**4. Histone modification**

Nucleosomes are considered to be the basic constituents of chromatin. Each nucleosome is an octamer of histones, which consist of two copies each of histone proteins H2A, H2B, H3, and H4 [46]. The most interesting epigenetic events in PDAC are histone modifications, since several studies revealed that the most frequently mutated epigenetic genes occurred in the histone family [13]. The posttranslational modifications include methylation, acetylation, citrullination, phosphorylation, SUMOylation, and ADP ribosylation. However, the most studied histone modifications in cancer are lysine alterations, including lysine methylation, acetylation, and phosphorylation [47–49]. In normal cell development, histone modifications regulate critical cell processes such as DNA replication and transcription or repair [46], while in cancer, histone modifications contribute to the maintenance of malignant phenotypes. In PDAC, the most common modification includes methylation and acetylation of lysine residues within the N terminal tails of histone proteins [11].

In the context of epithelial-mesenchymal transition (EMT) in PDAC, SNAIL is a critical transcription repressor of E-cadherin in EMT process. It plays a significant role in embryonic development and tumorigenesis [50]. Moreover, SNAIL has an essential function in histone modifications. This includes the activation of a set of chromatin modifiers such as lysine-specific demethylase, euchromatic histone lysine methyltransferase 2 (G 9a), suppressor of variegation 3–9 homolog 1 histone methyltransferases (Suv39H1), SIN3 transcription regulator family member A (SIN3A), and histone deacetylases (HDAC1 and HDAC2) [51, 52].

#### **4.1 Histone methylation**

 Methylation of histones is coordinated by histone methyltransferases (HMTs) and histone demethylases (HDMs). There are at least 17 different HMTs, all of which share the conserved (Su (var) 3–9, enhancer-of-zeste, trithorax) motif. The lysine methylation residue is most common and is mediated by histone lysine methyltransferases (HKMTs) [53]. Particularly, methylation at H3K9, H3K27, and H3K20 is associated with transcriptional repression, while methylation of H3K4, H3K36, and H3K79 causes transcriptional activation [47]. The silencing of tumor suppressor genes in cancer is caused by the corresponding activities of the HMT and HDMT

#### *Epigenetics: Dissecting Gene Expression Alteration in PDAC DOI: http://dx.doi.org/10.5772/intechopen.80585*

enzymes. On the other hand, the H3K27me3-specific HMT EZH2 (enhancer of zeste homolog 2), the catalytic subunit of PRC2, is overexpressed in a broad range of solid tumors, including prostate, lung, breast, colon, skin, and pancreatic cancers [54, 55].

The most frequently altered histone methylated genes in PDAC are KDM6A and MLL2 [33]. KDM6A is an H3K27me3 demethylase, which has a role in endoderm differentiation by regulating the expression of WNT signaling and HOX genes [56]. Other studies found that the loss of trimethylation at K27 of histone H3, which causes nuclear accumulation of EZH2, is strongly correlated with a poor PDAC outcome [57]. Various interactions have been shown to occur between DNA methylation and histone methylation. For example, the interaction between EZH2 and DNMTs renders the EZH2 gene a potential therapeutic target. Mucins (MUCs) are also known to play essential roles in tumor growth and invasion in pancreatic neoplasms. MUC1 and MUC4 are high-molecular-weight transmembrane mucins. Overexpression of mucins in cancer is associated with poor prognosis. It has been shown that mucin expression changes in PDAC are due to DNA methylation of H3 at the lysine9 residue [58, 59].

### **4.2 Histone acetylation**

Histone acetylation is the first discovered histone modification. The acetylation of lysine residues neutralizes their positive charge, which induces chromatin relaxation and activates a set of genes associated with transcription. On the other hand, removal of the acetyl groups is associated with gene silencing. Histone acetylases (HATs) and deacetylases (HDAC) are the required enzymes for this process [60, 61] (see **Figure 3a**).

#### **Figure 3.**

*Schematic diagram on the role of HDACs in PDAC. (A) HDACs mediate E-cadherin translational repression by activating the binding of EMT transcription factors to the E-boxes present in the E-cadherin promoter. (B) SIRT6 mediates the deacetylation of p53, FOXO3A, and C-Myc, which leads to increased metastasis and drug resistance.* 

Recent studies show a series of significant alterations of the acetylation process in PDAC, as well as mutations in the histone acetylase EP300 [33]. Furthermore, the SIRT6 gene is associated with the deacetylation of histone H3 at lysine residues 9 and 56, thus increasing the expression of the SIRT6 gene associated with PDAC metastasis by deacetylation of p53 and FOXOA3 [62](see **Figure 3b**). For instance, the activation of KRAS and increased expression of the c-Myc transcription factor promote PDAC metastasis [13]. Also, expression of HDAC7 and HDAC2 has been found increased in PDAC [63]. In addition, HDACs/HATs play important roles in the activation of several tumor suppressor genes in PDAC, such as p53 and EP300 [11, 14, 53, 64].

A recent study identified the acetylation of glutamate oxaloacetate transaminases 2 (GOT2) at three lysine residues (K159, K185, and K404) in PDAC. This promotes the transfer of NADH from the cytoplasm into mitochondria, enhancing PDAC cell proliferation and tumor growth in vivo. On the other hand, the acetylation of GOT2 at only K159 is correlated with downregulation of SIRT3 expression [65].

#### **4.3 Histone phosphorylation**

Histone phosphorylation has been associated with different cell processes, including apoptosis, cell cycle, DNA transcription, DNA repair, chromosome condensation, gene regulation, cell signaling pathways, energy, and metabolic pathways [66]. Phosphorylation of histones occurs on serine, threonine, and tyrosine residues, a process mediated by different kinases and phosphatases [46]. In cell development, the most important site for histone phosphorylation is the serine 10 of histone H3 (H3S10P), which is mediated by the Aurora-B kinase. This modification is a critical event in cell mitosis and meiosis [67].

Several studies identified histone phosphorylation changes during DNA damage, such as the phosphorylation of serine 139 on the histone H2A(X). On the other hand, phosphorylation of serines, e.g., 10 and 28 on H3, and serine 32 on H2B have been contributed by the activation of the epidermal growth factor (EGF). Moreover, H3ser28p mediated the expression of c-fos and α-globin [68–70].

 It has been shown that H2A T120 is phosphorylated in PDAC by VRK1 on the promoter region of CCND1, which consequently activates the transcription of cyclin D1 [71]. Besides, KRAS is most well-studied and known activated oncogene in PDAC [72]. Other studies have implicated the activation of the Ras-MAPK pathway with the upregulation of phospho-ERK1/2 and their downstream levels of H3 S10ph [73].

### **5. MicroRNA**

 MicroRNAs (miRNAs) are small (20–23 nucleotides), endogenous, noncoding, single-stranded RNA molecules, which control the expression of around 60% of the protein-coding genes [74]. Moreover, they can control both physiological and pathological processes, such as development and cancer [75]. In addition, the miRNA machinery is of great importance for drug development, since a functional miRNA machinery is a compulsory prerequisite for any RNA interference (RNAi)-based therapy approach. A total of 700 miRNAs have been discovered in human diseases, and more than 1000 predicted miRNA genes are yet to be experimentally validated [76].

Mature microRNAs require several steps of preprocessing before they can become functional. After they are transcribed by RNA polymerase II/III from intragenic regions or from regions that code for introns, the primary transcript (pri-miRNA) is processed by the ribonuclease Drosha and DGCR8 in the nucleus. The process

#### *Epigenetics: Dissecting Gene Expression Alteration in PDAC DOI: http://dx.doi.org/10.5772/intechopen.80585*

produces pre-microRNAs, hairpin-shaped intermediates of 70–100 nucleotides. Exportin-5, a Ran-GTP-dependent dsRNA-binding protein, transports pre-microR-NAs into the cytoplasm where they are further processed by the ribonuclease Dicer and TRBP (Tar RNA-binding protein) into a double-stranded miRNA. The strands separate and a mature single-stranded molecule join an RNA-induced silencing complex (RISC). The double miRNA strands are required to interact with RISC complex or to be degraded. Ordinarily, one miRNA strand can give rise to two individual mature miRNA sequences with different targets due to complementary seed sequence [74, 77, 78]. The single-stranded mature microRNA remains stable on the miRISC and induces posttranscriptional silencing of one or more target genes, usually through imperfect pairing with a target sequence in the 3' UTR [74]. However, this is not the only binding region for miRNAs, as there are also binding sites located in 5' UTR or even within the coding DNA sequence of mRNAs [77]. The seed sequence or seed region is a conserved heptametrical sequence, which is mostly situated at positions 2–7 from the miRNA 5' end [79, 80]. Furthermore, degradation of mature miRNAs appears to depend on their activity; in the absence of complementary targets, the miRNA could be released from miRNA-RISC complex, and then its 5′ end becomes accessible to the 5′ → 3′ exonuclease XRN2, which degrades the miRNA [81].

Cancer represents a heterogeneous group of diseases characterized by uncontrolled growth of cells, high proliferation rates, and apoptosis resistance. All of these features result from a complex of structural and expression abnormalities of genes, including those encoding microRNAs [75, 82]. The classification of cancer is more accurately defined with microRNA profiling than with mRNA profiling because of the strong correlation between microRNA expression signatures and tumor origin [75]. In general, microRNAs have two main functions in cancer; they can act as tumor suppressors (TSmiRs) or oncogenes (OncomiRs) [75, 76, 82, 83].

 One of the first indications that miRNAs serve as tumor suppressors (TSmiRs) came from Calin and colleagues when they discovered that miR-15a and miR-16-1 were deleted or downregulated in about 68% of chronic lymphocytic leukemia (CLL) samples. MiR-15a and miR-16-1 have been shown to control the expression of VEGF, a key proangiogenic factor involved in tumor angiogenesis. Furthermore, both of them induce the apoptosis of leukemic cells by affecting the antiapoptotic protein BCL2 [75, 84]. Another prominent TSmiR is the let-7 family, located at a chromosomal region, which is usually deleted in human cancers. It has been reported as a TSmiR in lung, breast [84], urothelial, and cervical cancers [85]. Recent studies found that let-7 was able to regulate the RAS oncogene in lung cancer. In addition, let-7 regulates late embryonic development by suppressing a number of genes such as c-Myc, RAS, and HAMGA2 [76, 82]. Taken together, reduced expression of TSmiRs in cancer releases oncogenic genes and promotes tumor initiation and progression.

In contrast, oncogenic miRNAs (OncomiRs) promote tumorigenesis by inhibiting tumor suppressor genes that play roles within other functions, such as cell differentiation and apoptosis. The first OncomiR that was discovered is the miR-17-92 cluster, which encodes miR-17-5p, miR-17-3p, miR-18a, miR-19a, miR-20a, miR19b-1, and miR-92-1. This cluster is located on chromosome 13 and is commonly found to be amplified in human B-cell lymphomas, lung cancer, and anaplastic thyroid cancer cells [86]. Another oncogenic miRNA, miR-21, has been validated in nine solid tumor types (lung, breast, head and neck, prostate, colon, pancreas, esophagus, stomach, and brain). Experimental data confirmed that miR-21 plays a significant role in cancer cell proliferation, apoptosis, and invasion. Accordingly, inhibition of miR-21 induces cell cycle arrest, increased apoptosis, and increased chemosensitivity to anticancer agents [87].

 The important emerging role of miRNAs in many cancer types, together with the fact that they can function as TSmiRs or OncomiRs, supports the potential of miRNAs as a new class of targets in the development of cancer therapies. Several studies have focused on targeting miRNAs as an experimental therapy in vitro or in vivo [85]. Notably, to modulate cancer-associated miRNAs in vivo, two main approaches were established: first, miRNA replacement therapy, which is based on adding the miRNAs missing in cancer cells for restoring their normal functions; second, inhibition of oncogenic miRNAs by using single-stranded chemically modified anti-miR oligonucleotides [85, 88]. The first successful in vivo experiment using anti-miRs in conjunction with locked nucleic acids was successfully applied in African green monkeys with hypercholesterolemia. The experiments resulted in the successful control of triglyceride and cholesterol levels, together with the management of disease manifestations with minimal side effects to herald a new research approach that is equally applicable in cancer [89].

PDAC shares many features with other solid tumors. Numerous studies have reported the significant roles, which miRNAs play in PDAC progression. Furthermore, these studies have also provided important information about cellular features, such as growth, invasive, and metastatic behavior that have been modified or altered in PDAC as a result of miRNAs, thus highlighting, to a large extent, the significance of miRNAs in PDAC progression [90]. High-throughput microarray technologies have been used to extensively profile miRNA signatures in cell lines, normal frozen tissues, formalin-fixed paraffin-embedded tissues (FFPE), blood, and fine needle aspiration biopsy (FNAB) samples, in order to establish a common expression pattern in PDAC [91]. Recently, a meta-analysis reviewed 11 miRNA profiling studies in PDAC and reported 439 miRNAs as deregulated in the 538 PDAC samples that were evaluated [92]. This analysis defines a common pool of


### **Table 2.**

*Top frequently deregulated miRNAs in PDAC.* 

*Epigenetics: Dissecting Gene Expression Alteration in PDAC DOI: http://dx.doi.org/10.5772/intechopen.80585* 

miRNAs that are atypically expressed in PDAC, and the potential renormalization of these miRNAs and/or expression patterns could help create a therapeutic approach in managing this aggressive disease [93].

 The commonly deregulated miRNAs are associated with major regulatory genes in several signaling pathways (**Table 2**), which are involved in most aspects of cellular physiology including regulation of cell cycle, differentiation, proliferation, and apoptosis. Notably, altered miRNA expression in PDAC contributes to metastasis and drug resistance [92, 94, 95]. The more frequently deregulated miRNAs in PDAC include miR-21, the expression of which is regulated by KRAS, and correlates with the degree of tumor progression [90]. KRAS is an important molecule in PDAC and is a direct target of miR-96, miR-217, miR-126, and miR-200c. The overexpression of these miRNAs reduces the level of KRAS expression, resulting in decreased cell invasion, migration, and tumor growth [96, 97]. Strikingly, two of these miRNAs, miR-145 and miR-200c, function as a regulatory network in the AKT-PI3K signaling pathway [98]. Conversely, it has been reported that KRAS activation suppresses the expression of the miR-134/145 cluster via the Ras responsive element-binding protein (RREB1) [99]. The miR-200 family is also frequently deregulated in PDAC and plays a significant role in EMT inhibition. One study demonstrated that miR-200 negatively regulates ZEB1 and ZEB2, which are both direct repressors of E-cadherin [100]. In the context of epigenetic modifications, several studies have found TSmiRs in PDAC, including miR-9-1, miR-124s, miR-192, miR-615-5p, and miR-1247, which were hypermethylated [44, 101–103].

### **6. Conclusions**

For the high mortality, poor prognosis, and undefined therapeutic targets in PDAC, the unraveling of the complex molecular layers driving this lethal cancer is a prerequisite for more effective therapeutic strategies and consistent diagnostic markers. The recent research on epigenetic mechanisms has significantly enriched our knowledge about the regulatory characteristics involved in the initiation, progression, and metastasis of PDAC. This book chapter has focused on the most critical epigenetics mechanisms, including DNA methylation, histone modifications, and modulated expression of miRNAs that play a significant role in PDAC tumorigenesis, and could serve as future therapeutic targets. Currently, significant emphasis is still given on detecting somatic genetic alterations in PDAC. However, it seems also promising to investigate the underlying epigenetic mechanisms for completing the full puzzle of altered gene expression in PDAC. The epigenetics field has developed strongly and will continue to advance into a frontier field for PDAC research. Additionally, it is essential to highlight the features of epigenetic mechanisms of gene regulation—their reversibility. This feature provides a ground for specifically targeting the epigenetic changes contributing to PDAC.

#### **Acknowledgements**

Alia Abukiwan was funded by the Heidrun-Seibert-Stiftung.

#### **Conflict of Interest**

The authors declare no conflict of interest.

*DNA Repair - An Update* 
