| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Biology Behind |
Author's Affiliation: Section of Hematology/Oncology, University of Chicago, Chicago, Illinois
Requests for reprints: Ezra E.W. Cohen, Section of Hematology/Oncology, University of Chicago, 5841 South Maryland Avenue, MC2115, Chicago, IL 60637. Phone: 773-702-4137; Fax: 773-834-0188; E-mail: ecohen{at}medicine.bsd.uchicago.edu.
In the motion picture series Star Wars, midichlorians are fictional intracellular organisms whose energy, the Force, could be harnessed by the Jedi knights, bestowing upon them extraordinary abilities. The creator of the series, George Lucas, has stated that the name and function for the midichlorians derived from mitochondrion. Mitochondria are the intracellular organelles responsible for energy production, the majority of cellular reactive oxygen species generation, and regulation of two forms of programmed cell death, apoptosis and autophagy; thus, perhaps Lucas' analogy was apropos.
In this issue of Clinical Cancer Research, Mithani and associates point out that mitochondria mutate late in head and neck squamous cell carcinomas (HNSCC; ref. 1). Over a decade before Lucas was born, Otto Warburg, in 1931, observed that cancer cells consume oxygen via mitochondrial oxidative phosphorylation at much higher rates than normal cells and hypothesized that cancer was actually the result of mitochondrial defects (2). Warburg's observations regarding oxidative phosphorylation have been confirmed and attributed largely to somatic nuclear DNA mutations and oncogenes (3). It was not until relatively recently that mitochondrial DNA (mtDNA) mutations have been linked with neoplasia.
The number of mitochondria and mtDNA copies per cell vary vastly depending on the cell type. The mitochondrial genome, which is exclusively maternally inherited, has been completely sequenced and consists of 16.6 kb of circular double-stranded DNA that encodes 37 genes, including 13 polypeptides essential to oxidative phosphorylation, 2 rRNA, and 22 tRNA. In fact, the maternal lineage of mtDNA allows definition of haplotypes and groups of similar haplotypes, called haplogroups, that, in turn, trace the origin of modern man to Africa 
200,000 years ago (4). Mutations in mtDNA are common and likely the result of the proximity of the mitochondrial genome to reactive oxygen species generation, making it especially vulnerable to oxidative damage, and the limited repair capacity of mtDNA. Mutations in mtDNA can be pathogenic, resulting in cell death; adaptive, conferring some advantage to the cell in its environment; or neutral (5). mtDNA mutations occur in nonmalignant tissue and have been associated with aging and environmental injury (6). In addition, germ line mtDNA mutations have been linked with specific diseases, including Leber's hereditary optic neuropathy, which is characterized by maternally transmitted, bilateral, central vision loss in young adults (7).
In 1996, Horton et al. (8) described a somatic mtDNA mutation in renal cell carcinoma patients comprising a critical 264 bp deletion in the first subunit (ND1) of NADH:ubiquinone oxidoreductase (complex I) of the electron transport chain. This report suggested that mtDNA mutations have functional significance and contribute to the cancer phenotype. There is evidence that mtDNA mutations occur in many cancers, including brain, lung, breast, ovarian, cervical, prostate, esophageal, gastric, colorectal, hepatocellular, pancreatic, bladder, thyroid, and HNSCC (5, 9). The mutations can be found in coding and noncoding regions and can be synonymous or nonsynonymous. As yet, it does not seem that a specific mutation is associated with a specific cancer site or vice versa. Interestingly, certain polymorphisms in germ line mtDNA seem to confer an increased risk of cancer, including breast cancer in African-American women, endometrial cancer, and prostate cancer (5).
The report in the current issue of Clinical Cancer Research from Mithani et al. is unique with respect to analysis of mtDNA mutations in matched normal, dysplastic, and malignant tissue, allowing one to theoretically assess the timing of these events with respect to cancer development. In fact, it is the same group of investigators who have proposed a widely cited model of HNSCC carcinogenesis based on protein and nuclear DNA alterations (10). In the current study, the mtDNA was sequenced from 12 patients in resected specimens that contained histologically normal, dysplastic, and malignant tissue. The 12 cases were selected based on the presence of known mtDNA mutations and a total of 23 mutations were observed in the tumor specimens, many of which resulted in amino acid alterations. Interestingly, only one third of the dysplastic and one sixth of the normal tissues contained mutations. Thus, the majority of mtDNA mutations were only identified in tumor tissue. Moreover, there were no unique mutations found in the dysplastic or normal specimens that were not present in the respective cancers. These data suggest that most mutations in mtDNA occur late in HNSCC carcinogenesis.
However, to place mtDNA mutations into context, one must also discuss the cell's natural process for eliminating dysfunctional organelles, autophagy. Derived from the Greek term for self-digestion, autophagy is the active process of detecting defective cytoplasmic material or organelles that are sequestered into double or multimembrane vesicles and delivered to cellular lysosomes for degradation and recycling (11). In long-lived postmitotic cells, such as neurons and myocytes, autophagy is thought to play a critical role in impeding age- or disease-related degeneration (12). Damaged mitochondria, through a process that is not completely understood, are specifically targeted, thus preserving energy homeostasis, reducing reactive oxygen production, and preventing cell death.
The functional significance of autophagy in cancer development and survival is currently being elucidated (13, 14). Evidence suggests that the process is crucial in preventing neoplastic transformation perhaps by eliminating inefficient mitochondria that produce higher amounts of mutagenic reactive oxygen species. Conversely, autophagy has also been cited as a mechanism of energy conservation that allows cancer cells to survive in nutrient-poor or hypoxic environments and as a protective process upon exposure to antineoplastic therapy. The interplay between mtDNA mutations and autophagy, therefore, seems to be complicated. It is entirely plausible that susceptibility to mtDNA mutations is an early element in carcinogenesis; however, these mutations are not observed in normal tissue due to autophagic elimination of defective mitochondria or cell death through autophagy or apoptosis. The mtDNA mutations that are detected, therefore, must not trigger autophagy or are being produced at rates that exceed clearance. The precancerous or malignant cell is protected against accumulation of defective mitochondria simply because each cell division has a dilution effect. Nevertheless, at some point, the neoplastic cell begins to accumulate mitochondria and mtDNA mutations. This cell's ability to survive in the presence of mtDNA mutations implies an inherent difference in autophagy or a survival advantage conferred to the cell by specific mtDNA mutations. The two, of course, are not mutually exclusive.
Concerning mtDNA mutations in HNSCC, these were first described by Fliss et al. in 2000 (15) with further studies shedding light on the nature of these events. It seems that mtDNA mutations are more common in noncancerous oral mucosa that has been exposed to known HNSCC risk factors, for example, betel quid (16–19) and tobacco (20, 21), as opposed to nonexposed normal mucosa. In fact, the site and type of mutation may vary with the carcinogen involved. Mucosa harboring premalignant changes has also been found to contain mtDNA mutations with the prevalence increasing in carcinoma in situ compared with hyperplasia (22). In addition to mtDNA mutations, mtDNA content per se seems to increase with increasing histopathologic grade from mild to severe dysplasia (23). Therefore, detection of mtDNA mutations has been proposed as a tool to guide early diagnosis of HNSCC, identify individuals at greater risk from carcinogen exposure, and establish clonality of separate lesions.
Of the two mutations described by Mithani et al. in normal tissue, one was nonsynonymous in the ND1 gene implicating functional consequence. Mutations in the ND1 gene have been previously described in prostate, pancreatic, colon, and thyroid cancers but not HNSCC (5, 9). One could hypothesize that the mutation found in adjacent normal mucosa in patients with HNSCC occurred early in the course of the disease and conferred a survival or proliferative advantage to the malignant clone. This is, of course, speculative, although it is striking that many mtDNA mutations, irrespective of the genomic region involved, are identical to adaptive mutations found in the population (5). This parallel phylogeny between normal adaptation and malignancy again implies a functional significance to these alterations that provides a survival advantage.
HNSCC affects 40,000 individuals in the United States each year but is a greater worldwide public health issue with an overall incidence that makes it the sixth most common cancer. Early-stage HNSCC is often curable with single modality therapy—radiotherapy or surgery—and relatively low long-term morbidity. Advanced-stage disease requires an intensive multimodality approach that yields a 40% to 50% long-term survival with treatment often compromising speech or swallowing function (24). A high-throughput, reliable, sensitive, and accurate method of detecting head and neck lesions at high risk for progression or predicting behavior of known cancers to guide intensity of therapy would be a welcome advance in the management of these patients. Adding to the complexity in HNSCC is that it is really a disease that encompasses several subsites—nasopharynx, oral cavity, oropharynx, hypopharynx, larynx, and paranasal sinuses—each with its own unique biology and behavior.
At this time, the analysis of mtDNA mutations is not mature enough to yield clinical utility. Nevertheless, the mitochondrial genome does offer one large advantage over its nuclear counterpart to researchers at present and to clinicians in the future. mtDNA contains only 16 kb and the entire content can be sequenced or arrayed using high-throughput methods (25). A gene array chip is now commercially available that can sequence the mitochondrial genome in just 48 h. A derivative of this relative facility in sequencing is the creation of a public database of mtDNA population variation in human evolution and disease, the MITOMAP.1 This internet-based tool provides the user-detailed information on mtDNA sequence, function, polymorphisms, haplogroups, and mutations—both germ line and somatic associated with disease states.
The fact that mtDNA mutations occur with relative frequency in cancer is indisputable. The next steps in our understanding and application of these observations are to further catalogue the location and type of mutations in specific cancers, determine resultant functional derangements, and develop therapeutic tools to target mitochondria. Just as in the case of nuclear DNA, it is very likely that patterns will begin to emerge associating specific mutations to particular cancers. Already, it seems that mutations in cytochrome oxidase subunit 1 are more common in prostate cancers (5, 9). At this point, however, due to the relative paucity of studies and samples analyzed, the data are mostly descriptive and definitive conclusions are difficult to draw. The sample size in the analysis from Mithani et al., as a case in point, was only 12 subjects.
Although the function of proteins encoded by mitochondrial genes has been described, the significance of the great majority of mutations in cancer is unknown. It is likely that some, if not many, of these mutations alter the apoptotic and energy balance within the cell, participate in carcinogenesis, and influence the malignant phenotype. It is also possible that some mutations, per se, alter the autophagic mechanisms within the cell. Studying the biochemical consequences of mtDNA mutations, however, is not straightforward as expressing or suppressing genes exclusively in the organelle can be difficult. One method is to use cytoplasmic hybrids that fuse whole cells with enucleated cells that have been depleted of mitochondria (26). Another is to use expression vectors that contain mitochondrial targeting sequences (27). Delineation of the functional consequences of mtDNA mutations would represent major progress in our understanding of their significance.
Furthermore, as we begin to understand which mutations contribute to the malignant phenotype, we will also explore therapeutic applications. In this context, researchers are identifying methods of targeting mitochondria through transfected targeting sequences, restriction endonucleases, and nanoparticles.
Thus, although it may seem like mitochondria evolved a long time ago in a galaxy far, far away, we are just at the beginning of an age of explosive discovery regarding the role of mitochondria and mtDNA mutations in malignant disease. Future breakthroughs will lead to a better understanding of the altered energy utilization and consumption in cancer cells; of their intrinsic resistance to apoptosis; and, eventually, to recognition of which cells will behave like Darth Vader and how to preferentially eliminate them without harming cells that turn out to act like Luke Skywalker.
Footnotes
Commentary on Mithani et al., p. 4331
Received 5/ 2/07; revised 5/17/07; accepted 5/23/07.
References
Key Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
