In late 1993, two groups at Johns Hopkins and Harvard were interested in a disease called Hereditary Non-Polyposis Colon Cancer (HNPCC), a form of colon cancer that may account for up to 15% of all colon cancers, about 20,000 cases per year in the US. It's called non-polyposis colon cancer to contrast it with a hereditary form of colon cancer that is preceded by a profusion of pre-malignant growths called polyps, as described in Lodish et al. (see Fig. 24-6). Because the polyps are often precursors to full blown colon cancer, they are often removed surgically when they are discovered (as happened to Pres. Reagan), but HNPCC gives no advanced warning that it will occur. So the folks looking for the cause of HNPCC did two things. First they studied families in which it occurred to see if they could identify the location of the mutant gene(s), and a Finnish group led by Albert de la Chapelle mapped one such gene to chromosome #2. Others looked at the DNA of tumor cells and discovered that they had an unusual abundance of "microsatellite" DNA which is DNA in which a two or three base pair sequence is repeated over and over. This kind of problem is supposed to be corrected by the mismatch repair system, because it arises by "slippage" of DNA polymerase during replication, resulting in a loop of unpaired bases. What might one conclude is the cause of the microsatellite DNA in the cancer cells--that the fixit mechanism is itself defective. It was known that the gene MSH2 was located on chromosome 2 in the same region where de la Chapelle mapped the mutation in HNPCC, and sure enough when the MSH2 gene was sequenced in people who had the disease in those families where it mapped to chromosome 2 (about 50-60% of all HNPCC patients), they were found to have mutations in MSH2 itself. This result was reported simultaneously by two groups--Vogelstein’s at Hopkins and Kolodner's at Harvard in December 1993. Three months later both groups announced that they had discovered a different gene that was involved in another 30% of HNPCC cases, and it was the gene for what? MLH1 (Kolodner collaborated with Mike Liskay of OHSU, who had been studying DNA repair in bacteria and who had cloned the bacterial Mut genes) . Thus most cases of HNPCC (and several other kinds of cancer, including ovarian, kidney and uterine cancers) are caused in part by defects in the DNA repair system. This shows that many cancers arise if a cell's own repair system becomes unable to detect and correct errors in DNA. In other words, even though the mutations show up in oncogenes or tumor suppressor genes and this is the immediate cause of cancer, it may be that the fundamental cause is inactivation of the DNA repair system in a cell, which allows other kinds of mutations to accrue at a high rate. As you might guess, DNA repair enzymes are suddenly very in with cancer researchers who are trying to identify all the DNA repair enzymes in humans and to determine if they are frequently mutated in cancer cells. So one hot area of study is DNA repair.
A second hot area of study is programmed cell death or apoptosis, which we've previously discussed. The finding that heated things up was the discovery also in the late 1980s that an oncogene isolated from cancerous B lymphocytes (lymphomas) and named B cell lymphoma gene #2 (Bcl-2) was an essential regulator of apoptosis. Bcl-2 normally inhibits cell death by inactivating caspase 9. Overexpression of bcl-2 in transgenic mice blocks cell death and therefore causes excess cell division, and it is thought that mutations in bcl-2 help to cause cancer by interfering with the apoptosis of defective cells. As we'll see in a few minutes, cells that have defective or damaged DNA are normally programmed to die, but a mutant bcl-2 can let these cells survive and proliferate, and of course they are more likely than most cells to accumulate mutations leading to cancer. A third line of evidence was that a known oncogene, myc, is an inducer of apoptosis. Overexpression of myc in transgenic mice causes massive cell death. Thus the guess is that mutations in myc also block cell death and allow defective cells to survive and that's why myc is an oncogene. So suddenly cell death became very interesting to cancer researchers.
Having now briefly summarized the suspected roles of DNA repair and of cell death genes in the causation of cancer, I want to describe a remarkable protein, that is instrumental in many cases of human cancer and whose many diverse functions keep turning up new surprises almost daily. The proteins is called p53, because it is a protein of 53, 000 molecular weight. It was discovered because its gene is often mutated in cancer cells. In fact it is estimated that at least half of all human cancers involve a defect in p53, and some estimates go as high as 80%. This of course suggests that it is somehow of central importance in the life of the cells, and increasingly the evidence supports that suggestion.
For a variety of reasons, p53 is known to be a transcription factor, i.e., a protein that regulates the transcription of genes and thus the production of other proteins (See Fig. 24-21). It has been crystallized and shown to bind to DNA and it has a sequence that appears to fit well into the major groove of DNA; indeed an common p53 binding sequence of nucleotides has been described. In addition many (but not all) of the effects of p53 seem to require transcription, because they're blocked by inhibitors of transcription. These results imply that p53 activates transcription of other genes, and that the proteins from these genes are the mediators of its effects.
What's known about p53. First of all, in normal cells, the levels of p53 are low, though it's rate of synthesis is pretty high. What does that imply? That p53 is unstable, and turns over rapidly. In fact it turns over with a half life of about 20 minutes; i.e., in a 20 minute period half the p53 molecules existing at the beginning will be gone at the end. (How would you measure that?) That's because p53 interacts with another protein called Mdm, that binds the N terminus of p53 and promotes ubiquination (and hence degradation of p53) (Fig. 24-21). However, if you put the cell under severe stress, such as exposing it to ultraviolet radiation, which causes structural damage to DNA, the levels of p53 increase dramatically because its rate of degradation drops greatly, as a result of phosphorylation of the p53 at a number of different residues (see Fig. 24-21 again), probably by a protein called ATM, among others (which stands for Ataxia Telangiectasia--mutated, a protein that was identified as mutant in a hereditary disease that increases susceptibility to cancer (Table 24-1)). The phosphorylated p53 no longer can bind Mdm, and so it is not ubiquitinated and degraded as rapidly; this prolongs its lifetime in the cell. The increase in p53 concentration in the cell allows p53 to activate transcription of many cellular proteins, including p21cip, which is an inhibitor of the cyclin-dependent kinases that are required for the cell to enter S and M phases (Fig. 13-36). The damaged cell freezes in the G1 or G2 phase of the cell cycle and doesn't enter S phase or M phase unless its DNA is repaired. If the repair doesn't occur within some amount of time, the exact duration of which isn’t yet known, the cells commits suicide by apoptosis. This indicates that there is a mechanism for detecting DNA damage in cells and preventing cell division, which would pass the damage on to daughter cells; if the damage is repaired then the cell cycle resumes; if not the cells undergoes programmed cell death.
As I showed you earlier, (Fig. 13-35), cells that have mutant p53 proteins often don't pause in G1 after they've been irradiated. Rather they continue through the cell cycle. It's also been found that in "knock out" mice where the p53 genes have been inactivated genetically, some cells are highly resistant to apoptosis. For example, thymocytes, which are immature T lymphocytes that are undergoing maturation in the thymus, are highly sensitive to X-irradiation, and can be induced to commit apoptosis by relatively low levels of radiation. But in p53 knock-out mice, the thymocytes can withstand 20 times higher doses of X-rays without committing suicide. These kinds of experiments suggest that p53 is involved in both cell cycle control and in control of apoptosis.
Another effect of p53 has recently been discovered. Another gene turned on by p53 has been identified and it is called Gadd45 (growth arrest and DNA damage inducible). Well, guess what? Gadd45 activates DNA excision repair, as does ERCC3, a protein to which p53 binds directly. These results suggest that p53 not only induces the block of stressed or damaged cells at the G1/S interface in the cell cycle but it also turns on some of the genes that encode enzymes that can fix the damage to DNA; it's also recently been shown to activate synthesis of ribonucleotide reductase, an enzyme necessary for synthesis of the nucleotides used for DNA repair. So p53 initiates a DNA repair program, as it holds the cell in G1 or G2.
And finally, p53 is clearly involved in inducing apoptosis. As I mentioned thymocytes in p53 deficient mice are much more resistant to radiation-induced apoptosis than normal thymocytes. In another experiment, a group introduced a temperature sensitive p53 protein into a cell line, then irradiated the cells at low temperature, when the p53 would function normally, and at elevated temperature, when p53 would be inactive. They found that the cells underwent apoptosis only at the permissive temperature when p53 was active, implying that p53 was crucial for apoptosis. One of the genes induced by p53 is Bax, a pro-apoptotic protein (Fig. 23-50) So p53 is now know or suspected to induce apoptosis, to inhibit cell division and to promote DNA repair in damaged cells, so no wonder that a defect in p53 is associated so often with cancer.
One of the cool recent trends in cancer biology has been the association of "molecular fingerprints" with particular kinds of cancer. That is, certain mutagenic agents seem to cause mutations at specific sites in specific genes, and so it's possible to determine with a high degree of certainty what the cause of some cancers was. This is especially true for cancers caused by cigarette smoking, because a major carcinogenic component of smoke is a complex organic chemicals called benzo(a)pyrene, that forms adducts with DNA; that is, it reacts with the nucleotide bases in DNA. It's been well established now that benzo(a)pyrene induces a characteristic suite of mutations into the p53 gene, at codons 175, 248 and 273. Thus it's possible to extract DNA from lung tumors and analyze the DNA sequence of the p53 genes. If those three codons are mutated, the probability is quite high that cigarette smoke was the cause of the tumor. Other mutagens affect p53 also, but cause mutations in other areas, or in different combination; as shown in Fig. 24-21, there are a number of "hot spots" where mutations that alter p53 function tend to occur. Several lawsuits are now in the courts in which the plaintiffs are using DNA evidence from their tumors to claim that they have been damaged by tobacco products. That is, tobacco companies often defend themselves by claiming that the tumor might have been caused by chemicals in air pollution, auto exhaust, etc., and not by smoking, but this defense is eliminated if the plaintiff can show that the tumor contains the benzo(a)pyrene-specific alterations of the p53 genes.
Finally, I want to mention one final protein mutations in which have recently been implicated as an important cause of cancer, though it has been . The protein is called p16, for its molecular weight, or CDKN2, for reasons I’ll describe in a minute. It was discovered about 4 years ago and shown to be an important regulator of the cell cycle. It appears to be a member of the increasingly important class of regulators called cyclin dependent kinase inhibitors, which we’ve already briefly discussed when we talked about the cell cycle in yeast. As shown in Fig. 24-19, p16 functions as an inhibitor of cyclin dependent kinases (and thus it's a tumor suppressor). Loss of p16 allows uncontrolled progression through the cell cycle, in ways similar to loss of Rb or overexpression of cyclin D1. When it was first discovered (in 1993), people reported finding that p16 was mutated in over 80% of "transformed" (i.e., cancer-like) cells in tissue culture, but it is not so common in tumors though some kinds of tumors, like melanomas and esophageal cancer, typically contain inactive p16 proteins.
To summarize: it appears that most cancers are caused by mutations in cellular genes. The rest are probably caused by viruses that either overactivate or inactivate cellular genes or that introduce viral genes that promote cell growth. The cellular genes involved in cancer mostly fall into two broad categories--oncogenes and tumor suppressors. Oncogenes encode proteins that are involved in mitogenic pathways and/or in movement of cells through the cell cycle. Overproduction or overactivity of these proteins will stimulate cells to divide in inappropriate circumstances. The other kind of mutated genes are called tumor suppressors. These mostly code for proteins that normally inhibit cell division or that are activated in damaged cells to block cell division and to induce apoptosis (programmed cell death). When these tumor suppressor genes are inactivated, then the brake that they apply to prevent excess cell division is released, which in turn causes overgrowth of the cells. Finally, a third form of cancer-inducing mutation is a defect in the normal DNA repair system of cells. Defects in this system allow mutations to accumulate at a much higher rate than normal, greatly increasing the chance that a mutation will occur in oncogenes and tumor promoters, leading eventually to cancer. Cancers appear to be caused by the accumulation of several mutations in various oncogenes and/or tumor promoters--not just one and perhaps as many as half a dozen or so--and it’s presumably the combination of overstimulation of cell division and loss of inhibitory brakes on cell division that leads eventually to the uncontrolled growth called cancer.
No comments:
Post a Comment