by translocation resembles MYC deregulation, whereas in myeloid cancers and soft-tissue sarcomas, gene fusion is more common (see Table 2 in the Supplementary Appendix).
Mutations
When an oncogene is activated by mutation, the structure of the encoded protein is changed in a way that enhances its transforming activity. Many types of mutation occur in oncogenes.43 Examples are the RAS oncogenes (KRAS, HRAS, and NRAS), which encode proteins with guanosine-nucleotide–binding activity and intrinsic guanosine triphosphatase activity. When mutated in codon 12, 13, or 61, the RAS genes encode a protein that remains in the active state and continuously transduces signals by linking tyrosine kinases to downstream serine and threonine kinases. These incessant signals induce continuous cell growth. Mutation of oncogenes in the RAS family has been associated with exposure to environmental carcinogens. Mutations of KRAS are common in carcinomas of the lung, colon, and pancreas,43 whereas mutations of NRAS occur principally in acute myelogenous leukemia and the myelodysplastic syndrome.44
Activating point mutations of the BRAF gene occur in 59% of melanomas, 18% of colorectal cancers, 14% of hepatocellular carcinomas, and 11% of gliomas.45 Most of the BRAF mutations change the valine residue at position 599 to glutamic acid (V599E). This change occurs within the kinase domain of the BRAF protein, resulting in a constitutively active protein that uncontrollably stimulates the MAP kinase cascade, thereby deregulating genes involved in cell proliferation, differentiation, and survival.45,46 In melanoma, BRAF mutations can precede neoplastic transformation; several types of nevi carry BRAF mutations.
Gene Amplification
An example of gene amplification, which usually occurs during tumor progression, is the amplification of the dihydrofolate reductase gene (DHFR) in methotrexate-resistant acute lymphoblastic leukemia.47 Amplification of DHFR is accompanied by cytogenetic alterations that mirror amplification of oncogenes.48,49 The amplified DNA segment usually involves several hundred kilobases and can contain many genes. Members of four different oncogene families are often amplified: MYC, cyclin D1 (or CCND1), EGFR, and RAS. MYC is amplified in small-cell lung cancer, breast cancer, esophageal cancer, cervical cancer, ovarian cancer, and head and neck cancer, whereas amplification of NMYC correlates with an advanced tumor stage.50 The t(11;14) translocation juxtaposes CCND1 and immunoglobulin enhancer elements and is characteristic of mantle-cell lymphoma.14 CCND1 amplification also occurs in breast, esophageal, hepatocellular, and head and neck cancer. EGFR (ERBB1) is amplified in glioblastoma and head and neck cancer. Amplification of ERBB2 (also called HER2/neu) in breast cancer correlates with a poor prognosis.51 A monoclonal antibody against the product of this oncogene (trastuzumab) is effective in breast cancers that overexpress HER2/neu. Table 3 in the Supplementary Appendix lists oncogenes that are amplified in different types of cancer.
Oncogenes in Cancer Initiation and Progression
When chronic myelogenous leukemia converts to acute leukemia, the malignant clone acquires an additional t(9;22) translocation, an isochromosome 17, or trisomy of chromosome 8. When follicular lymphoma becomes aggressive, the lymphoma cells often bear a t(8;14) translocation in addition to the original t(14;18) translocation. These findings support the hypothesis that most hematopoietic tumors and soft-tissue sarcomas are initiated by the activation of an oncogene, followed by alterations in tumor-suppressor genes and other oncogenes. In contrast, most carcinomas are initiated by the loss of function of a tumor-suppressor gene, followed by alterations in oncogenes and additional tumor-suppressor genes.52,53 This multistep process in human cancer has also been found in mouse models carrying activated oncogenes or inactivated tumor-suppressor genes, in which the duration and aggressiveness of the disease can be changed by introducing into the mouse genome the same sequential genetic alterations observed in human tumors. Methylation of CpG islands located in the promoter regions of a number of tumor-suppressor genes has also been considered an important epigenetic step in the process of carcinogenesis. This topic will be covered later in this series.
Oncogenes as Therapeutic Targets
Oncogenic proteins in cancer cells can be targeted by small molecules and, when the oncogenic protein is expressed on the cell surface, by monoclonal antibodies. Table 1 contains a summary of the targets and drugs (small molecules and monoclonal antibodies) being used in the treatment of a variety of human cancers.
|
(a nuclear receptor for retinoic acids57,58,59; see Table 2 in the Supplementary Appendix), retinoic acid can induce terminal differentiation and death of APL cells. This modality is called differentiation therapy. MicroRNA Genes
MicroRNA genes, unlike other genes involved in cancer, do not encode proteins. Instead, the products of these genes consist of a single RNA strand of about 21 to 23 nucleotides; their function is to regulate gene expression. A microRNA molecule can anneal to a messenger RNA (mRNA) containing a nucleotide sequence that complements the sequence of the microRNA (Figure 5). In this way, the microRNA blocks protein translation or causes degradation of the mRNA. Examples of the role microRNA plays in cancer pathophysiology involve miR-15a and miR-16-1, which are deleted or down-regulated in most indolent cases of chronic lymphocytic leukemia, suggesting an early event in the pathogenesis of this disease.59
|
Expression profiling of microRNA genes has revealed signatures associated with tumor classification, diagnosis, staging, and progression, as well as prognosis and response to treatment.61,62,63 For example, microRNA expression profiling can distinguish between indolent and aggressive forms of chronic lymphocytic leukemia,62 and expression of a small panel of microRNA genes correlates with prognosis in stage 1 lung cancer.63 Some microRNA genes that are deregulated in chronic lymphocytic leukemia have germ-line or somatic mutations in a microRNA precursor that affect the processing of short single-stranded microRNA molecules.62 MicroRNA genes can be up-regulated or down-regulated in cancer cells.64 The up-regulated genes function as oncogenes by down-regulating tumor-suppressor genes, whereas the down-regulated genes function as tumor-suppressor genes by down-regulating oncogenes. The function of microRNA genes depends on their targets in a specific tissue. A microRNA gene can be a tumor suppressor if in a given cell type its critical target is an oncogene, and it can be an oncogene if in a different cell type its target is a tumor-suppressor gene.
Up-regulation of microRNA genes can be due to amplification, deregulation of a transcription factor, or demethylation of CpG islands in the promoter regions of the gene. For example, the ALL1 (MLL) fusion proteins of acute lymphoblastic leukemia or acute myeloblastic leukemia carrying chromosome 11q23 translocations target the Drosha nuclease complex to specific microRNA genes, including miR191, thereby enhancing the processing of their microRNA precursors.65 The miR191 gene is also up-regulated in numerous types of solid cancers,64 suggesting that it is the downstream target of signal-translocation pathways involved in cancer.65 MicroRNA genes functioning as tumor suppressors can be down-regulated because of deletions, epigenetic silencing, or loss of the expression of one or more transcription factors.
The miR155 gene is overexpressed in diffuse large B-cell lymphoma,66 the aggressive form of chronic lymphocytic leukemia,62 and in breast, lung, and colon cancers. In transgenic mice carrying this gene under control of the Eµ enhancer of immunoglobulin genes,67 overexpression of miR155 causes acute lymphoblastic leukemia or high-grade lymphoma, indicating that deregulation of a single microRNA gene can cause malignant transformation.67 Since it takes several months for the tumors in these mice to become aggressive, it is likely that additional genetic alterations are needed for the development of frank neoplasia.
Members of the LET7 microRNA family, which are deleted or underexpressed in lung cancer, target RAS68; loss of LET7 results in overexpression of RAS.68 MiR15a and miR-16-1, the microRNAs that are deleted or down-regulated in chronic lymphocytic leukemia, cause overexpression of BCL2, which protects cells from apoptosis (Figure 4).69 The expression of a set of 21 microRNAs is altered in at least three types of solid tumors.64 One of these 21 genes, miR21, is of particular interest because it inhibits expression of the tumor suppressor PTEN.70 PTEN encodes a phosphatase involved in the PI3K kinase signaling pathway and is deleted, mutated, or silenced in advanced breast, lung, gastric, and prostate cancers.70
Summary
The identification of oncogenes involved in the initiation and progression of tumors has generated targets for the development of new anticancer drugs. Several new drugs, small molecules, and monoclonal antibodies directly affecting oncogene products have been developed, and more will follow. Considerable progress has been made in producing small molecules capable of inhibiting the enzymatic activity of ABL, KIT, EGFR, and ERBB2. For cases in which the oncogene products are not enzymes, it has been much more difficult to develop new agents.
The advantage of targeted therapy is the dependency of cancer cells on the oncogene product for growth and survival. Thus, cancer cells are more sensitive to the treatment than are normal cells. All targets, however, are not equivalent. It is possible to foresee the development of multiple drugs that have multiple targets involved in the development of cancer. The discovery of the involvement of microRNAs in the initiation and progression of human cancer may provide additional targets for anticancer treatments.
No potential conflict of interest relevant to this article was reported.
Source Information
From the Department of Molecular Virology, Immunology, and Medical Genetics and the Human Cancer Genetics Program, Ohio State University Medical Center, Columbus.
Address reprint requests to Dr. Croce at the Human Cancer Genetics Program and Department of Molecular Virology, Immunology, and Medical Genetics, Ohio State University Medical Center, 385L Wiseman Hall, 400 W. 12th Ave., Columbus, OH 43210, or at carlo.croce{at}osumc.edu.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
