**2. The Epstein Barr virus and mechanisms of infection**

Many epidemiological studies have been performed concerning NPC and three well-dened etiological factors involved in its pathogenesis have been identied. These include genetic susceptibility, early-age exposure to chemical carcinogens and an association with a latent EBV infection (Chan, Teo et al. 2002; Busson, Keryer et al. 2004).

It is widely accepted that EBV infection plays a major role in the pathogenesis of NPC in endemic areas, although not entirely clear in non-endemic regions (Henle and Henle 1976; Pathmanathan, Prasad et al. 1995). The association between NPC and EBV was initially discovered from serological studies and was later supported by demonstration of EBV DNA and nuclear antigen proteins (EBNA) in NPC tumor cells (zur Hausen, Schulte-Holthausen et al. 1970; Henle and Henle 1976). The notion that EBV plays an important role in the development of NPC is further supported by the observation that early nasopharyngeal lesions (dysplasia or carcinoma-in situ) are already EBV-positive, harbouring latent and clonal viral genomes as well as viral oncoproteins such as latent membrane protein (LMP) (zur Hausen, Schulte-Holthausen et al. 1970; Pathmanathan, Prasad et al. 1995).

In 1958, Denis Burkitt, a British surgeon working in Africa, described a common cancer primarily affecting children in specific regions of Africa (Burkitt 1958). Burkitt believed that a virus might be implicated in cancer development, given the climatic and geographic distribution of the cases (Burkitt 1962). EBV was first identified in 1964 when Anthony Epstein, Yvonne Barr and Bert Achong detected virus-like particles by electron microscopy in a cell line that had been established from a Burkitt's lymphoma biopsy (Epstein, Achong et al. 1964). The causal link between EBV and BL was corroborated by evidence showing that BL patient sera had elevated antibodies to EBV antigens (Henle and Henle 1966) . This group also established a link between primary EBV infection and infectious mononucleosis and, subsequently, the association of EBV with nasopharyngeal carcinoma (Henle, Henle et al. 1968; Henle, Henle et al. 1970; zur Hausen, Schulte-Holthausen et al. 1970).

Nasopharyngeal cancer, particularly when showing histological patterns of nonkeratinization (WHO/1978 II or III), is associated with previous EBV infection with sequential immortalization of the virus in the host's B-lymphocytes (Niedobitek, Hansmann et al. 1991).

*Bmi*-1 correlates with invasion of NPC and otherwise a possible mechanism of tumorigenesis involves repression of the *Ink4a-Arf* locus, which encodes the tumor

Furthermore, there is an ongoing debate over whether CSCs represent a mature tissue stem cell which has undergone malignant change or whether more differentiated cells re-initiate a 'stemness' programme as part of, or following, malignant transformation (Bomken, Fiser et

These and other questions are yet to be answered, namely how to explain the relationship between CSC, NPC and EBV. The recent accomplishment that LMP2A EBV encoded latent protein, induces epithelial–mesenchymal transition and stem-like cell self-renewal in NPC,

Many epidemiological studies have been performed concerning NPC and three well-dened etiological factors involved in its pathogenesis have been identied. These include genetic susceptibility, early-age exposure to chemical carcinogens and an association with a latent

It is widely accepted that EBV infection plays a major role in the pathogenesis of NPC in endemic areas, although not entirely clear in non-endemic regions (Henle and Henle 1976; Pathmanathan, Prasad et al. 1995). The association between NPC and EBV was initially discovered from serological studies and was later supported by demonstration of EBV DNA and nuclear antigen proteins (EBNA) in NPC tumor cells (zur Hausen, Schulte-Holthausen et al. 1970; Henle and Henle 1976). The notion that EBV plays an important role in the development of NPC is further supported by the observation that early nasopharyngeal lesions (dysplasia or carcinoma-in situ) are already EBV-positive, harbouring latent and clonal viral genomes as well as viral oncoproteins such as latent membrane protein (LMP)

In 1958, Denis Burkitt, a British surgeon working in Africa, described a common cancer primarily affecting children in specific regions of Africa (Burkitt 1958). Burkitt believed that a virus might be implicated in cancer development, given the climatic and geographic distribution of the cases (Burkitt 1962). EBV was first identified in 1964 when Anthony Epstein, Yvonne Barr and Bert Achong detected virus-like particles by electron microscopy in a cell line that had been established from a Burkitt's lymphoma biopsy (Epstein, Achong et al. 1964). The causal link between EBV and BL was corroborated by evidence showing that BL patient sera had elevated antibodies to EBV antigens (Henle and Henle 1966) . This group also established a link between primary EBV infection and infectious mononucleosis and, subsequently, the association of EBV with nasopharyngeal carcinoma (Henle, Henle et

Nasopharyngeal cancer, particularly when showing histological patterns of nonkeratinization (WHO/1978 II or III), is associated with previous EBV infection with sequential immortalization of the virus in the host's B-lymphocytes (Niedobitek, Hansmann

(zur Hausen, Schulte-Holthausen et al. 1970; Pathmanathan, Prasad et al. 1995).

al. 1968; Henle, Henle et al. 1970; zur Hausen, Schulte-Holthausen et al. 1970).

suppressors INK4A and ARF, by direct binding of *Bmi*-1.

is a strong evidence of this association (Kong, Hu et al. 2010).

**2. The Epstein Barr virus and mechanisms of infection** 

EBV infection (Chan, Teo et al. 2002; Busson, Keryer et al. 2004).

al. 2010).

et al. 1991).

EBV is a member of the herpesvirus family. As other herpesviruses, EBV is an enveloped virus that contains a DNA core surrounded by an icosahedral nucleocapsid and a tegument. Family members include herpes simplex 1 (HSV-1) and 2 (HSV-2) and varicella-zoster virus (VZV) - alphavirus subfamily; cytomegalovirus (CMV) and human herpesvirus 6 (HHV-6) and 7 (HHV-7) - betaherpesvirus subfamily; and human herpesvirus 8 (HHV-8) and EBV gammaherpesvirus subfamily. The oncogenic potential of EBV was recognized through the association with numerous human malignancies. In addition to endemic Burkitt's lymphoma and NPC, EBV was later found in Hodgkin's lymphoma cases, post-transplant lymphoproliferative diseases, some T-cell lymphomas and a proportion of gastric carcinomas (Young and Rickinson 2004). Research is currently ongoing to determine the role of EBV-encoded gene products in these different cellular environments in an attempt to understand the role that EBV plays in the pathogenesis of these malignancies.

It is now known that EBV infects 90% of the world's adult population. Although herpesviruses are ubiquitous in nature, humans are the only natural host for EBV. Upon infection, the individual remains a life-long carrier of the virus (Henle and Henle 1976). During acute infection, EBV seems to primarily infect and replicate in the stratified squamous epithelium of the oropharynx, followed by a latent infection of the B lymphocytes. EBV infection of B lymphocytes is thought to occur in the upper aerodigestive tract lymphoid organs and the virus persists in circulating memory B cells (Sixbey, Nedrud et al. 1984; Farrell 1995). Taking in account that EBV infection is a common event in the entire World, the reasons why only some individuals will develop EBV-related malignancies, are interesting aspects yet to clarify.

Much of the known biology of EBV relates to its interaction with B-lymphocytes. This is mainly a result of the ability of EBV to readily infect and transform normal resting Blymphocytes in vitro, which also conrms the B-lymphotropic nature of this virus. EBV latent gene expression in various EBV-associated malignancies and EBV-derived cell lines has led to the identication of three different and distinct latency patterns. These latency patterns characterize a heterogeneous group of diseases and are based on EBV genome expression arrangements as the result of differential promoter activity inuenced by host cell factors (Thompson and Kurzrock 2004; Young and Rickinson 2004). Latency type I is generally associated with Burkitt's lymphoma and is characterized by expression of the EBV-encoded RNAs (EBERs) and the BamHI-A rightward transcripts (BARTs), in addition to EBV nuclear antigen-1 (EBNA1) expression. Latency type II is associated with NPC, gastric cancer and EBV-positive Hodgkin lymphoma and is characterized by expression of EBERs, BARTs, EBNA1 and the latent membrane proteins (LMP1, LMP2A and LMP2B). Latency type III is associated with lymphoblastoid cell lines and post-transplant lymphoproliferative diseases: the full range of latent gene products are expressed, including EBNAs 1, 2, 3A, 3B, 3C and -LP, the expression of all three latent membrane proteins (LMP1, LMP2A and LMP2B) and EBERs and BART RNAs. Although these classications of latency are useful in characterizing the distinct gene expression patterns, they are by no means completely denitive (Thompson and Kurzrock 2004). In recent years, there has been increasing interest in the presence of different viral and cellular micro-RNAs in EBVinfected B cells and epithelial cells. Future works should investigate the role of EBV-encoded micro-RNAs and its correlation with transcriptional regulation of both the viral and cellular genome.

Nasopharyngeal Cancer – An Expanding Puzzle – Claiming for Answers 233

The development and progression of NPC involves the accumulation of genetic alterations. Both genetic and epigenetic changes can affect the development of NPC, by altering the

Host factors previously shown to be associated with NPC development include *HLA* class I and II alleles and polymorphisms of genes responsible for critical cellular functions, as carcinogen metabolism and detoxification, cell cycle control, DNA repair and immune response. Published studies demonstrated that some genetic variations are associated with a higher risk of NPC development (Hildesheim, Anderson et al. 1997; Nazar-Stewart, Vaughan et al. 1999; Cho, Hildesheim et al. 2003; Catarino, Breda et al. 2006; Sousa, Santos et al. 2006). This genetic predisposition to NPC may be modulated by environmental factors

Familial clustering of the disease has been widely documented in the Chinese population and several polymorphisms have been associated with NPC development, namely in nitrosamine metabolizing genes cytochrome P450 2A6 (CYP2A6), P450 2E1 (CYP2E1) and P450 2F1 (CYP2F1), Glutathione S- transferase M1 (GSTM1), XRCC1 (codons 194 arg-trp), have been suggested to influence the susceptibility to NPC (Hildesheim and Levine 1993; Pathmanathan, Prasad et al. 1995; Sousa, Santos et al. 2006; Tiwawech, Srivatanakul et al.

Less in known in non-endemic areas, but it has been suggested that the individual genetic background may influence the onset of NPC, also in these regions. Regarding cell cycle control, some studies investigated the role of polymorphisms in cyclin D1 (CCND1) and TP53 genes in the genetic susceptibility to NPC (Pathmanathan, Prasad et al. 1995; Cho, Hildesheim et al. 2003; Catarino, Pereira et al. 2008). These genetic variants have been associated with increased individual risk of NPC development. Regarding the key regulator of cell cycle, cyclin D1, the study by Catarino et al, demonstrate that individuals carrying the genetic variant present a 2-fold increased risk for the development of NPC, the proportion of cases attributable to the genotype being 14.76% (Catarino, Breda et al. 2006). Moreover, another study indicates that cyclin D1 variants influence the age of onset of oncogenic virusassociated cancers, namely NPC (Catarino, Pereira et al. 2008). The results of this study demonstrate that the waiting time for onset of oncogenic virus-associated neoplasia in patients carrying the cyclin D1 variant was 12 years earlier in comparison with the other patients. Another study by Sousa et al, focus on the polymorphic variants (Arg/Pro) on *TP53* codon 72 in nasopharyngeal cancer development. This study revealed a three-fold risk for carriers of Pro/Pro genotype, suggesting that Pro/Pro genotype represents a stable risk

Another study revealed that a MDM2 polymorphism is associated with increased risk to develop NPC adjusted for age and gender, with particular effect in undifferentiated types and early clinical stages. The study also indicates that the median age of onset of NPC was significantly different according to the genetics variants (Sousa, Pando et al. 2011). Regarding the immune response markers, a study revealed an increased frequency of a tumor necrosis factor-alpha (TNF-α) polymorphic variant in patients with NPC. The authors indicate that this variant is associated with a 2-fold increased risk for the development of this disease. Moreover, this effect seems to be stronger in undifferentiated types (Sousa,

function of genes that are critical for proliferation, apoptosis and differentiation.

and this may have a differential impact in endemic and non-endemic NPC.

factor for nasopharyngeal carcinoma (Sousa, Santos et al. 2006).

2006).

Breda et al. 2011).

Analysis of EBV in NPC tumors has revealed the presence of clonal EBV genomes, suggesting that these carcinomas arise from the clonal expansion of a single EBV-infected progenitor cell (Raab-Traub and Flynn 1986). Similar to most EBV-associated malignancies, the exact role of EBV in NPC pathogenesis remains poorly understood. Unlike EBVassociated B cell tumors, where the virus is considered to be an initiating factor in the oncogenic process, virus infection in the context of NPC pathogenesis seems to act as a tumor-promoting agent.

Recent studies investigated circulating tumor-derived EBV DNA in the blood of NPC patients. Using real-time PCR methodology, studies indicate that the level of pre-treatment EBV DNA has prognostic significance. Furthermore, post-treatment EBV DNA levels seem to be correlated with treatment response and overall survival of NPC patients (Lo, Chan et al. 2000; Lo, Leung et al. 2000; Lin, Chen et al. 2001; Chan, Lo et al. 2002). This methodology should be applied in large-scale trials as an approach for early disease diagnosis.
