**3.3 MicroRNAs in the development of NPC**

MicroRNAs (miRNAs) are short non-coding RNA molecules of about 20-23 nucleotides in length, involved in post-transcriptional gene regulation. In animals, miRNAs control the expression of target genes by inhibiting translation or degradating target mRNAs through binding to their 3′UTR. MicroRNAs are involved in regulating a broad range of biological processes, such as development, differentiation, proliferation, apoptosis, and signal transduction pathways often deregulated in cancers. Some miRNAs can function as tumor suppressors or oncogenes (McManus 2003; Ventura and Jacks 2009).

Several biological pathways that are well characterised in cancer are significantly targeted by the downregulated miRNAs. These pathways include TGF-Wnt pathways, G1-S cell cycle progression, VEGF signalling pathways, apoptosis and survival pathways, and IP3 signalling pathways (Chen et al. 2009). Several known oncogenic miRNAs, such as miR-141 (Zhang et al. 2010) miR-17-92 cluster and miR-155 (Chen et al. 2009)were found to significantly up-regulated in NPC tumors. While some tumor suppressive miRNAs, including miR-34 family, miR-143, and miR-145, miR-218 (Alajez et al. 2011), mir-29c, miR-200a, miR-26a and let-7 (Wong et al. 2011) are significantly down-regulated in NPC. Among them, let-7 inhibits cell proliferation through down-regulation of c-Myc expression while miR-26a inhibits cell growth and tumorigenesis through repression of another oncogene: *EZH2* (Lu et al. 2011).

EBV is reported to be present in almost all NPCs and can transform cells, which subsequently induces cell proliferation and tumor growth. In addition to EBV-encoded protein-coding genes such as EBNA1 and LMP1, NPC cells and tissues also express high levels of non-coding EBV RNAs, including EBER1, EBER2 and multiple microRNAs (miRNAs). EBV was the first human virus found to encode microRNAs (Barth, Meister, and Grasser 2011). By small RNA cloning and sequencing, Zhu JY et al. characterized the miRNA expression profile of NPC tissues. Their study revealed an NPC-specific miRNA signature. EBV expresses all miRNAs from the BART cluster in NPC tissues, while no miRNA originating from the BHRF1 region of the EBV genome was found. Their study suggested that BART-derived miRNAs may have an important function in maintaining the virus in NPC tissues, whereas BHRF1 origin miRNAs might not be required for NPC

methylation status of the regulatory sequences (LRS, LMP1 regulatory sequence) located 5' from LMP1p, as LMP1 is expressed in NPCs with unmethylated LRS but is absent from NPCs with highly methylated LRS. A good correlation exists between LRS methylation and silencing of LMP1p in EBV-carrying lymphoid cell lines and tumors as well (Li and

On the other hand, EBV regulates the expression of critical cellular genes using cellular DNA methylation machinery. LMP1 has been shown to interacting with methyltransferase and further induce the cellular gene E-cadherin (*CDH1*) promoter methylation. Increased methylation may occur through the activity of DNA methyltransferases 1, 3a, and 3b that in turn are induced through JNK/AP1 signalling by LMP1. Transfection of LMP1 into cancer cells suppressed E-cadherin expression, thereby facilitating a more invasive growth of NPC cells (Tsai et al. 2006). It will be interesting to discover novel target genes regulated by

MicroRNAs (miRNAs) are short non-coding RNA molecules of about 20-23 nucleotides in length, involved in post-transcriptional gene regulation. In animals, miRNAs control the expression of target genes by inhibiting translation or degradating target mRNAs through binding to their 3′UTR. MicroRNAs are involved in regulating a broad range of biological processes, such as development, differentiation, proliferation, apoptosis, and signal transduction pathways often deregulated in cancers. Some miRNAs can function as tumor

Several biological pathways that are well characterised in cancer are significantly targeted by the downregulated miRNAs. These pathways include TGF-Wnt pathways, G1-S cell cycle progression, VEGF signalling pathways, apoptosis and survival pathways, and IP3 signalling pathways (Chen et al. 2009). Several known oncogenic miRNAs, such as miR-141 (Zhang et al. 2010) miR-17-92 cluster and miR-155 (Chen et al. 2009)were found to significantly up-regulated in NPC tumors. While some tumor suppressive miRNAs, including miR-34 family, miR-143, and miR-145, miR-218 (Alajez et al. 2011), mir-29c, miR-200a, miR-26a and let-7 (Wong et al. 2011) are significantly down-regulated in NPC. Among them, let-7 inhibits cell proliferation through down-regulation of c-Myc expression while miR-26a inhibits cell growth and tumorigenesis through repression of another oncogene:

EBV is reported to be present in almost all NPCs and can transform cells, which subsequently induces cell proliferation and tumor growth. In addition to EBV-encoded protein-coding genes such as EBNA1 and LMP1, NPC cells and tissues also express high levels of non-coding EBV RNAs, including EBER1, EBER2 and multiple microRNAs (miRNAs). EBV was the first human virus found to encode microRNAs (Barth, Meister, and Grasser 2011). By small RNA cloning and sequencing, Zhu JY et al. characterized the miRNA expression profile of NPC tissues. Their study revealed an NPC-specific miRNA signature. EBV expresses all miRNAs from the BART cluster in NPC tissues, while no miRNA originating from the BHRF1 region of the EBV genome was found. Their study suggested that BART-derived miRNAs may have an important function in maintaining the virus in NPC tissues, whereas BHRF1 origin miRNAs might not be required for NPC

Minarovits 2003)).

epigenetic mechanism of EBV.

*EZH2* (Lu et al. 2011).

**3.3 MicroRNAs in the development of NPC** 

suppressors or oncogenes (McManus 2003; Ventura and Jacks 2009).

pathogenesis. In the same study, they also identified two novel and highly abundant EBV miRNA genes, namely, miR-BART21 and miR-BART22 (Zhu et al. 2009). A parallel study demonstrated that LMP2A is the putative target of miR-BART22 in NPC. LMP2A is a potent immunogenic viral antigen that is recognized by the cytotoxic T cells, down-modulation of LMP2A expression by miR-BART22 may permit escape of EBV-infected cells from host immune surveillance (Lung et al. 2009). Similar regulations were also addressed on LMP1: EBV-encoded BART miRNAs target the 3′ UTR of the LMP1 gene and negatively regulate LMP1 protein expression. These miRNAs also modulate LMP1-induced NF-κB signalling and alleviate the cisplatin sensitivity of LMP1-expressing NPC cells (Lo et al. 2007).

## **4. Epigenetic alternations in relation to clinical parameters of NPC, and their roles as biomarkers**

Frequent aberrantly methylated TSGs in tumors have been used as molecular markers for the detection of malignant cells from various clinical materials. It provides possibilities of both cancer early detection and dynamic monitoring of cancer patients after treatment (Schulz 2005).

DNA methylation biomarkers hold a number of advantages over other biomarker types, such as proteins, gene expression and DNA mutations (Balch et al. 2009; Laird 2003). Methylated DNA sequences are more chemically and biologically stable, and more easier to be amplified, thus greatly enhancing detection sensitivity. DNA methylation are often cancer specific, and restriction to limited regions of DNA in the CpG islands. Compared to genetic alternations such as gene mutation or amplification, aberrant methylation on TSG promoters is rather prevalent and tumor-specific among NPCs. As mentioned above, NPC tumor progression is well characterized by a number of combinatorial epigenetic aberrations distinct to other malignancy, including DNA methylation of more than 30 genes. Consequently, these methylated DNA sequences represent potential biomarkers for diagnosis, staging, prognosis and monitoring of response to therapy or tumor recurrence (Balch et al. 2009; Laird 2003).

#### **4.1 DNA methylation, results from tumor tissues**

It has been shown that some genes are high frequently methylated in tumor tissue DNA obtained from NPC primary tumors, but not in normal tissues (Pan et al. 2005; Sun et al. 2007; Zhang et al. 2007; Li, Shu, et al. 2011). These genes are ideal candidate to serve as biomarkers for detection of NPC. Some of these TSGs are not only methylated in NPC, but also commonly methylated in other cancers. So methylation assessment of single genes lacks sufficient specificity for NPC diagnosis. It is believed that panels of multiple methylation biomarkers may achieve higher accuracy required for discriminate NPC from other cancers (Kwong et al. 2002; Hutajulu et al. 2011). This notion was supported by a study of Esteller et al, which showed that a panel of three to four markers could define an abnormality in 70– 90% of each cancer type through detecting their aberrant methylation (Esteller et al. 2001). Some studies have been conducted using different combination of gene panels, though there is overlap among them. Combination of methylation markers not only improved the discrimination between NPC and non-NPC diseases, but also the sensitivity of cancer

Epigenetics of Nasopharyngeal Carcinoma 17

demethylating agents might restore normal cell growth control, or induce cell immunity against cancer cells. Demethylating agents would also reactivate the expression of EBV early and lytic genes in latently infected NPC cells, which will lead to further tumor cell death.

Epigenetic therapeutic agents include DNA methyltransferase inhibitors and histone deacetylase (HDAC) inhibitors. 5-Azacytidine and 5-aza-2'-deoxycytidine are the most widely studied DNMT inhibitors. Clinical trials using such agents have been carried out on a series of cancer patients. In several phase I/II/III studies, decitabine (5-aza-2' deoxycytidine) has also shown promising data in patients with MDS and AML (Kantarjian et al. 2007; Issa et al. 2004). In patients with NPC and EBV-positive AIDS-associated Burkitt lymphoma, azacitidine effectively induces demethylation of all the latent and early lytic EBV promoters and some viral antigens, indicated the potential of epigenetic therapy for

Adjei, A. A. 2001. Blocking oncogenic Ras signaling for cancer therapy. *J Natl Cancer Inst* 93

Ayadi, W., H. Karray-Hakim, A. Khabir, L. Feki, S. Charfi, T. Boudawara, A. Ghorbel, J.

Balch, C., F. Fang, D. E. Matei, T. H. Huang, and K. P. Nephew. 2009. Minireview: epigenetic

Biel, M., V. Wascholowski, and A. Giannis. 2005. Epigenetics--an epicenter of gene

Burgos, J. S. 2003. Absence of p53 alterations in nasopharyngeal carcinoma Spanish patients

Chan, K. C., J. M. Ko, H. L. Lung, R. Sedlacek, Z. F. Zhang, D. Z. Luo, Z. B. Feng, S. Chen, H.

Chan, S. L., Y. Cui, A. van Hasselt, H. Li, G. Srivastava, H. Jin, K. M. Ng, Y. Wang, K. Y. Lee,

nasopharyngeal and esophageal carcinomas. *Lab Invest* 87 (7):644-50.

Chen, K. W. Chan, S. W. Tsao, D. T. Chua, E. R. Zabarovsky, E. J. Stanbridge, and M. L. Lung. 2010. Catalytic activity of matrix metalloproteinase-19 is essential for tumor suppressor and anti-angiogenic activities in nasopharyngeal carcinoma. *Int J* 

G. S. Tsao, S. Zhong, K. D. Robertson, S. Y. Rha, A. T. Chan, and Q. Tao. 2007. The tumor suppressor Wnt inhibitory factor 1 is frequently methylated in

Bird, A. 2002. DNA methylation patterns and epigenetic memory. *Genes Dev* 16 (1):6-21. Boguski, M. S., and F. McCormick. 1993. Proteins regulating Ras and its relatives. *Nature* 366

Bos, J. L. 1989. ras oncogenes in human cancer: a review. *Cancer Res* 49 (17):4682-9.

with Epstein-Barr virus infection. *Virus Genes* 27 (3):263-8.

regulation: histones and histone-modifying enzymes. *Angew Chem Int Ed Engl* 44

Barth, S., G. Meister, and F. A. Grasser. 2011. EBV-encoded miRNAs. *Biochim Biophys Acta*. Berx, G., and F. van Roy. 2009. Involvement of members of the cadherin superfamily in

biopsies from Tunisian patients. *Anticancer Res* 28 (4B):2161-7.

changes in ovarian cancer. *Endocrinology* 150 (9):4003-11.

cancer. *Cold Spring Harb Perspect Biol* 1 (6):a003129.

Daoud, M. Frikha, P. Busson, and A. Hammami. 2008. Aberrant methylation of p16, DLEC1, BLU and E-cadherin gene promoters in nasopharyngeal carcinoma

NPC (Chan et al. 2004).

(14):1062-74.

(21):3186-216.

(6456):643-54.

*Cancer*.

**6. References** 

detection. The detection rate can reach 98% when combined analysis of five methylation markers (*RASSF1A, p16, WIF1, CHFR* and *RIZ1*) in a recent study (Hutajulu et al. 2011).

#### **4.2 Methylation markers in circulating DNA**

Cancer specific DNA methylation can be detected in tumor-derived free DNA in the bloodstream, e.g. in serum or plasma. High frequency of methylated *DAPK* gene were found not only in NPC tumors, but also could be detected in plasma and buffy coat of NPC patients (Wong et al. 2002). Methylated DNA was detectable in plasma of NPC patients before treatment including 46% for *CDH1*,42% for *CDH1*, 42% for *p16*, 20% for *DAPK* ,20% for *p15 and 5%* for *RASSF1A*. Aberrantly hypermethylated promoter DNA of at least one of the five genes was detectable in 71% of plasma of NPC patients before treatment. Hypermethylated promoter DNA of at least one of the three genes (*CDH1*, *DAPK1*, and p16) was detectable in post-treatment plasma of 38% recurrent NPC patients and none of the patients in remission. Suggesting that cell-free circulating methylated DNA might be a useful serological marker in assisting in screening of primary and potentially salvageable local or regional recurrent NPC (Wong et al. 2004).

#### **4.3 Methylation markers in other body fluids and nasopharyngeal swabs**

In addition to tissue analysis, methylated DNA has been detected in the mouth and throat rinsing fluid, saliva and nasopharyngeal swabs of NPC patients. Methylated DNA found in cancer patient serum correlated reasonably well with methylation levels in tumor tissue, and it is also believed that the source of serum DNA is necrotic tumor cells. Hypermethylated *RIZ1* gene was detected in 60% of NPC primary tumors, but not in any of the normal controls. Of 30 matched body fluid samples, methylated *RIZ1* DNA was found in 37% of NP swabs, 30% of rinsing fluid, 23% of plasma, and 10% of buffy coat samples. The results in NPC tumor and NP swab samples from the same patients show good concordance. Our early study also reported that the high sensitivity (81%) and specificity (0% false positives) of detecting aberrant methylation of *CDH13* (encoded a cell adhesion molecule H-cadherin) from nasopharyngeal swabs suggested it could be utilized as a tool for early diagnosis.
