User:Khanhhoa22/sandbox

Cancer is a disease of altered cell proliferation, in which normal homeostatic regulatory pathways are deranged. Animals rely on cell proliferation to build and maintain their bodies. Even when proliferation ceases, its potential remains latent within the genome of each cell. In animals, cells are social. They listen to cues from surrounding environment to maintain order. Cancer ignores or is deaf to these cues. Cell and tissue-level safeguards limit inappropriate proliferation. The longer-lived an animal and the more cell divisions it undergoes in its lifetime, the more likely it is to develop cancer.

Cancer is a genetic disease. Tumorigenesis is a multi-step process that involves a limited number of rate-limiting, random events. It is ultimately caused by changes to the genome of an individual cell. Cancers are monoclonal. They arise from a single altered cell. Cancer involves both genetically altered and surrounding ‘normal’ cells. These tumor-host interactions in some cases facilitate cancer growth and spread. In other cases, they limit cancer. A given tumor is a heterogeneous mix of cells of different genotypes. Only a subset of these cells are the primary ‘drivers’ of the tumor.

Genomic studies of cancer
The goal of genomic studies of cancer is to discovery the genetic changes that cause a normal cell to become a cancer. The idea is to look at not one or a defined group of genes, but it is to look simultaneously at all the genes in the genome. Three main approaches are:

·     Global gene expression profiling using microarray and RNA-sequencing to measure expression levels of all transcribed genes

·     Global DNA copy number profiling using Array Comparative Genomic Hybridization (aCGH) to measure amplification/deletion of all genes

·     Cancer genome sequencing to identify all changes in genomic DNA of tumor or can just be exome.

Cancer genomics research has revealed many mutated genes in cancers and led us to variety of methods of diagnosing and treating cancers. For example, the mutations in HER2 gene are found to correlate to a number of cancers, including ovarian and breast cancer. Tumor cells are intermixed with normal cells. They are surrounding normal epithelia, resident stroma and reactive immune cells. The tumor itself is also genetically diverse, different sub-clones evolve with different genomes. New technologies enable genomic studies on small group or even single cells.

Discovery of mammalian tumor viruses
In 1911, Peyton Rous discovered that a filterable infectious agent from an avian virus (now called Rous sarcoma virus) can induce tumors in chickens. Other tumors transmitted by filterable agents were discovered, such as Ciuffo human warts in 1907, Shope rabbit papilloma in 1933, and Bittner mouse mammary carcinoma “milk-agent” aka mouse mammary tumor virus in 1936. These discoveries led the researchers to examine the potential causal relationship between viruses and human cancers.

Identify cellular genes corresponding to the tumor viral genes
Although Rous proved that the avian sarcoma virus can induce tumors in chickens, the molecular explanation for virus-induced tumors was not quite well understood until decades later.

(Figure 1) In 1976, Bishop and Varmus speculated that the sarcoma virus’s src gene has a cellular origin. To test their hypothesis, they created a src-specific complementary DNA (cDNA) probe using reverse transcription. This virus-derived src-probe (v-src for viral src) was found to hybridize to a cellular gene of normal cell called c-src (captured “cellular” gene), suggesting that the sequence of src gene is conserved. Unlike the viral v-src genes, c-src encodes a cytoplasmic tyrosine kinase that is normally auto-inhibited. Phosphorylated tyrosine at C-terminus keeps c-src inactive until cellular signals remove this phosphorylated tyrosine to activate Src. V-Src contains mutations and C-terminus deletions that prevent auto-inhibition. The cellular gene transduced by a tumor virus is called a proto-oncogene.

Some tumor viruses as causes of human cancers, such as Hepatitis B virus (Hepatocellular carcinoma) and Epstein-Barr virus (Burkitt’s lymphoma), human papillomavirus SV40 (cervical cancer mesotheliomas). Viruses involved in human cancer are very common in human population, but cancer is usually an infrequent outcome of infection. There is usually a long interval between initial infection and cancer development. Cancers are usually clonal. Chemical and physical agents act as cofactors. In conclusion, these viruses are not “complete carcinogens”. DNA viruses involved in activation of proto-oncogene are about 15% of human cancers. The other 85% of human cancers are not caused by viruses. Since proto-oncogenes are conserved in animals, they can be transformed into oncogene by other mechanisms that are not necessary done by viruses.

Transformation of proto-oncogene to oncogene by mutation, translocation and intrachromosomal rearrangement
Ras is the most frequently mutated oncogene in human cancer. Ras is a small GTPase that functions as a cellular switch. Mutations, e.g. G12V, inactivate the GTPase activity of Ras and lock it in the active (ON) state. In an experiment, an activated ras transgene is microinjected into pro-nucleus and then put into mice. After 6 months to 1 year, hyperplasia benign tumors can be seen in these transgenic mice. Eventually these tumors develop into cancers.

Another example of proto-oncogene becoming oncogene is the abl/Philadelphia chromosome paradigm by gene fusion and chromosome translocation. (Figure 2) The Philadelphia chromosome results from a 9-22 reciprocal translocation, which is detected in chronic myelogenous leukemia (CML) and other leukemia. The translocations that activate the ABL tyrosine kinase occur in several different leukemia. BCR-ABL is a hyperactive kinase for at least 2 reasons: 1) loss of normal N-terminal auto-inhibitory domain and 2) BCR induces clustering, which increases auto-activation by other associated BCR-ABL kinase molecules. A drug that targets activated ABL catalytic domain of BCR-ABL causes remission of CML is called Gleevec.

Translocation can also affect proto-oncogene expression. An example is where myc is translocated to immunoglobulin heavy chain locus. This translocation can also occur to one of the two light chain loci. (Figure 3) Myc translocated to the IgH locus comes under the control of the strong IgH enhancer, losing its own enhancer in the process. Although this myc translocation leads to no structural change to the protein coding sequence, it results in deregulated expression of Myc. In many cases, the translocation appears to result from mistakes in V-D-J joining or class-switch recombination.

Amplified segments can be detected by array comparative genome hybridization (aCGH). It can be carried out on intact chromosomes. (Figure 4) Genome wide array CGHs can help identify candidate oncogenes by looking for amplified or deleted genes For example, MYCN, the c-myc gene, is amplified in about 22% of neuroblastomas, and the level of MYCN amplification strongly correlates with poor outcome. Another example of amplification of oncogene is ERBB2 or HER2 that encodes epidermal growth factor receptor. It is amplified in about 30% of breast cancers, especially the more aggressive tumors. Monoclonal antibodies, e.g. Herceptin, are used to treat cancer patients with amplified ERBB2.

The Cancer Genome Atlas
The Cancer Genome Atlas (TCGA) has improved our understanding of the biology of cancer through the application of genome analysis technologies. The goal is to sequence 500 primary tumors per adult tumor type and then compare the sequence of tumor DNA with that of matched DNA from normal tissue of the same patient. One of the major findings is that each tumor contains multiple mutations, and not all of these mutations play a causal role in cancer progression. Some are “driver” mutations and some are “passenger” mutations. Mutations that are recurrent are regarded as drivers. Most mutations specific to a single tumor are regarded as passengers, and they randomly occur in cancer with no functional effects. The Cancer Genome Atlas showed that the “average number of mutations in mutated genes varies across tumor types; most tumors have 2-6, indicating that the number of driver mutations required during oncogenesis is relatively small”

Deregulation of cell cycle checkpoints in cancers
(Figure 5) In the early 1970s, Hartwell screened for cell division defective mutants and identified key and conserved regulators of the eukaryotic cell cycle. He discovered one of the most important cdc (cell division control) genes--cdc28. The cdc28 mutant yeast had a large cell phenotype because its growth was not blocked, but its progression into S phase and mitosis was. Cdc28 is concluded to control G1/S (start of the cell cycle) as well as G2/M transition. This screen also revealed that there are checkpoints in the cell cycle that ensure physiological normality before proceeding.

Uncontrollable cell proliferation is one of the hallmarks of cancer. The homologous cdc gene of yeast in human is the cdk genes. In cancer, CDK proteins and pathways are consistently altered. Upregulation of CDK4 pathway has been found in more than 90% of melanoma cases. Levels of Cyclin D1, which works along with CDK4 at the G1/S checkpoint, are increased in ½ of breast cancers. Endogenous CDK inhibitors, e.g. p16INK4A, play an important role in regulating the activities of CDKs in order to keep the cell cycle in check, however they are often silenced in tumors. p16INK4A is an inhibitor of CDK4. Its promoter is often hyper-methylated (silenced), resulting in upregulation of CDK4 that then promotes cell proliferation in tumors.

Tumor suppressors
By binding to CDK proteins and preventing progression of cell cycle, CDK inhibitors help suppress the development of tumor cells. The existence of tumor suppressor genes can be traced back to the cell fusion experiment done by Henry Harris in 1969. (Figure 6) In this experiment, a normal cell is fused with a cancer cell, forming a tetraploid heterokaryon that was then injected to mice. It didn’t form tumors in these mice, suggesting that normal cells contain genes that constrain or suppress cell proliferation. The same observation was observed in human cell fusion experiments. The first tumor suppressor gene was described in Drosophila. In 1967, Gateff and Schneiderman reported that fly cells harboring a recessive mutation called lethal giant larvae (lgl) grow rapidly and invasively, and behave like a malignant tumor. The first human tumor suppressor gene was identified through the study of a rare childhood tumor, retinoblastoma. The inheritance of familial retinoblastoma conforms to the behavior of a Mendelian dominant allele. In 1972, Knudson hypothesis shows that the rate of familial retinoblastoma is consistent with one single random event while sporadic cancer appears to require 2 random events. From 1978 to 1981, several studies of the chromosome deletions in a number of retinoblastoma tumors assigned the chromosome location of the Rb gene to 13q14.11 (Sparkes et al, 1980; Ward et al, 1984). The esterase D gene is localized to the same band on chromosome 13. In 1986, Wen-Hwa Lee and his colleagues cloned the esterase D gene first. In 1987, Wen-Hwa Lee reported in Science that Rb gene is expressed in normal fetal retina and other tumor cells, but this expression is altered or absent in all retinoblastoma cells examined. Sequencing of the RB gene from more retinoblastoma tumor uncovered small internal deletions and point mutations that inactivate Rb function. Subsequent studies found that Rb is mutated in many types of human cancers.

Retinoblastoma is often detected before the age of 5. The theoretical frequency of two random mutation events that have to occur in sporadic retinoblastoma (10-12 per cell generation) does not agree with the observed frequency of the human disease. It has been shown that inactivation of the second allele of a tumor suppressor gene often occurs at a higher frequency through a mechanism called loss of heterozygosity. Loss of heterozygosity can be achieved through multiple mechanisms, such as mitotic recombination, allelic deletion, and mitotic nondisjunction (Hagstrom et al., 1999; Jasin, 2000; Lasko et al., 1991). Tumor suppressor genes can also be silenced by DNA methylation. As the tumor becomes more malignant, the overall level of methylation throughout the genome decreases. Meanwhile, the promoter region of tumor suppressor genes experience hyper-methylation, resulting in silencing of tumor suppressor genes.

Another important tumor suppressor is p53 protein. P53 target genes include genes in cell cycle arrest, apoptosis, DNA repair and anti-angiogenesis. Cells lacking a functional p53 tend to accumulate mutations at an unusual high rate, leading to genomic instability. People with Li-Fraumeni syndrome, first discovered in 1982, are susceptible to a wide array of cancers due to TP53 mutations. Majority of tumor-associated p53 mutation are missense mutations rather nonsense mutation. The mutant p53 alleles in the tumor cells may be dominant negative over the WT p53 allele. This dominant negative effect of mutant p53 is due to the fact that P53 forms homo-tetramers, and this oligomerization is required for its function. (Figure 7) With equal amount of WT and mutant p53, only 1/16th of the p53 complexes in the cell contain only WT p53. Missense mutants are more efficient than null mutants in that it inactivates 15/16th of p53 rather than leaving ½ of p53 in the WT form. Loss of heterozygosity often gets rid of the left-over 1/16th of the functional p53. P53 half-life of less than 20 minutes, indicating a high degradation rate—an unstable protein. MDM2 maintains the short half-life of p53 in normal unstressed cells by activating p53 degradation. MDM2’s activity can be inhibited by p14ARF. P14ARF expression is activated by excess amounts of E2F. Deregulation of Rb leads to excess E2F, signaling a danger to the organism. Excess E2F activates p14ARF, which then increases the expression of p53 and ultimately leads to apoptosis, which serves as the second line of defense to get rid of the tumor cell. Since MDM2 inhibits the tumor suppressor p53, it is considered as an oncogene. In human cancers, a single nucleotide polymorphism at nucleotide 309 increases binding of a transcriptional activator to the mdm2 promoter, leading to elevated levels of MDM2 protein, which then reduces p53 level and its response to stress signal. Smoking, UV, and aflatoxin have been shown to abolish the DNA binding and transcription activity of p53.