Survey							
                            
		                
		                * Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
INTRODUCTION Cancer is one of the three leading causes of death in industrialized nations. It is not just a single disease, but many diseases together. Cancer cells are the progeny of a single transformed cell that undergo unregulated cell proliferation. Its two main characteristics are uncontrolled growth of the cells in the human body and the ability of these cells to metastasize or migrate from the original site and spread to distant sites. If the spread is not controlled, cancer can be fatal. By definition, cancer is considered as a disease of the genes. A gene is a small part of DNA, which is the master molecule of the cells. Many of these genes produce proteins that are involved in controlling the process of cell growth and division. Cancer is caused by mutations in these genes (Kumar et al., 2010). Yet, there are two key differences between cancer and other genetic diseases. First, cancer is caused mainly by mutation in somatic cells, whereas other genetic diseases are caused solely by mutation in the germ line. Secondly, an individual cancer does not result from a single mutation, but rather from the accumulation of as few as 3 to perhaps as many as 20 mutations, depending on the type of cancer in genes that normally regulate cell multiplication. Because years may be required for many mutations to accumulate, cancer is mainly a disease of the aged. It has been reported that about 60% of all cancers are diagnosed in people who are older than 65 years of age. The most common cancers are skin cancer, lung cancer, colon cancer, breast cancer (in women), and prostate cancer (in men). In addition, cancer of the kidneys, ovaries, uterus, pancreas, bladder, rectum and blood and lymph node cancer (leukemia and lymphomas) are also included among the 12 major 2 cancers (Carson-DeWitt et al., 2006). Hepatocarcinoma (HCC) is the 5th most common cancer worldwide with poor diagnosis and accounts for approximately 549 thousand deaths each year (Parkin et al., 1984). Throughout people’s lives, the body cells are growing, dividing, and replacing themselves. An alteration or mutation in the DNA molecule can disrupt the genes and produce faulty proteins, which causes any cell to become abnormal and lose its restraints on growth. The abnormal cell begins to divide uncontrollably and eventually forms a new growth known as a ’tumor’, also called neoplasm. Although the term ‘tumor’ may be applied to any swelling, a mass formed by clone of unwanted cells (Vincent, 1985), but some tumors do not have serious health consequences, while those composed of cells that spread throughout the body usually cause cancer. Tumors arise with great frequency, especially in older animals and humans, but most pose little risk to their host because they are localized and of small size such as warts. Broadly, tumors are of two types, benign or malignant. Benign tumors contain cells that closely resemble, and may function like, normal cells. The surface interaction molecules that hold tissues together keep benign tumor cells, like normal cells, localized to appropriate tissues. Benign liver tumor stays in the liver and benign intestinal tumors stay in the intestine (Lodish et al., 2000). A benign tumor is not considered cancer as it is slow growing and does not spread or invade surrounding tissues. Once it is removed, it does not usually recur. In contrast, malignant tumor is cancerous as it invades surrounding tissues and spreads to other parts of the body. If the cancer cells have spread to the surrounding tissues, then, even after the malignant tumor is removed, it generally recurs. The cells composing a malignant tumor or cancer express some proteins 3 characteristics of the cell type from which they arose, and a high fraction of those cells grow and divide more rapidly than normal. Some malignant tumors remain localized and encapsulated, at least for a time; an example is carcinoma in situ in the ovary and breast. Most, however, do not remain in their original site; instead they invade surrounding tissues, get into circulating system and set up areas of proliferation away from the site of their original appearance. The spread of tumor cell and establishment of secondary areas of growth is called metastasis. Most malignant tumors have the potential to invade the surrounding tissues including blood vessels and lymphatic channels and metastasize to distant sites of the body (Abercrombie and Ambrose, 1962). The transformation of normal cell into malignant cell requires a number of genetic changes, and some of the genes. Tumor suppressor genes and ‘oncogenes’ involved in this process are now known. Malignant tumors frequently kill their host because invasion and metastasis cause therapeutic failure (Marc, 1986). In malignant cells, normal cellular processes are bypassed due to the actions of a selected group of genes, called oncogenes, which regulates cellular activities. A group of these highly conserved genes exist in normal cells and are called proto-oncogenes. These genes appear to be important in regulating cellular growth during embryonic development. It is thought that in carcinogenesis these proto-oncogenes become unmasked or changed during the breakage or translocation of chromosomes. These genes, not active before, become active and functional, and in some instances lead to the excessive production of growth factors which could be important in the neoplastic stage. As Haber and Harlow (1977) explained, tumor suppressor genes (about 20, known in humans) normally act as cell’s break. In contrast to oncogenes, tumor suppressor genes encode proteins that restrain cell growth 4 and prevent cell from becoming malignant. The transformation of normal cell to cancerous cell is accompanied by the loss of function of one or more of these genes. Most of the proteins encoded by tumor suppressor genes act as negative regulators of cell proliferation which may function as transcription factors (p53 and WIL), cell cycle regulators (RB and p16), components regulating signaling pathways (NFI) or regulating RNA polymerase II elongation (VHL). Thus their elimination contributes and promotes uncontrolled growth. In healthy individual the immune system can recognize the neoplastic cells and destroy them before they get a chance to multiply. However some mutant cells may escape immune detection and survive to become tumors or cancers. Cancer is thought to be the result of several sequential events, including genetic predisposition, transformation by viruses or environmental mutagens such as radiations and chemicals, and tumor promoters. The theory of immune surveillance (Swann and Smyth, 2007) says that the immune system continually recognize and eliminates tumor cells; when a tumor escapes immune surveillance and grow too large for the immune system to kill, cancer is the result. Immune surveillance is most likely to be successful against virus-induced tumors which express foreign peptides. Tumors vary greatly in their immunogenicity (Pearson et al., 1978) and even tumors with antigens which can be recognized by the host immune system can evade immune elimination. Lack of tumor rejection by immune system is not always due to the absence of T cells which can recognize those antigens (Tanaka et al., 1999). Tumor-specific lymphocytes can be found in the blood drawing lymph nodes (as well as in tumor itself) of patient with actively growing tumors. These lymphocytes can kill tumor cell in vitro but fail to do so in vivo. 5 The presences of these cells have helped immunologists to identify antigens on some tumor cells (Janeway et al., 1999). Some cancers have a genetic basis. In other words, an individual could inherit faulty DNA from his parents, which could predispose him to getting cancer (Kumar et al., 2010). While there is scientific evidence that both factors (environmental and genetics) play a role, less than 10% of all cancers are purely hereditary. Cancers that are known to have hereditary link are breast cancer, colon cancer, ovarian cancer and uterine cancer. Besides genes, certain physiological traits could be inherited and could contribute to cancer (Frank, 2004). Cancers are generally classified into three broad groups; carcinomas, sarcomas, and leukemia or lymphomas (Cairns, 1986: 4-14). Carcinomas are cancer that arises in the epithelium (the layers of cells covering the body’s surface and lining the internal organ and various glands). About 90% of human cancers fall into this category. Sarcomas are cancer of the supporting tissues of the body, such as bone, cartilage, fat, muscle and blood vessels. Lymphomas and myeloma are cancer that begins in the cells of immune system. Leukemia is the cancer that starts in blood forming tissues such as the bone marrow and causes large number of abnormal blood cells to be produced and enter the blood. A tumor isolated from a patient is a large aggregation of cancer cells, all of them descended from a single ‘founder’ cell. The single ancestor was once a normal cell, with a function in a particular tissue. Oncogenes become activated in the conversion of the 6 normal founder cell into a cancer cell. Once activated, the genes function continuously to direct cells toward abnormal behavior that is called the cancerous state (Weinberg, 1983). 1.1 CHEMICAL CARCINOGENS AND CARCINOGENESIS A majority of cancer are caused by changes in the cell’s DNA because of damage due to the environment. Environmental factors that are responsible for causing the initial mutation in the DNA are called carcinogens. Substances that cause DNA mutation are known as mutagens, and mutagens that cause cancer are known as carcinogens. Many mutagens are also carcinogens but some carcinogens are not mutagens. Alcohol is an example of a chemical carcinogen that is not a mutagen (Seitz et al., 1998). Such chemicals may promote cancer through stimulating the rate of cell division. Carcinogenesis is a multistep process. The fundamentals of the carcinogenesis model include an initiation step involving changes at genetic levels, which is followed by promotion, conversion and progression steps to clinical malignancies (Tanaka, 1992). If the process ended and the cancerous cell did not grow and replicate, no cancer would form. Radiation is an agent directly involved in causing cancer; this is due to its ability to damage the genome or disrupt the cellular metabolic processes. Several radioactive compounds are also carcinogens, but their carcinogenic activity is attributed to the radiation such as gamma rays and alpha particles that they emit (Land, 1995). Carcinogens may increase the risk of cancer by altering cellular metabolism or damaging DNA directly in cells, which interferes with biological processes, and induces the uncontrolled, malignant division ultimately leading to the formation of tumors. 7 Usually DNA damage, if too severe to repair, leads to programmed cell death, but if the programmed cell death pathway is damaged, then the cell cannot prevent itself from becoming a cancer cell (Isaacs, 1993). Direct-acting or DNA-reactive, activation-independent carcinogens that bind covalently to cellular genomic DNA is genotoxic and are mutagenic. However, there are other carcinogens that require prior metabolism or metabolic activation to become carcinogenic (procarcinogens). In this process one or more enzymes catalyses reactions converting procarcinogens to active carcinogens (called as proximate carcinogens, an intermediary compound) and the ultimate carcinogen, which is the final compound that reacts with cellular components such as DNA (Philips, 1985). A possible sequence can be represented as follows: Procarcinogens  Proximate carcinogens  Ultimate carcinogens. Procarcinogens are the parent compounds that cause cancer, but yet are metabolically upstream from the damaging agent. Ultimate carcinogen refers to the reactive metabolite that covalently modifies DNA. Proximate carcinogen is an intermediate species between pro and ultimate carcinogen (Bradfield, 2009). 1.1.1 Nitrosamines as a common environmental carcinogen Nitrosamines are chemical compounds having chemical structure R1N (-R2)-N=O, some of which are found to be carcinogenic. Research has revealed that approximately 90% of nitrosamine compounds were rendered carcinogenic (Scanlan, 2000). Nitrosamines were first discovered in a study of workers who were exhibiting jaundice and liver damage as a result of occupational exposure to an unknown agent then. 8 Nitrosamines have been reported to be used in a number of industries, e.g. rubber industries. Many experiments on animals revealed that, dibutylnitrosamine (DBN) is hepatotoxic to the rats and caused a high incidence of liver tumors apart from bladder and colon tumors in them. In fact, it is one of the most carcinogenic compounds known after DEN (diethylnitrosamine). Based upon experiments, a mechanism was proposed that is common for all the carcinogens, i.e. metabolic bio-activation pathway to a highly reactive ‘ultimate carcinogen’ that can add an alkyl group to the O6 and N7 of guanine (Bradfield, 2009). Fig. 1: Proposed metabolic pathways of N, N-dibutylnitrosamine (DBN) in case of guanine, where DBN gets metabolic activated by cytochrome p450 enzymes in liver to become ultimate carcinogen, which adds butyl group at O6 & N7 positions resulting in mispairing of guanine with cytosine resulting in mutation in DNA. 9 It was hypothesized that O6 alkylation of guanine leads to the inability of the guanine residue to undergo normal base-paring with cytosine and thus may lead to transition mispairing, resulting in mutation (Fig.1). The inability of the system to remove O6 alkyl guanine from DNA, and therefore to reduce the level of this abnormal purine in DNA before DNA replication, may be of critical importance for determining susceptibility to carcinogenesis by N-Nitroso compounds (Arcos et al., 1982). Nitrosamines are a class of chemical compounds that were first discovered and reported by two British scientists, J. Barnes and P. Magee, who reported that nitrosamines and their derivatives like dimethylnitrosamine produced liver tumors in rats. This discovery was made during a routine screening of chemicals used as solvents in the dry cleaning industry (Scanlan, 2000). This discovery resulted in an interest among scientists around the world to investigate and identify the carcinogenic properties of other nitrosamines and N-nitroso compounds. Approximately 300 of these compounds were tested, and 90% of them were found to be carcinogenic when experimented in animals. Most nitrosamines are mutagens, often organ specific. For example, dimethylnitrosamine causes liver cancer in experimental animals, whereas some of the tobacco specific nitrosamines cause lung cancer. A number of nitrosamines are transplacental carcinogens also. Since nitrosamines are metabolized in the same way in human as well as animal tissues, it seems highly likely that humans are susceptible to the carcinogenic properties of nitrosamines (Scanlan, 2000). It has also been found that carcinogenic N-nitroso compounds are 10 formed from the reaction of naturally-occurring amines and nitrites that may be added to foods or produced by bacterial reduction of nitrate. N-Nitroso compounds can be produced during processing, storage and preparation of foods as well as in the mammalian stomach (Issenberg, 1976). It is said that occurrence of nitrosamines are commonly because of their chemical precursors, amines and nitrosating agents, and the chemical reaction for nitrosamine formation is quite facile. Cured meats can contain nitrosamines because meats contain amines, and sodium nitrite, a source of nitrosating agents, added to cured meats as a preservative. Of all the cured meats, bacon has received the most attention. It was found to contain detectable levels of nitrosamines, like nitrosopyrrolidine and, to a lesser extent, dimethylnitrosamine (Larsson et al., 2006). It was only in the late 1970s that attention was focused on the occurrence of nitrosamines in cured meats, and the removal of sodium nitrite as a food additive. However, the prospect of sodium nitrite removal presented a formidable dilemma for the regulatory agencies; sodium nitrite removal is supposed to prevent nitrosamine formation, but it might also increase the risk of botulism poisoning. Sodium nitrite and sodium chloride together are particularly effective against Clostridium botulinum. The solution to the dilemma was to limit the addition of sodium nitrite to 120 ppm, the lowest level found to be effective in controlling growth and toxin production by Clostridium botulinum (Scanlan, 2000). Researchers in late 1960s at the University of Nebraska Medical Center were studying nitrosamine formation from a drug called aminopyrine. Mysteriously, when they used a new batch of aminopyrine, no nitrosamines were formed. Further investigation 11 revealed that the new batch of aminopyrine was formulated with ascorbic acid as a preservative, whereas the original batch that readily formed nitrosamines was not. Nitrosamines can form in the gastric juice of the human stomach. This is commonly known as endogenous nitrosation. Bacteria in the mouth chemically reduce nitrate, which is prevalent in many vegetables, to nitrite, which in turn can form nitrosating agents. Many foods found to contain amines that can react with nitrosating agents in the acidic stomach to form nitrosamines but it has been demonstrated that ascorbic acid can reduce nitrosation in the stomach (Scanlan, 2000). The potency and target specificity of an NNitroso compound depends not only on the structures of the chemical but also on the dosage and treatment protocol, the route of administration, the animal species, and various other factors such as the age, sex and diet of the animals (Arcos et al, 1982). Nitrosamines are carcinogenic in animals. A 1981 report from the National Academy of Sciences (NAS) estimated that the per capita exposure is about 1 g/day from foods and beverages, mainly from fried bacon and beer. Current exposure is probably closer to 0.1 g/day due to successful efforts over the past 20 years to reduce nitrosamine formation in foods and beverages. In contrast, the NAS report estimated an exposure of 17 g/day from cigarette smoking, although the use of filters has somewhat lowered smokers' exposure. Recent reports indicate that industrial exposure, such as found in a rubber or chemical manufacturing plant, can be relatively high (Scanlan, 2000). These investigation and findings provide indirect evidence indicating that nitrosamines are human carcinogens. For instance, tobacco-specific nitrosamines are one 12 of the major groups of chemical carcinogens in tobacco products, and indicative of a causal link between tobacco use and cancer. But it is still difficult to evaluate the risk of cancer from daily exposure of 1 g of nitrosamines obtained from foods and beverages. The same difficulty applies to the risk assessment of the exposure to trace amounts of aflatoxin, polycyclic aromatic hydrocarbons, and heterocyclic amines in a variety of foods and beverages (Scanlan, 2000). 1.1.2 N-N-Dibutylnitrosamine (DBN) Chemically N-Nitrosodibutylamine is a volatile, pale yellow oily substance with a characteristic odour. It is sparingly soluble in water and is lipid soluble and miscible with hexane, dichloromethane, and many other organic solvents. It is sensitive to light, especially to ultraviolet light, and undergoes relatively rapid photolytic degradation. When heated to decomposition, it emits toxic fumes of nitrogen oxides (Hazardous Substances Data Bank, 2000; WHO-IARC, 1978). DBN is one of the potent carcinogens among the dialkylnitrosamines and has ability to induce liver and lung tumors in rodents, when administered. In mice, it has been have reported that 3 injections of 1l DBN to infant animals induced tumors of liver and lung (Wood et al., 1970). DBN is a nitrated amine derivative. Amines are chemical bases. They neutralize acids to form salts and water. These acid-base reactions are exothermic. The amount of heat that is evolved per mole of amine in neutralization is largely independent of the strength of the amine as a base. Amines may be incompatible with isocyanates, halogenated organics, peroxides, phenols (acidic), epoxides, anhydrides, and acid halides. Amine generates flammable gaseous hydrogen in combination with strong reducing agents, such as hydrides. The metabolic fate of N, N-dibutylnitrosamine (DBN) was 13 studied in the rat, to elucidate the possibility of a correlation between its metabolism and its organotropic carcinogenicity to the urinary bladder and other organs. It was extensively metabolized in the rat, no unchanged DBN being found in the urine. DNA underwent metabolic transformation in at least three ways. The major pathways demonstrated on the basis of urinary metabolites were omega- and (omega-1)-oxidations of one butyl chain to give N-butyl-N-(3-carboxypropyl) nitrosamine (BCPN) and Nbutyl-N-(3-hydroxybutyl) nitrosamine, respectively. The third minor pathway was (omega-2)-oxidation of the butyl chain to afford N-butyl-N-(2-hydroxybutyl) nitrosamine. Both hydroxylated metabolites were excreted into the urine as such and as their glucuronic acid conjugates. The omega-oxidation of DBN to BCPN is responsible for the induction of bladder tumors in rats, while the products of the (omega-1)- or (omega-2)-oxidation may be involved in tumor induction in the liver (Suzuki et al.,1980). In a study (Fujii et al., 1977) pulse doses of N-dibutylnitrosamine (DBN), Nbutyl-N-(4-hydroxybutyl) nitrosamine (BBN) and N-butyl-N-(3carboxypropyl) nitrosamine (BCPN) suspended in 1% gelatin, were administered subcutaneous to infant CDF1 mice, and the experiment terminated at one year of age. Tumors were induced in lungs and liver. The incidences of lung adenomas were 73-95% in all treated mice, with no sex differences. Hepatocellular adenomas and a carcinoma were found with an incidence of 81% (21/26) in DBN, 59% (13/22) in BBN, and 32% (9/28) in BCPNtreated males and the incidence was 23% (5/22) in DBN-treated females. Only one papilloma of the fore-stomach was induced in mice treated with DBN. These findings indicated that the subcutaneous administration of DBN, BBN, and BCPN induced tumors of the lung and liver, but no tumors of the urinary bladder, under these experimental 14 conditions. The carcinogenic effect on mice at the treated dose level was DBN > BBN > BCPN. 1.2 ANTIGENS ON TUMOR CELLS The discovery of immune responses to tumors was made possible by the finding that tumors could be induced in mice after treatment with chemical carcinogen or irradiation and the subsequent development of inbred strains of mice (Janeway et al., 1999). Ludwik Gross discovered that inbred mice could be immunized against a tumor that developed in a mouse of the same inbred strain (Old, 1977). Tumor antigens fall into two broad categories: Tumor specific antigens and Tumor associated antigens. 1.2.1 Tumor specific antigens (TSA) They are unique, found only on tumor cells and are eminently positioned as targets for immunologic attack. The tumor expresses antigenic peptides that can become targets of a tumor-specific T-cell response. The antigens expressed by experimentally induced murine tumors often termed as tumor-specific transplantation antigens (TSTA), or tumor rejection antigens (TRA), or tumor specific antigen (TSA), are usually specific for an individual tumors (Janeway et al., 1999). TSA are not present on non-tumor cells hence these peptides can become the targets of a tumor specific-cell response. They might be proteins normally expressed only in the embryo, but which become expressed abnormally in tumor cells in adults, or they might be normal self proteins that become over-expressed. TSA usually appears when an infecting virus has caused the cell to become immortal and to express virus antigens. TSAs not induced by virus are idiotypes of BCR or B cell lymphomas or TCR on T cell lymphomas. TSTA are peptides of tumor 15 cell proteins that are presented to T-cells by MHC molecules. Unique tumor specific antigens are found only on tumor cells and not on other cells of the host. There are several examples of unique tumor specific antigens (TSAs). 1.2.2 Tumor associated antigens (TAA): These are found in tumor cells and some normal cells as well. Antibodies contained in cancer sera react with a unique group of autologous cellular antigens called tumor associated antigens (TAAs). Cancer is as a multi-step process which involves not only genetic changes conferring growth advantage but also factors that disrupt regulation of growth and differentiation. It is possible that some of these factors can be identified and their functions evaluated with the aid of autoantibodies arising during tumerigenesis. The multi-factorial and multi-step nature in the molecular pathogenesis of human cancers must be taken into account in both the design and interpretation of efforts to identify biomarkers meant for early detection of cancer. Recent studies suggest that the combination of antibodies against a group of TAAs might acquire higher sensitivity for diagnosis of cancer. It is possible that autoantibody profiles involving different arrays of TAAs might be developed in the future and the results could be useful for cancer diagnosis (Zhang, 2004). The discovery of cell-surface antigens, which distinguish cancer cells from normal cells, lead to the starting point to explain the malignant transformation in terms of cell-surface changes and of attempts to control cancer by immunological means (Old, 1977). TAAs are more commonly found on tumor cells and on normal cells during fetal life (oncofetal antigens). In postnatal life, TAAs are present in selected organs or in many 16 cells but at much lower concentration than on tumor cells. Alpha fetoprotein (AFP) is a secreted tumor antigen and is the fetal equivalent of albumin. It is found in the serum of patients with hepatomas and teratomas and can be used as a marker for the presence of such cancers. It is, however, not a suitable target for tumor rejection (Zhang, 2004). Immune responses to TAA may be suppressed because they are considered self. In order for the immune system to react against a tumor, the tumor must have antigens that are recognized as foreign. A number of alterations occur in the cell during tumorigenesis. Due to their genetic instability, tumor cells usually express abnormal proteins, i.e., tumorassociated antigens (TAAs), which have no or very limited expression on normal cells. Such TAAs expose new, potentially immunogenic epitopes, which can be recognized by the immune system of the host. Interestingly, the endogenous immune response against these epitopes has only a marginal effect on the tumor. For example, a minority of patients suffering from colorectal cancer (CRC) have been shown to develop endogenous tumor antigen specific antibodies, and these only at low and ineffective titers. With progression of tumor growth, the cytotoxic immune response at the tumor site has been found to be inhibited, and invading T cells and antigen-presenting cells (APCs) often appear to be non-functional (Schuster et al., 2006). Most relevant from the point of view of immune-surveillance and surface membrane molecules, this might be antigenically novel for suppression of membrane proteins which are essential for immune recognition and activation (Zhang, 2004). Several TAA and other cancers have been identified and are important development in the field of cancer immunology. Better understanding of the molecular 17 mechanisms of antigen processing and presentation has allowed the development of specific vaccines (Minev et al., 1999). Though tumor associated antigens are, to a large extent, specific to the tumors that display them, some normal cells may also express such antigens at particular stages of differentiation. Monoclonal antibodies have now permitted the characterization of these antigens and are now widely used to diagnose some of these tumors. Despite the discovery of TAA, a major limitation in cancer immunotherapy has been the lack of antigens applicable to a majority of patients with common cancers, i.e. universal TAAs. Clinical approaches have generally been used to test one malignancy at a time, and in some cases (such as the immunoglobulin idiotypic antigen in B-cell malignancies), patient by patient. To circumvent this obstacle, an attempt was made to find universal antigens that could trigger CTL responses against a broad range of tumor types (Vonderheide, 2002). The strategy was used, in which tumor antigens and their CTL epitopes are deduced from genes known to be expressed selectively in tumors. TAA represent a heterogeneous group of macromolecular structures (glycoproteins, glycolipids) in or on tumor cells (LeGruce et al., 1980) and can serve an effective target for active immunotherapy against tumor. In this context, the preparations containing extracted TAA of tumor cells (TAA extract) and purified TAA have been explored for active immunization. Extraction of viable tumor cells with low concentrations of butanol has become increasingly popular for obtaining TAA. With this procedure mainly peripheral membrane components are released, thus avoiding extraction of cytoplasmic or integral membrane components (LeGruce et al., 1985). 18 Butanol extraction does not exert cytolytic effects as viability and proliferation capacity of tumor cells is preserved (Alam et al., 2001). 1.3 GLYCOPROTEINS Glycoproteins are complex compounds composed of a protein and a carbohydrate. There are many different types of glycoproteins that are found in abundance in all kinds of cells. As small molecules they are found on the exterior surface of cells. Different types of cells have specific, unique types of glycoproteins attached to them. Glycoproteins are vital to a number of important biological functions. They allow certain types of cell-to-cell communication, help co-ordinate complicated cellular responses to stimuli and activate the action of other types of cells (Craig, 2008). Many proteins found in nature are glycoproteins because they contain covalently linked oligosaccharide and polysaccharide groups. Known glycoproteins include structural proteins, enzymes, membrane receptors, transport proteins, and immunoglobulins, among others. They play many roles in cells; some of these are well known, but others yet un-discovered. The carbohydrate portion of glycoproteins is usually made of combinations of sugar molecules such as glucose, galactose, mannose and fucose, which are collectively known as oligosaccharides (ISMRD, 2009). In eukaryotic cells most proteins are subject to post-translational modification, of which glycosylation represents one of the most common post-translation events, more than half of all proteins that have been characterized are glycoproteins. The carbohydrate components of glycoproteins perform critical biological functions in protein sorting, immune and receptor recognition, inflammation, pathogenicity, metastasis, and other cellular processes (Gates et al., 2004). 19 Variations in structure and degree of glycosylation site saturation can contribute to overall mass heterogeneity. The terminal residues of these glycans are commonly Nacetylneuraminic acid (sialic acids). The degree of sialylation affects both the mass and charge of a glycoprotein. Other modifications to the protein such as sulfation or phosphorylation also affect charge. O-Linked glycans often have lower mass than Nlinked structures, but can be more abundant and heterogeneous. Mammalian glycoproteins contain three major types of oligosaccharides (glycans): N-linked, Olinked, and glycosylphosphatidylinositol (GPI) lipid anchors (Gallagher, 1985). 1.3.1 N-linked glycans N-linked glycoproteins are expressed by all eukaryotic cells. Protein glycosylation of N-linked glycans is a co-translational event, occurring during protein synthesis. Nlinked glycosylation requires the consensus sequence Asn-X-Ser/Thr. N-Linked glycans are linked to the protein backbone via an amide bond to asparagine residues in an Asn-XSer/Thr motif, where X can be any amino acid, except Pro. Carbohydrate is not attached to the polypeptide one sugar at a time. Rather, a large preformed carbohydrate containing fourteen or more sugar residues is attached to the aspargine as the protein is translated in the rough endoplasmic recticulum. The carbohydrate on the glycoprotein is then modified by enzymes that remove some sugars and attach others as a newly formed glycoprotein and moves from the rough endoplasmic recticulum to golgi apparatus and other location in the cell. Many N-linked glycoproteins eventually become part of the cell membrane or are secreted by the cell (Gates et al., 2004; Noiva, 2004). Glycosylation occurs most often when this consensus sequence occurs in a loop in the peptide. Oligosaccharide intermediates destined for protein incorporation are 20 synthesized by a series of transferases on the cytoplasmic side of the endoplasmic recticulum (ER) while linked to the dolichol lipid. Following the addition of a specific number of mannose and glucose molecules, the orientation of the dolichol precursor and its attached glycan shift to the lumen of the ER where further enzymatic modification occurs. The completed oligosaccharide is then transferred from the dolichol precursor to the Asn of the target glycoprotein by oligosaccharyltransferase, OST (Gates et al., 2004). Further processing includes trimming of residues such as glucose and mannose, and addition of new residues via transferases in the ER and, to a great extent, in the Golgi. In the Golgi, high mannose N-glycans can be converted to variety of complex and hybrid forms which are unique to vertibrates. Inhibition or elimination of glycosylation in the study of N-linked glycans can be brought about by a number of compounds. In the presence of compactin, coenzyme Q, and exogenous cholesterol, N-glycosylation is greatly inhibited. Treatment with tunicamycin completely blocks deglycosylation in that it inhibits GlcNAc C-1-phosphotransferase, which is critical in the formation of the dolichol precursor necessary for synthesizing of N-glycans (Gates et al., 2004). The diverse assortment of N-linked glycans is based on the common core pentasaccharide, Man3GlcNAc2. Further processing in the Golgi results in three main classes of N-linked glycan sub-types; High-mannose (Fig.2.1(B.1.) Fig.2.1(B.2)), Hybrid, and Complex (Fig. 2.1(C.1) and Fig. 2.1(C.2)). Complex glycans contain the common trimannosyl core. Additional monosaccharides may occur in repeating lactosamine units. Additional modifications may include a bisecting GlcNAc at the mannosyl core and/or a fucosyl residue on the innermost GlcNAc. Complex glycans exist in bi-, tri- and tetraantennary (Fig.2.1.(D)) forms (Gates et al., 2004). 21 22 23 1.3.2 O-linked glycans O-linked glycoproteins are usually synthesized by the addition of sugar residues to the hydroxyl side chain of serine or threonine residues in polypeptides in the golgi apparatus. Unlike N-linked glycoproteins, O-linked glycoproteins are synthesized by the addition of a single sugar residue at a time. Many O-linked glycoproteins are secreted by the cell to become a part of the extracellular matrix that surrounds it. O-Linked glycosylation is a true post-translational event and does not require a consensus sequence and no oligosaccharide precursor is required for protein transfer. The most common type of O-linked glycans contain an initial GalNAc residue (or Tn epitope), these are commonly referred to as mucin-type glycans. Other O-linked glycans include glucosamine, xylose, galactose, fucose, or manose as the initial sugar bound to the Ser/Thr residues. O-Linked glycoproteins are usually large proteins (>200 kDa) that are commonly biantennary with comparatively less branching than N-linked glycans. 24 Glycosylation generally occurs in high-density clusters and may contribute as much as 50-80% to the overall mass (Gates et al., 2004). . . O-Linked glycans tend to be very heterogeneous; hence they are generally classified by their core structure. Non-elongated O-GlcNAc groups have been recently shown to be related to phosphorylation states and dynamic processing related to cell signaling events in the cell. O-Linked glycans are prevalent in most secretory cells and tissues. They are present in high concentrations in the zona pelucida surrounding mammalian eggs and may function as sperm receptors (ZP3 glycoprotein). O-Linked glycans are also involved in hematopoiesis, inflammation response mechanisms, and the 25 formation of ABO blood antigens. Elongation and termination of O-linked glycans are carried out by several glycosyltransferases. One notable core structure is the Gal-(1-3) GalNAc (core 1) sequence that has antigenic properties. Termination of O-linked glycans usually includes Gal, GlcNAc, GalNAc, Fuc, or sialic acid (Gates et al., 2004). By far the most common modification of the core Gal-(1-3)-GalNAc is mono-, dior tri-sialylation (Fig.2.2(A) and Fig.2.2(B)). A less common, but widely distributed Olinked hexasaccharide structure contains (1-4)-linked Gal and (1-6)-linked GlcNAc as well as sialic acid (Fig.2.2(C)). 1.4 BIOCHEMICAL ROLE OF SIALIC ACIDS The presence of sialic acids residues as terminal components of cell surface glycoprotein and glycolipids, plays an important role in the chemical and biological diversity of glycoconjugates. Glycotransferases (sialyltransferases) property of cell-typespecific expression, leads to specific sialylation patterns of oligosaccharides which can be considered as key determinants in the makeup of cells. Many differences have been found 26 in the sialoglycosylation patterns of cells during development, activation, aging and oncogenesis. Structurally, sialic acids comprise a family of 36 naturally occurring derivatives of neuraminic acid that are usually N-acylated to form N-acetylneuraminic acid (Neu5ac) or N-glycoloylneuraminic acid (Neu5Gc) as a fundamental molecules. Oacetylation at one or several of the hydroxyl functions, at C-4, C-7, C-8, and C-9 or introduction of double bond between C-2 and C-3 are some additional modifications. Apart from cell surface membrane and body fluid of all mammals, sialic acids in these structural forms are also found in variety of organisms such as bacteria, viruses, protozoa and higher animals. The saturated sialic acids usually occupy the terminal, non reducing position of the oligosaccharide chains of complex carbohydrates on outer and inner membrane surface in various linkages mainly to galactose, N-acetylgalactosamine, and sialic acid itself. Being present on terminal end, sialic acids are among the first molecules encountered by other cells or compounds coming in contact with the cell. Modifications, especially enzymatic N-acetyl-hydroxylation, O-acetylation, and O-methylation, are results of biosynthesis of modified sialic acids (Schauer et al., 2005). Researchers have investigated on the biological role of sialic acids such as significance of O-acetylation, N-acetyl-hydroxylation, and sialic acid binding proteins. The functions of sialic acids have been grouped in three categories. The first is its physiochemical effects on the glycoconjugates to which they are bound, and on the environmental molecules in cell membranes. Therefore, sialyl groups influence and stabilize the conformation of both the glycan and the protein parts of glycoconjugates, resulting in modified properties such as higher thermal and proteolytic stability of glycoproteins. Other two functions of the sialic acids deal with molecular and cellular 27 recognition. In the case of molecular recognition, sialic acids acts as masks to prevent biological recognition, e.g., of subterminal galactose residues. In the case of cellular recognition, sialic acids serve as a recognition sites. The best example of the masking function is the binding and uptake of desialylated serum glycoproteins by hepatocytes, similarly sialidase treated erythrocytes, lymphocytes, and thrombocytes are bound and partly phagocytized by macrophages, mediated by a galactose-specific lectin. Studies have revealed that during apoptosis, the rodent thymocytes are phagocytized after the loss of cell surface sialic acids. These studies evidence the role of sialic acids in maintaining the life span of molecules and cells which, together with demasked galactose residues, regulates these and many more biological processes including the pathological events like the spreading of cancer (Schauer et al., 2005) Carbohydrates in the form of asparagine-linked (N-linked) or serine/threoninelinked (O-linked) oligosaccharides are major structural components of many eukaryotic proteins. They perform critical biological functions in protein sorting, immune recognition, receptor binding, inflammation, pathogenicity, and many other processes. The diversity of oligosaccharide structures often results in heterogeneity in the mass and charge of glycoproteins. N-linked oligosaccharides may contribute 3.5 kDa or more per structure to the mass of a glycoprotein. Variations in the structures and different degrees of saturation of available glycosylation sites in a glycoprotein all contribute to mass heterogeneity. The presence of sialic acid (N-acetylneuraminic acid) affects both the mass and charge of a glycoprotein. Other modifications to the carbohydrate such as sulfation or phosphorylation also affect charge. O-linked sugars, although usually less massive than N-linked structures, may be more numerous and are also heterogeneous in 28 structure. To study the structure and function of a glycoprotein, it is often desirable to remove all or just a select class of oligosaccharides. This approach allows the assignment of specific biological functions to particular components of the glycoprotein. For example, the loss of ligand binding to a glycoprotein after removal of sialic acid may implicate this sugar in the binding process (Gates et al., 2004). Removal of carbohydrates from glycoprotein is useful for a number of reasons: for simplifying amino acid sequence determination of glycoproteins, to remove heterogeneity in glycoproteins for X-ray crystallographic analysis, to remove carbohydrate epitopes from antigens, to enhance or reduce blood clearance rates of glycoprotein therapeutics (Szkudinski et al., 1995), to investigate the role of carbohydrates in enzyme activity and solubility, to investigate ligand binding for quality control of glycoprotein pharmaceuticals, to study the peptide portion of the glycoprotein by SDS-PAGE, etc. Both chemical and enzymatic methods exist for removing oligosaccharides from glycoproteins. Hydrazinolysis of glycoproteins (Kuraya et al., 1992), although capable of removing both N-linked and O-linked sugars, results in the complete destruction of the protein component and is, therefore, not suitable if recovery of the protein is required. Milder chemical methods such as use of trifluoromethanesulfonic acid (TFMS) even when optimized results in incomplete sugar removal and partial protein destruction (Sojar et al., 1987). The amino acid-linked sugar residue of both N-linked and O-linked oligosaccharides is retained. Only the enzymatic method provides complete sugar removal with no protein degradation. 29 1.5 NEURAMINIDASE: Neuraminidases (sialidases) belong to a family of exoglycosidases that catalyze the removal of terminal non-reducing sialic acid residues ketosidically linked to momosaccharide or olligosaccharide chains of glycoconjugates. They are widely distributed in viruses, bacteria, fungi, mycoplasma, protozoa, avian and mammalian species. Among the various sialidase species, viral and bacterial enzymes have been purified and studied extensively (Saito and Yu, 2005). For example in virus, neuraminidase cleaves off the terminal sialic acids on the glycosylated HA during virus budding to facilitate virus release. Additionally, it helps virus spread through the circulation by further removing sialic acids from the cell surface. These cleavages prevent self-aggregation and ensure the efficient spread of the progeny virus from cell to cell. Otherwise, infection would be limited to one round of replication. It is described as a receptor-destroying enzyme because it cleaves a terminal sialic acid from the cellular receptors. It may facilitate viral invasion of the upper airways by cleaving the sialic acid moieties on the mucin of the airway epithelial cells. It is likely to play a role in the budding process through its association with lipid rafts during intracellular transport. It may additionally display a raft-association independent effect on budding. It plays a role in the determination of host range restriction on replication and virulence. Sialidase activity in late endosome/lysosome traffic seems to enhance virus replication (Berman et al., 2000). Neuraminidases hydrolyse the non-reducing, terminal sialic acid linkage in various natural substrates, such as glycoproteins, glycolipids, gangliosides, and polysaccharides (Monti et al., 2002). There are two major classes of neuraminidase that 30 cleaves exo-poly-sialic or endo-poly-sialic acids; exo-neuraminidase (or exo-α-sialidase) hydrolyses (23),(26),(28) glycosidic linkages of the terminal sialic acid residues (Schauer, 1982) and endo-neuraminidase (or endo-α-sialidase) hydrolyses (28)--sialosyl linkages in oligo or poly(sialic) acids (Cabezas,1991). In mammals, neuraminidases occur in the lysosome and the cytosol, and are associated with the plasma membrane. Neuraminidases have also been known to be a factor in the pathogenesis of many diseases. For example, in viruses neuraminidases enable the transport of the virus through mucin, the eruption of the virus from the infected host cell, and the prevention of self-aggregation of virus particles through the destruction of the host cell receptor recognized by the virus (Bert et al., 2003). Eukaryotic, bacterial and viral neuraminidases share highly conserved regions of beta-sheet motifs. Bacterial neuraminidases often possess domains in addition to the catalytic neuraminidase domain; for instance the neuraminidase from Micromonospora viridifaciens contains three domains, of which the catalytic domain is the N-terminal domain (Gaskell et al., 1995). Similarly, leech neuraminidase is a multidomain protein, where the catalytic domain is the C-terminal domain (Lvo et al., 1999). In several paramyxoviruses, neuraminidase forms part of the multi-functional haemagglutinin-neuraminidase glycoprotein found on the viral envelope (Varghese et al., 2004). Recombinant -(23,6)-Neuraminidase from Clostridium perfringens is a highly purified enzyme, which hydrolyses non-reducing, terminal -(23) and -(26) sialic acids from complex glycans and glycoproteins (Fig.3). The relative rate of cleavage of (23) linkages is reported to be a greater with this enzyme than that for -(26) linkages (Corfield et al., 1983; Roggentin et al., 1988). Use of -(2-3,6,8,9)- 31 Neuraminidase for (2-8) or branched sialic acids is suitable. To cleave only nonreducing terminal -(2-3) unbranched sialic acid residues, use of -(2-3) Neuraminidase is important. This enzyme has molecular weight of  41 kDa, pH range of activity is 4.5 – 7.0, where its optimal pH is 6.0 (Corfield et al., 1983; Dwek et al., 1993; Kobatta et al., 1979). Neuraminidase is a comparatively stable enzyme in solution. When dissolved at 120 mg/mL in 10 mM phosphate buffer, pH 6, containing 25 to 150 mM KCl, the enzyme not only retained activity at 0-4°C for over 30 months but, actually increased in activity in storage. However, for an enzyme concentration of 2 mg/mL under the same conditions, activity is reported to be lost unless albumin (BSA) was added at 0.03% (Cassidy et al., 1965). A sample at 0.01 unit/mL in 0.2 M phosphate buffer (pH 6.0) held at 57-58°C steadily lost activity, approximately 70% over three hours (Hatton et al., 1973). 32 1.6 CANCER IMMUNOTHERAPY: Depending on the type and stage of tumors, current cancer therapies involves treatment with a combination of surgery, radiation and/or chemotherapy. Whereas the primary tumor can, in most cases, be efficiently treated by a combination of these standard therapies, preventing the metastatic spread of the disease through disseminated tumor cells is often not effective (Schuster et al., 2006) . Cancer immunology research aims at development of methods to harness and enhance the body's natural ability to defend it against malignant tumors. Persistent and long-drawn efforts of numerous scientists have finally resulted into a breakthrough in immunotherapy for cancer, called biological response modifiers or biological therapies. Active cancer immunotherapy (against passive immunotherapy discussed later) aims at induction of an endogenous, long-lasting tumor antigen-specific immune (preventive or therapeutic) response. For this, specific vaccination approach defined TAAs or material obtained after biopsy of the tumor is used. The immune response against the tumor can be further enhanced by unspecific stimulation of the immune system using stimulators such as adjuvant or so-called biological response modifiers e.g., cytokines (Schuster et al., 2006). Interferons belong to a group of proteins known as cytokines. They are produced naturally by white blood cells in the body (or in the laboratory) in response to infection, inflammation, or stimulation. They have been used as a treatment for certain viral diseases, including hepatitis B. Interferon-alpha was one of the first cytokines to show an antitumor effect, and it is able to slow tumor growth directly, as well as help to activate the immune system. Interferon-alpha has been approved by the FDA (food and drug 33 administration) and is now commonly used for the treatment of a number of cancers, including multiple myeloma, chronic myelogenous leukemia, hairy cell leukemia, and malignant melanoma. Interferon-beta and interferon-gamma are other types of interferon that have been investigated (Park et al., 2001). Other cytokines with antitumor activity include the interleukins (e.g., IL-2) and tumor necrosis factor. IL-2 is frequently used to treat kidney cancer and melanoma. Some of the problems with these cytokines, including many of the interferons and interleukins, are their side-effects, which include malaise and flu-like syndromes. When given at a high dose, the side-effects can be greatly magnified. IL-2 therapy has a clinical efficacy in patients with advanced melanoma and renal cell carcinoma. IL-2 has played an important role in treating such patients, either alone or in combination with other chemo/immunotherapeutic agents and treatment strategies. IL-2 is the only FDAapproved immunotherapeutic agent used for patients with metastatic melanoma, associated with a low, but consistent, rate of overall response of 13 – 17% (7 – 9%) partial response and 6 – 8% complete response (Riker et al., 2007). Another important biological therapy involves antibodies against cancer cells or cancer-associated targets. Monoclonal antibodies are artificial antibodies against a particular target (the ‘antigen’) and are produced in the laboratory. The original method involved hybridoma cells (a fusion of two different types of cells) that acted as factories of antibody production. A major advance in this field was the ability to convert these antibodies, which originally were made from mouse hybridomas, to ‘humanized’ antibodies that more closely resemble human natural antibodies. Even newer techniques can be used to generate human antibodies from genetically engineered mice or bacteria 34 containing human antibody genes. Monoclonal antibodies (mAb) have been widely used in scientific studies of cancer, as well as in cancer diagnosis. A promising approach to improve the potency of antibody based cancer therapies is to conjugate the mAb with cytotoxic chemotherapeutic drugs. It is expected that antibody-drug-conjugates (ADCs) will eventually fulfill the promise of specific delivery of cytotoxic drugs to tumor cells (targeted therapy), thus avoiding the dose-limiting toxicity of chemotherapy that occurs as a result of its effects on normal cells (Ducry et al., 2010). The mAb could be a vehicle to bring immunoconjugate therapy to the clinic by conjugating monoclonal antibodies to drugs, toxins, and radioisotopes using the specificity of the monoclonal antibody to carry enhanced killing capacity directly to the tumor cells (Oldham et al., 2008). Vaccines have revolutionized public health by preventing the development of many important infectious diseases, including polio, small pox, and diphtheria. But, it has been much more difficult to develop effective vaccines to prevent cancer, or to treat patients who already have cancer. Attempts to develop such cancer vaccines, despite many decades of experimental work, have yet to yield proven results. In spite of this, a notable increase in interest has been generated by recent advances in the areas of immunology and cancer biology, which have led to more sophisticated and promising vaccine strategies than those previously available. Cancer vaccines typically consist of a source of cancer-associated material (antigen), along with other components, to further stimulate the immune response against the antigen. The challenge has been to find better antigens, as well as to package the antigen in such a way as to enhance the patient's immune system to fight cancer cells that have the antigen (Park et al., 2001). 35 Increasingly, cancer vaccines have been shown to be capable of improving the immune response against particular antigens. The result of this immunologic effect is not always sufficient to reverse the progression of cancer. However, cancer vaccines have been generally well tolerated, and they may provide useful anticancer effects in some situations. For example, in malignant lymphoma, a number of laboratory studies have indicated that vaccination using lymphoma-associated proteins called idiotype can stimulate the immune systems of mice sufficiently to help them resist the development of lymphomas. Cancer vaccines continue to be evaluated in these diseases as well as most other cancer types. The many new strategies for vaccine construction and immune stimulation may lead to the emergence of clinically useful cancer vaccines. An example of one exciting new approach being tested in melanoma and other cancers is the use of dendritic cell vaccines. Dendritic cells help to turn on the immune response (Park et al., 2001). 1.6.1 TAA based cancer immunotherapy Specific active immunotherapy with antigenic tumor cells which have been rendered non-tumorigenic or extracted TAA is one of the approaches in immunotherapy. The objective is to stimulate the host immune response against the tumor (Ansel and Blangy, 1984; Marx, 1989). However, one of the main problems associated with specific immunotherapy is generally the weak immunogenicity of TAA (Law, 1984). Thus, a major challenge in tumor immunology is to develop methods that augment an effective immune response to TAA (Longkumer, 2001). Many methods are known to enhance the immunogenicity of TAA such as, by enzymatic unmasking of TAA with neuraminidase (Simmons et al., 1971), by rigidification of cell membranes with lipids (Skornick et al., 36 1984) or by additional immunogenic cell surface determinants (Heicappell et al., 1986). Immunity against these more immunogenic tumor cells usually cross-reacts with the original tumor cells. This approach has good prospects. A second approach involves the reduction of suppressor T cells; this kind of treatment is based on observations that tumor-bearing animals may have T cells that can specifically suppress the immunological response to TAA (North et al., 1987). To achieve this goal most of the attention is focused upon cyclophosphamide (Longkumer, 2001). Another approach is to employ immune-stimulants as immunological adjuvants. In the past, Bacillus Calmette-Guein (BCG) and Corynebacterium parvam mixed with irradiated tumor cells have especially been studied to achieve improved immunological responses (Hoover and Hanna, 1989). Better defined and probably less toxic immune-stimulants, such as cytokines and the analogs of lipopolysaccharides (LPS) have been studied to enhance the immunological response of TAA (Livingstone et al., 1985; Maas et al., 1989; McCune and Marquis, 1990). There is also an urgent need for the development of a carrier system and strong adjuvants suitable for a TAA-based cancer immunotherapy. Liposomes and their further development as virosomes with added adjuvancy may address both these issues (Adamina et al., 2006). Instead of whole tumor cells, preparations containing extracted TAA of tumor cells (TAA extract) have recently been explored for active immunization (Fidler 1994; Kleinerman et al., 1993). Immunization with purified TAA may be more effective than whole tumor cells. In the whole tumor cells, the relevant or desired TAA may be present at a low density and consequently swamped by other determinants. Isolation and immunochemical characterization of certain TAA enable the study of immune reactions 37 against highly purified TAA with encouraging results in cancer patients (Presant et al., 1993; Hellström and Hellström, 1989; Longkumer, 2001). Many human tumor- associated antigens (TAAs) have been identified and molecularly characterized. When bound to major histocompatibility complex molecules, TAA peptides are recognized by T cells, which have a therapeutic potential of active immunization or vaccination with TAA peptides in patients with metastatic cancer. A few TAA peptides, mostly those recognized by CD8 (+) T cells in melanoma patients, have been clinically tested (Parmiani et al., 2002). In the present investigation, an earlier identified TAA of approximately 58 kDa glycoprotein (Alam et al., 2005) that was found over-expressed in liver cells of mice upon exposure to a potential hepatocarcinogen N-nitrosodibutylamine (DBN) was isolated, purified and characterized for its immunological reactivity in its native form and after unmasking of the carbohydrate content linked to the polypeptide backbone as well. The following objectives were conceived for the above investigation embedded in this thesis. 1. Induction of hepatocellular carcinogenesis in Swiss albino mice by intravenous administration of DBN. 2. Isolation and purification of membrane surface glycoprotein (s) from liver tissue of DBN-treated mice. 3. Immunochemical characterization of both purified and deglycosylated TAA.