Keywords

19.1 Introduction

Carcinogenesis is a multi-step process which starts with tumor initiation and then advances with tumor progression, subsequently metastasizing to the distant body parts. The role of de novo as well as inherited mutations in carcinogenesis is well established. Mutational changes drive to cellular transformations by two important mechanisms, through inactivation of tumor suppressor genes and through conversion of proto-oncogenes into oncogenes. In addition to these genomic changes, there are many different epigenetic modifications which have the capability to regulate the process of gene expression without directly altering the DNA sequences. These epigenetic alterations include methylation of CpG islands in the promoter regions of genes, modifications of the histone tails leading to changes in chromatin structure as well as microRNAs (miRNAs), the short regulatory RNAs (20–30 nucleotides in length) which undergo sequence-specific binding to the 3’ untranslated region (3’UTR) of their target genes and lead to their degradation or inhibition of translation.

The epigenetic machinery gets significantly altered in the process of breast carcinogenesis leading to drastically changed genetic expression profiles in cancerous cells. The altered expression patterns facilitate these cells to achieve selective advantage over the non-cancerous cells. Nearly all the cellular pathways for example those involved in cell growth, proliferation and survival are affected by these epigenetic alterations.

In this chapter on epigenetic factors in breast cancer progression, we are focusing on the basics of epigenetics, the major types of epigenetic modifications such as DNA methylation, histone modifications, and miRNA silencing in breast cancer progression along with the epigenetic targeting strategies against prevention and therapeutics of breast carcinogenesis.

19.2 Epigenetics of Breast Cancer

Breast cancer is the most frequently diagnosed cancer in females and the leading cause of cancer death among women globally. Although acquired genetic alterations in the important tumor suppressor genes such as BRCA1 or proto-oncogenes such as c-Myc are considered as the driving forces for breast cancer initiation, epigenetic malfunction has also been proven to contribute equally in the process of breast carcinogenesis. Epigenetics is defined as the study of heritable changes in gene expression which occur without alterations in the underlying DNA sequences. Epigenetic mechanisms constitute a regulatory layer which coordinates all the crucial biological processes and alterations in the dynamics of these critical players is associated with a variety of human cancers, including breast cancer.

DNA methylation, histone modification and miRNA silencing are the three major epigenetic players of transcriptional regulation of gene expression. The first key player in epigenetic regulation machinery, DNA methylation, is the process of covalent addition of a methyl group to the fifth carbon of the cytosine nucleotide in the CpG dinucleotide sequences. The methylation is catalyzed by a group of enzymes known as the DNA methyltransferases (DNMTs) and the methyl group is donated by S-adenosyl methionine (SAM). The process of DNA methylation is involved in many important biological phenomenon such as genomic imprinting, X-chromosome inactivation as well as developmental processes. The genes with lower promoter methylation status are in general transcriptionally active, for example the housekeeping genes have lower level of promoter methylation, while those with higher promoter methylation are in general transcriptionally repressed as illustrated in Fig. 19.1a. In breast cancer, the aberrant methylation of numerous genes results in altered level of their expression. The retinoblastoma (Rb) gene was the first gene reported to be hypermethylated in cancer [1].

Fig. 19.1
figure 1

Role of DNA methylation and Histone acetylation in transcriptional regulation. Schematic diagram illustrating the transcriptional changes associated with DNA and histone modifications. (a) The DNA methyltransferases add methyl groups to the cytosine residues of the CpG dinucleotides in the gene promoters which render the MSDBPs unable to bind to the promoter and transcriptional inactivation of genes takes place. The demethylating agents remove the methyl groups from the cytosine residues leading to binding of transcriptional complexes (TC) and the gene becomes transcriptionally active. (b) The histone acetyltransferases transfer the acetyl moieties to the histone tail and lead to unwinding of the DNA resulting in the transcriptional activation of the genes. The histone deacetylases counterbalance the process by removing the acetyl groups and lead to conversion of chromatin from an active (euchromatin) to inactive form (heterochromatin). Abbreviations: Me methyl groups; Ac acetyl groups; MSDBPs methylation sensitive DNA binding proteins; TC transcriptional complex

Full size image

Histone tail modifications are the second most important players in transcriptional regulation. Histones are the basic proteins which constitute the nucleosome core around which DNA is tightly packaged to form chromatin. The amino-terminal tails of these histones protrude out from the nucleosomal core and are the subject of numerous bio-chemical modifications. Lysine residues are the most often modified amino acid residues with a variety of modifications including acetylation, methylation, ubiquitination and sumoylation. Other histone modifications include methylation and deimination of arginine, phosphorylation of serine or threonine, proline isomerization and ADP-ribosylation of glutamic acid residues. The modes of alteration in transcriptional regulation are specific to the type and position of the modified amino acid residues.

The various types of histone modifications are regulated by many different histone modifying enzymes including histone acetyltransferases (HATs) and histone deacetylases (HDACs), involved in lysine acetylation and deacetylation, respectively; histone lysine methyltransferases (KMTs) and histone lysine demethylases (KDMs) implicated in lysine methylation and demethylation, respectively; E3 ubiquitin ligase, ubiquitin protease and nuclear deubiquitinase involved in ubiquitination; E1 activating and E2 conjugating enzymes involved in histone lysine sumoylation; histone arginine methyltransferase and peptidyl arginine deiminase 4 involved in arginine methylation and deimination; different kinases and phosphatases which contribute to serine or threonine phosphorylation and dephosphorylations; proline peptidyl isomerase, an important enzyme for proline isomerization and poly (ADP-ribose) polymerase leading to ADP-ribosylation of glutamic acid. These enzymes write the histone code of cells predicting the expression profiles of genes. These modifications lead to either eu- or hetero-chromatin conformations and facilitate the accessibility or inaccessibilty of DNA to the transcription complexes. Fig. 19.1b shows the role of histone acetylation in transcriptional regulation.

Micro-RNAs (miRNAs) are small, non-coding, phylogenetically conserved RNAs which are regulatory in function. The third most important key players in post transcriptional regulation, miRNAs bind to 3’-untranslated region (3’UTR) of their target mRNA transcripts by perfect complementarity and lead to their degradation. They can function as tumor suppressors when they target proto-oncogenic transcripts or as oncogenes, targeting tumor suppressor gene transcripts. These regulatory RNAs are also involved in promoting epithelial to mesenchymal transition (EMT) and malignancy. Global down-regulation of miRNA expression is a very frequent event in breast carcinogenesis. Figure 19.2 illustrates miRNA mediated silencing in breast carcinogenesis.

Fig. 19.2
figure 2

miRNA-mediated transcriptional silencing. The process of gene transcription involves the synthesis of primary RNA transcripts which undergo post-transcriptional mRNA processing involving splicing, capping and polyadenylation. The mature mRNAs then enter the cytoplasm. The miRNA are synthesized as the primary miRNA transcripts which are cleaved by the nuclear enzyme drosha to form precursor miRNAs. These precursor miRNAs containing hair-pin loop then enter the cytoplasm, their entry being facilitated by a nuclear membrane protein exportin. In the cytoplasm, dicer cleaves the hair-pin loop leading to the formation of mature miRNA, which binds to the mRNAs containing region of sequence complementarity. The RNA-induced silencing complex (RISC) cleaves the double stranded mRNA-miRNA complex leading to post-transcriptional repression

Full size image

19.2.1 DNA Methylation in Breast Cancer

Approximately, more than half of the genes in the human genome have CpG islands in their promoters. Both normal cellular development and disease processes are regulated by patterns of methylation of CpG islands. Promoter methylation has dual roles in terms of transcriptional inactivation and chromatin remodelling. The methylation profiles of the breast cancer tissue and corresponding normal tissues are quite different. The local hypermethylation of CpG islands in the tumor suppressor gene promoters and global hypomethylation of the genome is a commonly observed paradox in case of all cancer types including breast cancer. The DNA methyltransferases (DNMTs) are the group of enzymes actively involved in cytosine methylation in the CpG islands. These enzymes are of two types depending on their target DNA. The DNMT1 is the maintenance methyltransferase which methylates the daughter DNA strands based on the pattern of methylation in parental strands. The other two DNA methyltransferases, DNMT3A and DNMT3B are de novo DNA methyltransferases which by themselves set up the patterns of methylation early in the development.

19.2.1.1 Hypermethylation in Breast Cancer

The active promoters are less methylated at their CpG islands and are considered open to the transcription factors and other transcriptional co-activators. The breast cancer cells achieve substantially altered methylation profiles in the process of carcinogenesis. Analysis of DNA methylation has been advanced to an enormous degree by genome-wide high throughput sequencing approaches, which allow quantitative and cost-effective analysis of the target genome for methylation studies. A genome-wide high-throughput methylation enriched immunoprecipitated-DNA (MeDIP-DNA) sequencing based study confirmed a global promoter hypomethylation at the genomic level and a cell-type specific regulation of gene methylation at the gene level in eight different breast cancer cell lines [2]. The major genes which are reported to be hypermethylated in breast carcinogenesis are regulatory genes coding for hormonal receptors (ERα, ERβ, PR), genes regulating cell cycle progression (p16 INK4A , CCND2, p57 KIP2), growth inhibitory genes (TGFβ, RASSF1A, SOCS1, RARβ, SYK, HIN-1, NES1), genes involved in DNA repair mechanisms (BRCA1, MGMT, GSTP1 etc.), pro-apoptotic genes (DAPK, FHIT, Twist, HOXA5, TMS1, GPC3), genes regulating angiogenesis (MASPIN, THBS1), genes involved in cell to cell and cell to matrix interactions and cytoskeletal changes (E-Cadherin, CDH13, APC, prostasin, TIMP-3, BCSG1) [3, 4].

p16INK4A hypermethylation has been clearly associated with the immortalization potential of the normal human mammary epithelial cells (HMECs). The HMEC cells having hypermethylation at p16INK4A promoter have demonstrated to bypass senescence barrier known as stasis which is mediated by Rb and another p53 involving telomere-length-dependent pathway. When these cells overcome another barrier known as agonescence or crisis, they undergo immortalization. The hypermethylation profiles of these immortalized HMECs for a cluster of 30 genes known as ‘cancer proliferation cluster’ were found to be remarkably similar to the methylation profiles of the non-invasive breast cancers [5]. The methylation status of p16INK4A was also found to be significantly associated with frequencies of promoter methylation of BRCA1, BRCA2, ERα, and RARβ2. Thus, the hypermethylation of p16INK4A gene is indicative of global pattern of epigenetic modifications at the time of early breast carcinogenesis [6]. The methylation profiling of women at high breast cancer risk without having BRCA1 and BRCA2 mutations have demonstrated higher frequencies of promoter methylation of the several tumor suppressor genes such as RARβ, ER, p16INK4A, BRCA1, PR-A, PR-B, RASSF1A, HIN-1 and CRBP1 in periareolar fine-needle aspirate samples [7].

Hypermethylation of promoters in RARβ2, APC and RASSF1A have been considered as early events in breast carcinogenesis. More specifically, RASSF1A is the marker for breast cancer as well as benign breast susceptibility, while RARβ2 methylation is indicative of personal history of breast cancer. APC promoter hypermethylation correlated inversely with the parity of women. Parity has been associated with the decreased risk of breast cancer [8, 9].

The incidence of hypermethylation at certain key breast cancer regulatory gene promoters is an early biomarker to detect breast carcinogenesis. This hypermethylation can also be used as a marker of pre-invasive and invasive breast cancer [4, 10]. In contrast, a genome-wide analysis showed that the presence of a breast CpG island methylator phenotype (B-CIMP), having higher proportion of genes hypermethylated, can predict the possibility of metastasis or survival, where higher methylation was associated with lower risk of breast cancer metastasis [11]. Quantitative multiplex methylation specific PCR (QM-MSP) is a high-throughput, very sensitive DNA methylation detection method which is used to quantitate cumulative gene promoter hypermethylation in samples with very limited content of DNA.

Promoter methylation of genes encoding for the regulatory miRNAs is another mechanism involved in deregulation of cellular activities observed in case of breast carcinogenesis. The level of miRNA expression is the measure of aggressiveness of the breast tumor. Hypermethylation of genes for tumor suppressor microRNAs such as let-7 family, miR-206, miR-17-5p, miR-125a, miR-125b, miR-200, miR-34, and miR-31, have been shown to be a common event in breast tumors [12, 13]. Another human micro RNA, miR-335 involved in tumor recurrence in breast cancer, was reported to be inactivated through dual mechanisms, deletion and promoter hypermethylation [14].

19.2.1.2 Hypomethylation in Breast Cancer

In normal cells, hypermethylation of repetitive DNA sequence elements is a universal regulatory mechanism leading to silencing of these elements to circumvent the chromosomal rearrangements such as translocations and gene disruptions through insertions of reactivated transposable elements [15, 16]. Cancerous cells have globally hypomethylated genome which leads to chromosomal aberrations, loss of genomic imprinting, activation of repetitive elements and chromosomal instability due to increased mutational events [17]. Promoter hypomethylation of numerous proto-oncogenes leading to uncontrolled proliferation and metastasis (genes for synuclein γ and urokinase) and development of drug resistance (genes encoding N-cadherin, ID4, β-catenin, annexin A4 and WNT11) is the other major epigenetic mechanism of breast carcinogenesis [17]. Even the early cancerous lesions display widespread global hypomethylation patterns and patterns of hypomethylation of CpG sequences in satellite DNA, suggesting it to be an early event in breast carcinogenesis [18]. Analysis of three repetitive DNA elements, long interspersed repetitive DNA elements (LINE1), short interspersed repetitive DNA elements (Alu) and satellite DNA elements (Sat2) lead to the conclusion that their methylation level is significantly lower to those of adjacent normal breast tissue [19]. BRCA1 mutations have been proved to be associated with aberrant regulation of DNMTs leading to global hypomethylation and increased expression of various proto-oncogenes such as c-Fos, Ha-Ras and c-Myc [20].

19.2.1.3 Methylation-Sensitive Transcription Factors in Breast Cancer

The mechanism of transcriptional repression by hypermethylation of CpG islands in gene promoters may function in two different ways. The first mechanism may be by creating hindrance in binding of methylation-sensitive transcription factors to the gene promoters leading to transcriptional repression [21]. The other mechanism may be by recruiting the methylation-dependent transcriptional repressors to the hypermethylated CpG islands, which subsequently recruit the HDACs. The HDACs in turn deacetylate the histone proteins leading to conversion of euchromatin into heterochromatin. The members of a family of nuclear proteins, methyl-CpG-binding domain proteins (MBDs) which comprises of MBD1, MBD2, MBD3, MBD4 and MeCP2 are capable of binding specifically to methylated DNA (with the exception of MBD3). These nuclear proteins are involved in DNA methylation mediated transcriptional repression. MBD3 lacks a functional methylated DNA binding domain but is an integral subunit of the histone deacetylase Mi2-NuRD complex recruited by MBD2 [22, 23]. The MBD2 expression is found to be involved in repression of tumor-promoter gene human telomerase reverse transcriptase (hTERT) in breast as well as in other cancer cell lines [24]. MIRA-assisted microarray profiling of global DNA methylation patterns in ductal carcinoma led to the conclusion that hypermethylation of the transcription factors encoded by homeobox genes is also a very common event leading to their silencing which suggests a critical role of homeobox gene methylation in breast carcinogenesis [25].

19.2.2 Histone Modifications in Breast Cancer

Among all the earlier discussed histone modifications, histone acetylation and methylation are relatively stable and are considered to be potential marks of the epigenetic modification carried over through multiple cell division cycles. The histone modifications have the tendency to open-up the chromatin or leading to its compaction, giving rise to ‘open’ or ‘closed’ conformations of chromatin. The chromatin in an open organization makes the DNA thread approachable to transcription factors and co-activators and such chromatin is transcriptionally active, called as euchromatin. The compacted state of chromatin is transcriptionally inactive and is known as heterochromatin. In general, histone acetylation at lysines (symbolized by K) H3K5ac, H3K8ac, H3K9ac, H3K12ac, H3K18ac and H4K16ac as well as H3K4me and arginine dimethylation H4R3me2 are considered as the markers of euchromatin [26–29]. The heterochromatin histone markers include mono (me)-, di (me2)-, or tri (me3)-methylated histone H3 at lysine 9 (H3K9me, H3K9me2 and H3K9me3), H3K27 (H3K27me, H3K27me2 and H3K27me3), and H4K20 (H4K20me, H4K20me2 and H4K20me3) which initiate and maintain the heterochromatin state [17, 26, 30, 31]. Mutations of p300 and its co-activator CBP lead to aberrant histone acetylation patterns in many cancer types including breast cancer [32]. Aberrant expression of other histone methyltransferases may also be implicated in breast cancer. The HDACs counterbalance the actions of HATs by deacetylating the lysine moieties leading to the compaction of chromatin. These enzymes are divided into four types based on their homologies to the yeast counterparts, cellular localization and acetylation activities.

Class I HDACs include HDACs-1, 2, 3, and 8, class II HDACs have two sub-classes: class IIA having HDACs-4, 5, 7 and 9; class IIB including HDACs-6 and 10. The class III HDACs include seven HDACs namely sirtuin 1-7 and class IV HDAC has a single member HDAC11. Class I HDACs (except for HDAC8) have been found to be up-regulated in breast cancer. A tumor suppressor gene, ARH1, is repressed through multiple histone deacetylases in breast cancer [33]. The inhibition of class II HDACs have effects on cell cycle progression, apoptosis, gene expression and estrogen receptor signalling in breast tumor cells [34]. Overexpression of HDACs-1, 6 and 8 is implicated in breast cancer invasion and also in the expression of matrix metalloproteinase-9 (MMP-9) [35]. The regulation of transcription by methylation of lysine moieties is dependent of location of that particular lysine and the degree of methylation. Histone methyltransferases (HMTs) are the enzymes responsible for this histone methylation process. Over expression of EZH2, a type of HMT has been associated with the breast cancer invasion and progression [36]. This protein works in association with HDACs and KMTs leading to transcriptional repression of associated genes [37]. Mono-, di- and tri-methylations on the histone H3K27 are the histone modifications which generally induce transcriptional repression and thus are regulatory in controlling gene expression patterns. Arginine methylation at the histone H3R2 by the enzyme PRMT6 antagonizes the binding of histone remodelling enzymes and transcriptional co-activators and thus leads to the transcriptional repression [38]. The tissue microarray of 880 breast tumor cases and normal breast tissues have shown a higher levels of acetylation at H3K18 while lower H4K16 acetylation in the breast tumors. Assessment of the relationship between histone modifications and patient outcome showed that, in the whole cohort, low-level detection of histone modifications was associated with adverse patient outcome. The histone modifications H3K4me2, H4K16ac and H4K20me3 in normal breast acini were reported to be higher than corresponding cancerous tissues. Longer disease free survival was reported in women with high levels of histone modifications such as H3K18ac, H4R3me2 and H3K9ac. In poorer prognostic subtypes of breast cancers, including basal carcinomas and HER2-positive tumors, moderate to low levels of lysine acetylation (H3K9ac, H3K18ac and H4K12ac), lysine methylation (H3K4me2 and H4K20me3), and arginine methylation (H4R3me2) were observed, which indicates the involvement of these histone markers in repression of transcription of the tumor suppressor genes [39].

19.2.3 miRNA Silencing in Breast Cancer

In 2005, altered miRNA expression in case of human breast cancer was first reported [40]. miRNAs interact with a variety of genes involved in different cellular pathways such as genes involved in cell division and cell cycle regulation. The silencing of miRNAs may take place either through hypermethylation of promoter sequences of genes coding for them, or through copy number variations as demonstrated by a study that 73 % of miRNA encoding genes are present in the chromosomal regions which are either frequently deleted or amplified in breast cancer [41]. Significant changes in the level of expression of the enzymes involved in miRNA processing (Dicer and AGO1) have been observed during breast carcinogenesis. The alternative mechanism of miRNA deregulation may be the alterations in the miRNA processing machinery [42].

Aberrant expression of miRNAs in human breast cancers is a frequent biological event [43]. miR-21 is an oncogenic miRNA which targets multiple tumor suppressor genes involved in p53 suppression pathway. This miRNA also promotes breast cancer invasion and metastasis [44]. miR-27a is a breast cancer oncogenic miRNA which down regulates expression of cell cycle inhibitors, leading to unregulated cell proliferation [45]. Up-regulation of miR-10b has been proven to be involved in breast cancer invasion and metastasis by targeting the HOXD10 mRNA [46]. Another group of researchers proved that miR-373 and miR-520c help in cancer cell migration and invasion, working as metastasis-promoting miRNAs through a CD44 suppression mechanism [47]. Conversely, some miRNAs (miR-335, miR-126, and miR-206) have been found to be human breast cancer metastasis-suppressor miRNAs. miR-335 functions in metastasis repression by targeting mRNA transcripts of a transcription factor SOX4 [48].

Five known groups of miRNAs are found to be the direct regulators of cell cycle progression. Let-7 is the most well studied family of miRNAs, which is involved in the targeting of Ras, HMGA2, and caspase-3 genes [49–51]. Many important cell cycle regulatory genes are repressed by let-7 such as cyclin D1, cyclin D3, cyclin A, CDK4, CCNA2, CDC25A, CDK6 and CDK8 [50, 52].

19.3 Epigenetic Targets and Therapeutic Strategies in Breast Cancer Treatment

19.3.1 Role of DNA Methyltransferases in Breast Cancer

Epigenetic targeting of DNMTs to reverse the down-regulation of methylation-silenced tumor suppressor genes is a promising approach for breast cancer prevention. The DNMT1 was found to be over-expressed in pre-invasive breast tumors when compared with the normal breast tissue [53]. DNMT1 inhibition by ASO98 in MDA-MB-231 cells lead to re-expression of estrogen receptor α (ERα) mRNA in these cells with other methylation silenced genes such as PGR, RARα and cyclinD1 [54]. Inhibition of DNMT1 has also been shown to inhibit anchorage-independent growth and tendency to invade, but on the contrary combined inhibition of DNMTs leads to activation of pro-metastatic genes leading to increased invasiveness [55]. Over-expression of DNMT3B was found to be a frequent event in breast carcinogenesis [56]. Inhibition of these enzymes to reactivate methylation silenced tumor suppressor genes as well as to inhibit the tumor promoter genes is currently the main therapeutic approach against breast cancer.

19.3.1.1 Nucleoside Analogue DNMT Inhibitors

A well known DNMT inhibitor, 5-azacytidine (AZC) is a ribonucleoside analogue and binds to RNA and DNA. It acts by inhibiting mRNA translation and by inhibiting DNA methylation by trapping DNMTs. At higher concentrations, this drug forms high levels of enzyme–DNA adducts. Another nucleoside analogue, 5-aza-2’-deoxycytidine, is a deoxyribonucleotide analogue which does not bind to RNA. This nucleoside analogue inhibits DNA methylation by the same mechanism as that of 5-azacytidine. Both these compounds are FDA approved for the treatment of myelodysplastic syndrome. 5-fluoro-2’-deoxycytidine is also a deoxyribonucleotide analogue of 5-azacytidine which induces DNA hypomethylation and cellular differentiation. Zebularine is another deoxyribonucleotide analogue, functioning in the similar way as other deoxyribonucleotide analogues [57].

19.3.1.2 Non-Nucleoside Analogue DNMT Inhibitors

Epigallocatechin-3-gallate (EGCG) is a major and most effective constituent of green tea polyphenols and is involved in direct inhibition of DNMT by forming hydrogen bonds to hinder the entry of cytosine into its active site [58]. It also leads to decrease in the level of available S-adenosyl-l-methionine (SAM) and an increase in S-adenosyl-L-homocysteine (SAH) and homocysteine levels, thereby providing evidence of an indirect inhibition of DNA methylation [59]. MG98 is an anti-sense oligonucleotide which highly specifically targets the 3’UTR of DNMT1 causing decreased DNA methylation [60]. Hydralazine is a compound which binds to the DNMTs in a similar manner as of 2’-deoxycytidine, 5-azacytidine, and 5-aza-2’-deoxycytidine as predicted by molecular docking experiments [61]. Procainamide is a non-nucleoside analogue which reduces affinity of DNMT1 for both DNA and S-adenosyl-methionine (SAM) causing a decrease in DNA methylation [62]. The structures of some of the nucleoside and non-nucleoside analogues having DNMT inhibitory activity are given in Fig. 19.3.

Fig. 19.3
figure 3

Different nucleoside analogue and non-nucleoside analogue DNMT inhibitors

Full size image

19.3.2 Role of Histone Deacetylases in Breast Cancer

Aberrant expression of histone deacetylases (HDACs) is observed in the process of breast carcinogenesis leading to the silencing of different types of tumor suppressor genes such as the genes involved in cell cycle regulation and apoptosis. The histone deacetylases, HDAC1 and HDAC6 were found to be over-expressed in breast tumor tissues [63, 64]. Inhibition of HDACs may lead to activation of silenced genes through two important mechanisms. The HDAC inhibitors can either help in the opening up of the chromatin by causing the accumulation of hyper-acetylated histones H3 and H4 or they can alter the nuclear DNMT dynamics to prevent the hypermethylation of the tumor suppressor genes. Cellular inhibition of SIRT1 using RNAi constructs in breast cancer MDA-MB-231 and MCF-7 cell lines had revealed the re-expression of tumor suppressor genes despite full retention of hyper-methylation of the promoters [65]. HDACs inhibition was reported to cause degradation of DNMT1 protein through an ubiquitin-dependent proteasomal degradation mechanism [66]. The following types of histone deacetylase inhibitors have been tested for breast cancer treatment and some of them have proved to be successful in reactivation of the silenced tumor suppressor genes and suppression of proto-oncogenes in breast cancer.

19.3.2.1 Short Chain Fatty Acids

Butyrate is a short chain fatty acid which causes hyper-acetylation of H3 and H4. Valproic acid (VPA) inhibits HDAC activity and relieves HDAC-dependent transcriptional repression. It has an added benefit of anti-tumor effects on both estrogen-sensitive and estrogen-insensitive breast cancer cells [67, 68]. Valproic acid in combination with 5-flurouracil is undergoing phase II clinical trials.

19.3.2.2 Hydroxamic Acids

Trichostatin A (TSA) is a HDAC inhibitor which acts by enhancing acetylation as well as the stability of the ERα and p300 protein. TSA synergizes with the demethylating agent 5-aza-2’-deoxycytidine in the re-expression of ER-α gene in breast cancer cells [69]. Suberoylanilide hydroxamic acid (SAHA) is another hydroxamic acid which inhibits both the class I and class II HDAC enzymes and found to be effective in breast cancer patients with amplification of Her2/neu. This is an FDA approved drug for the treatment of cutaneous T cell lymphoma patients.

19.3.2.3 Cyclic Tetrapeptides

Trapoxin (TPX) binds covalently to the histone deacetylases through the epoxide moiety and thus inhibits histone deacetylation irreversibly. TPX is an inhibitor of the cell cycle in eukaryotes and it also reverses the morphological changes in transformed cells [70]. Depsipeptide (FK228, FR901228), a bicyclic bacterial product, is a pro-drug which is activated by the action of glutathione and yields two free sulfhydryl groups which are capable of chelating the zinc in the HDAC active site. This compound exhibits a stronger direct inhibition of class I HDACs in comparison with class II HDACs [71, 72]. Apicidin [cyclo(N–O-methyl-l-tryptophanyl-l-isoleucinyl-d-pipecolinyl-l-2-amino-8-oxodecanoyl)] has been shown to increases the levels of acetylated histone H3 and H4 through inhibition of histone deacetylases in H-Ras-transformed human breast epithelial (MCF10A-Ras) and non-transformed mammary epithelial (MCF10A) cell lines [73].

19.3.2.4 Benzamides

Entinostat (MS-27-275), an active benzamide derivative, causes hyper-acetylation of nuclear histones H3 and H4 by selectively inhibiting class I HDAC enzymes. This drug also sensitizes TRAIL-resistant breast cancerous cells to radiation treatment [74]. CI-994 (N-acetyl-dinaline), a novel oral histone deacetylase inhibitor, is another substituted benzamide derivative which leads to accumulation of acetylated histones possibly in an indirect manner [75].

19.3.2.5 Psammaplins

Psammaplins are marine metabolites which were previously considered to be having dual DNMT and HDAC inhibition activity. Recent studies demonstrated that psammaplin A can efficiently inhibit bacterial DNA methyltransferase, but not the human DNMT enzyme. But these compounds can efficiently inhibit histone deacetylation activity in breast (MCF7), lung (A549), and normal human lung fibroblast (WI-38) cell lines [76]. Figure 19.4 depicts structures of some of the compounds with HDAC inhibitory activity.

Fig. 19.4
figure 4

Different types of HDAC inhibitors in breast cancer treatment

Full size image

19.3.3 Role of Steroid Hormone Receptors in Breast Cancer

The two estrogen receptors, estrogen receptor α (ERα) and estrogen receptor β (ERβ) are the nuclear hormone receptors. In unstimulated state, the estrogen receptors (ER) reside in cytoplasm bound to molecular chaperones. There are two modes of estrogen receptor activation, ligand-dependent, where ER is activated by estrogen binding and ligand-independent, which is the alternative mode of activation through growth factor signalling by phosphorylation of the receptor. In a classical ER response mechanism, after dimerization, ER binds to the estrogen response elements (EREs) present in the promoter regions of certain genes and mediates cellular proliferation as depicted in Fig. 19.5.

Fig. 19.5
figure 5

Classical estrogen response pathway of breast cancer proliferation. Estrogens (estrone and estradiol) are synthesized by the enzyme aromatase from the precursors, androstenedione and testosterone. The steroid hormone estrogen binds with the estogen receptors (ER) present in the cytoplasm and this estrogen-bound ER then enters the nucleus. After dimerization, the complex binds to the promoter of the genes containing estrogen responsive element (ERE) and leads to their transcriptional activation ultimately leading to cellular proliferation. The selective estrogen receptor modulators (SERMs) compete with the binding of estrogen to the ER while the selective estrogen receptor down regulators (SERDs) function in the down regulation of ERs

Full size image

ER has also been shown to regulate transcription of various key regulatory genes by non-classical pathway through binding with the non-estrogen responsive element sites such as activator protein (AP1) and specificity protein 1 (SP1) on the promoter regions of other transcription factors such as Jun and Fos. In such cases, ER functions as a co-regulatory protein (Fig. 19.6). Another non-genomic function of ER has been shown to be the activation of other growth factor receptors by some still unknown mechanism. Due to these three different modes of actions, ER is proven to be important for breast cancer progression and thus targeting of this gene may prove fruitful in the breast cancer therapy [77].

Fig. 19.6
figure 6

Non-classical estrogen response pathway of breast cancer proliferation. The steroid hormone estrogen binds with the estrogen receptors (ER) present in the cytoplasm and this estrogen-bound ER then enters the nucleus. After dimerization, estrogen-bound ER promotes transcriptional activation by binding to the non-estrogen responsive element sites such as AP1 and SP1 on the promoters of other transcriptional factors such as Jun and Fos. Thus, estrogen may also function as a transcriptional co-activator leading to the transcriptional activation of genes associated with cellular proliferation. The selective estrogen receptor modulators (SERMs) compete with the binding of estrogen to the ER, while the selective estrogen receptor down regulators (SERDs) function in the down regulation of ERs

Full size image

Expression of progesterone receptor (PR) is dependent on the expression of ERα in breast cancer cells. Previously, the gene coding for PR was known to be involved in inhibiting the proliferation of the uterine endometrium. Due to this inhibitory role, this gene was thought to be of lesser importance in breast carcinogenesis. However, later studies proved that progesterone helps in proliferation of breast cells in a mechanism independent of estrogen, and thus it is associated with breast cancer risk. Progesterone primarily induces cellular proliferation in a paracrine fashion [78]. Paracrine progesterone secretion-induced development of alveoli is mediated by receptor activator for nuclear factor-κB ligand (RANKL)-mediated downstream signalling in PR-negative breast cancer cells. Progesterone receptor also affects proliferation through CCND1-dependent mechanism in PR-positive breast cancer cells in autonomous manner [79]. This dual role of progesterone in cell proliferation in both the PR-positive and PR-negative cancer types makes it a strong candidate for epigenetic targeting. The treatment of DNMT and HDAC inhibitor combination (5-azacytidine and Trichostatin A) lead to re-expression of PR gene in PR- negative breast cancer cell lines [80]. Inhibition of either the estrogen receptors or alternative inhibition of estrogen synthesis is an important strategy for treatment of hormonal receptor positive breast cancer.

19.3.3.1 Selective Estrogen Receptor Modulators

These are small synthetic molecules which compete with estrogen for binding to the ER and can exert different effects on the estrogen target genes. The typical example of this category is tamoxifen, which functions by acting as a competitive inhibitor of estrogen to show agonistic or antagonistic effects on the estrogen-related transcriptional patterns in breast cancer cells to inhibit cancer cell proliferation. The tamoxifen bound ER can dimerize and bind to the DNA, but the downstream effects are different. For example, the receptor bound balance of co-activators and co-repressors differs for estrogen-bound and tamoxifen-bound estrogen receptors. This results in different effects on different target genes. Tamoxifen-bound ER may block AF-2 mediated gene transcription but AF-1 mediated transcription remains unaffected [81]. Later on, to improve the agonist/antagonist profile, tamoxifen like triphenyethylene SERMs were developed. Toremiphene, a chlorinated analogue of tamoxifen, has no significant difference in efficacy or binding affinity to the former [82]. Droloxifene (3-hydroxytamoxifen) was shown to have a 10-fold higher binding affinity for ER relative to tamoxifen [83]. Idoxifene was another tamoxifen analogue which was later on found inefficacious. The fixed ring selective estrogen receptor modulators were developed as the second or third generation of SERMs. This class of SERMs include raloxifene, arzoxifene, acolbifene, ERA-923 etc. They also function in similar way to tamoxifen but have comparatively lesser side effects. Figure 19.7a shows the structures of some of the available SERMs.

Fig. 19.7
figure 7

Different types of estrogen receptor modulators, downregulators and aromatase inhibitors in breast cancer treatment

Full size image

19.3.3.2 Selective Estrogen Receptor Down-Regulators

These pure anti-estrogens induce destabilization and degradation of estrogen receptor and have no agonistic activity. They function by sterically hindering the dimerization of the ER leading to increased ER turnover and disruption of nuclear delocalization leading to lesser ER molecules in cells [84]. The typical example of this class is the steroid fulvestrant, which has been approved for the treatment of metastatic breast cancer (Fig. 19.7b). Fulvestrant-mediated destabilization of the ERs is dependent on pattern of ubiquitination of the protein [85]. Other examples of SERDs include IC182780 and ZK-703. Use of SERDs has been proved to be the most effective in the treatment of ER-positive tamoxifen-resistant breast cancers.

19.3.4 Role of Aromatase in Breast Cancer

Aromatase is an enzyme involved in synthesis of estrogens, estrone and estradiol from androgenic substrates androstenedione and testosterone respectively [86]. The enzyme is encoded by a CYP19A1 gene and the process of transcription of this gene is regulated by tissue specific promoters. Majority of breast cancer tumors show higher levels of aromatase activity than corresponding non-malignant mammary tissues. Inhibition of aromatase is a well-known therapeutic strategy to reduce estrogen expression for the treatment of estrogen receptor-positive breast cancer in post-menopausal women. But, nonspecific aromatase inhibitors (AIs) lead to reduction of aromatase activity in other tissues such as bone, brain and adipose tissue. Due to this non-specificity, they are associated with undesirable side effects such as bone loss and abnormal lipid metabolism. Breast tumor-specific inhibition of aromatase can prove to be a far better approach in this regard.

19.3.4.1 Aromatase Inhibitors

There are two types of aromatase inhibitors including irreversible steroidal activators and reversible non-steroidal imidazole-based inhibitors. Steroidal agents are irreversible inhibitors of aromatase, which inhibit aromatase by competing with the physiological aromatase substrates (androstenedione and testosterone). The non-steroidal inhibitors reversibly interact with the cytochrome P450 moiety of the enzyme and thereby inhibit enzymatic activity. Examples of this type of aromatase inhibitors are letrozole, anastrozole, exemestane [87]. The structures of some of the aromatase inhibitors are given in Fig. 19.7c.

The first generation aromatase inhibitor AG3 which was used for breast cancer treatment for more than one decade, showed potential side effects. It was then withdrawn from the market. Then, the second generation steroidal aromatase inhibitor, 4-hydroxyandrostenedione (4-OHA, formestane) had a limitation of pharmacological availability. The third generation aromatase inhibitors are being used as adjuvant therapy in place of tamoxifen, the ER inhibitor in the treatment of post menopausal breast cancer, because they are well-tolerated and are more effective. Exemestane (6-methylenandrosta-l,4-diene-3,17-dione) is a orally administered third generation steroidal aromatase inhibitor available in the market [88].

19.3.5 Role of Breast Cancer Susceptibility Genes in Breast Cancer

Although mutations of the breast cancer susceptibility gene 1 (BRCA1) are well known to be involved in a majority of breast cancers, the methylation status of the proximal portion of BRCA1 promoter is also highly influential in deciding the BRCA1 expression level in the breast cancer cells. Hypermethylation of the BRCA1 promoter significantly increases in the breast cancer cells in comparison to the normal tissues [89]. BRCA1 product, a multifunctional protein is involved in DNA repair, cell cycle check point control and chromatin remodelling as well as protein ubiquitination. It also interacts with a variety of proteins involved in epigenetic modifications such as histone deacetylases, and different components of chromatin remodelling complexes. Hypermethylation of BRCA1 promoter may also function as the second hit to silence the expression of this gene in a subset of BRCA1 mutation carrier women [90]. Higher levels of BRCA1 promoter methylation were also correlated with advanced breast cancer stages and increased mortality [91].

Mutations in the other breast cancer susceptibility gene BRCA2 are also implicated in the development of some sporadic breast cancers. This gene shows transcriptional activation either through the presence of HAT activity or through its association with transcriptional activator p300/CBP-associated factor (P/CAF). This gene was found to be over-expressed in breast cancer [92]. BRCA2 was found to be hypermethylated in approximately 70 % of the patients [93]. The breast cancer patients defective in BRCA1 and BRCA2 genes depend on an alternate DNA repair process mediated by poly (ADP-ribose) polymerase 1 (PARP1). Thus, tumors deficient in BRCA1 gene can be well targeted by PARP1 inhibitors leading to lack of DNA repair and enhanced cell death. The examples of PARP1 inhibitors include nicotinamide analogs (first generation), benzamide analogs (second generation) and 3-aminobenzamide analogs (third generation PARP1 inhibitors). AG014699, veliparib (ABT-888), olaparib and iniparib are the PARP1 inhibitors which have been undergoing the different stages of clinical trials [94].

19.4 Conclusion

The process of breast carcinogenesis is well-orchestrated by genetic and epigenetic mechanisms and therapeutic strategies targeting both these processes at the same time may prove to be a better approach for effective breast cancer treatment. The epigenetic mechanisms mainly DNA methylation and histone modification have been shown to play key regulatory role in breast cancer initiation and progression. Silencing of important tumor suppressor genes and activation of proto-oncogenes into oncogenes by epigenetic modifications facilitates the mammary cells to acquire transformed phenotype. Methylation status of frequently methylated genes such as RASSF1 and BRCA1 in breast cells has proven to be important early biomarker to detect breast cancer susceptibility. Further, genome-wide methylation profiling would add an extra mile to identify other important early biomarkers to detect the risk of developing breast cancers in highly susceptible women. Targeting one or more epigenetic modifications such as DNA methylation, histone modifications and miRNA-mediated silencing is currently growing therapeutic approach against breast carcinogenesis. Many different types of anti-cancerous compounds with epigenetic mechanism of action are being used in breast cancer therapy and a variety of new compounds are also being investigated in clinical trials. However, detailed understanding of these epigenetic modifications with respect to their part in breast carcinogenesis might lead to the significant advancement in the prevention and treatment strategies against breast cancer.