Cancer is currently recognized as a group of diseases characterized by an abnormal and uncontrolled division of cells, the formation of cellular masses or localized tumors, and the simultaneous formation of independent circulatory and nutrient systems (angiogenesis). Such cells can advance to an invasive-degenerative phenotype in both surrounding and distant tissues through the blood or lymphatic systems via a process known as metastasis, which is the principal cause of death in patients with advanced tumors. Currently, the different types of cancer that affect human beings are considered to be a worldwide public health problem, affecting nearly 13 million people and causing the deaths of eight million people annually [1]. Among the main molecular characteristics that precede the development of this disease are variations in gene expression and genomic instability; these variations are a consequence of disturbances in the processes that regulate protein expression at the genetic and epigenetic levels.

At the functional level, genetic information resides inside the chromatin, which is a structure in the cell nucleus consisting of repetitive units of smaller structures called nucleosomes [2,3]. Nucleosomes consist of 146 base pairs (bp) of DNA wrapped in 1.75 turns around a histone octamer, all of which is organized as a central tetramer of H3/H4 histones surrounded by two histone H2A/H2B dimers [4,5]. The DNA between nucleosomes can be associated with a fifth histone, H1, to achieve the high-order compression of the chromatin [3,6]. At the structural level, two main chromatin conformations are distinguishable: euchromatin, with a low degree of compression associated with a “relaxed” conformation of the histone-DNA macro-complex, and heterochromatin regions, with a high degree of compression that can be accessory (susceptible to changes in the degree of compression) or constitutive (which is observed during the cell cycle) [3,6]. In epigenetics, variations in the structure or the degree of chromatin compression lead to the regulation of the genes associated with a specific genomic region because these variations facilitate or prevent access of the nuclear proteins required for processes such as gene expression [7]. In this manner, epigenetics regulates the reversible change from heterochromatin (compressed structure associated with the repression of gene expression) to euchromatin (structure accessible to the transcriptional machinery). This regulation occurs via mechanisms that involve DNA methylation and post-translational histone changes such as methylation, citrullination, and ADP-ribosylation of arginine (the latter modification also occurs at glutamic acid residues); methylation, acetylation, biotinylation, ubiquitination, and sumoylation of lysine; phosphorylation of serine and threonine; and the cis-trans isomerization of proline (Figure 1). Each of these modifications has a characteristic effect on the level of gene expression. For example, the addition of acetyl groups to the histones is frequently related to transcriptional activation, whereas deacetylation produces the opposite effect. Conversely, the methylation-demethylation of lysine can be associated with transcriptional activation or repression depending on the function of the residue and the degree of methylation [8,9].

In summary, these processes contribute to a finely regulated mechanism of control of genetic expression and are responsible for maintaining the cellular “equilibrium”. In fact, alterations in one or more components of such mechanisms can result in alterations in gene expression and/or cellular phenotype, which are the main causes of the emergence of several pathologies, including neurological diseases, diabetes, and diseases associated with endocrine dysfunction, as well as the different varieties of cancer that can affect human beings [10,11,12,13,14,15]. Consequently, major efforts have been directed toward the identification and characterization of different epigenetic regulators whose phenotypes have been altered in tumor cell lines with the goal of identifying potential therapeutic targets, including DNA-methyltransferases [16,17,18], histone acetyltransferases/deacetylases [17,19,20,21,22,23,24], and histone methyltransferases/demethylases [25,26,27,28,29], for the development of more efficient cancer treatment strategies.

The current work provides an updated description of a large family of histone demethylases that are responsible for maintaining the cellular phenotype by regulating histone methylation levels and serve as promising targets in the development of new treatments against a large variety of forms of cancer. New knowledge will be discussed in relation to the search for new anti-cancer targets, which could be targeted by a new generation of specific drugs directed against these altered histone modifiers.

In a large number of cellular processes, covalent modifications induced by methylation can affect different nitrogen-bearing amino acids, such as arginine, histidine, and lysine [30,31,32,33,34]. Histones can be mono-, di-, or tri-methylated at their lysine residues, and they can be mono- or di-methylated at arginine residues in a symmetric or asymmetric fashion (Figure 2).

Previously, histone methylation was considered to be a stable and irreversible mark of chromatin [35]. However, in 2004, Shi et al. described the mechanism of action of the first protein with specific demethylase activity against a lysine residue, LSD-1/KDM1A (lysine-specific demethylase-1), breaking the paradigm of irreversible methylation and opening a new research outlook on these proteins and their association with epigenetic regulatory mechanisms [36].

The discovery of new sequencing and analysis technologies has allowed for the identification of the histone methylation pattern in the human genome [37], which, in turn, identifies associations between cellular phenotypes and methylation-demethylation as part of regular cell development and in processes related to the emergence of different pathologies. Inside the nucleosome, the H3 and H4 histones are the main targets of these modifications, which have been associated with the development of different pathologies, including cancer, that are associated with epigenetic defects (Figure 1 and Figure 3) [38,39]. Such observations have produced significant interest in the characterization of different proteins associated with the epigenetic regulation of histone methylation, among which the recently described histone demethylases are promising targets for future treatments against tumor cells.

After the first characterization of a protein with demethylase activity, much effort has been made to identify and characterize other proteins with this activity in histone residues. The main proteins that have been discovered to date can be classified into two super-families: those belonging to the amine oxidase superfamily, which are dependent on FAD as a co-factor, and those corresponding to the oxygenase superfamily, in which the demethylase activity is dependent on Fe(II) and α-ketoglutarate. This latter superfamily is associated with the presence of a characteristic domain termed Jumonji (Figure 4). Among the most important histone demethylases associated with the development of cancer are LSD-1/KDM1A, and members of the Jumonji family JARID1A-1C/KDM5A-5C, JHDM1B/KDM2B, JMJD2C/KDM4C, JMJD2A/KDM4A, JMJD3/KDM6B, and UTX/KDM6A (Table 1), which will be analyzed in detail in the following paragraphs.

Recently, an indirect or direct association has been established between histone demethylases and the emergence and progression of cancer. Among the potential strategy for targeting KDMs, the identification of molecules with the capacity to specifically inhibit the enzymatic activity of these proteins represents a promising path for treatment [10,188]. As the first member of this important family of proteins, LSD-1 represents an important target for the so-called “epi-drugs”. Among the molecules with pharmaceutical potential are oligoamine analogs, originally designed based on the significant similarity between LSD-1 and other FAD-dependent polyamine oxidases (Figure 5a, top panel) [109,189], and specific inhibitors possessing a common 2-phenyl cyclopropyl-1-amine nucleus [108,118,190,191]. These specific inhibitors for LSD-1 were designed based on the mechanism of action of the enzyme (which is similar to that of monoamine oxidases), and are able to reach inhibition constants in the micromolar range (Figure 5a, lower panel) [108,118,190,191].

These findings have promoted the search for new inhibitors and have resulted in the recent description of namoline, a gamma-pyrone capable of inhibiting LSD-1 and altering the growth of prostate cancer tumor cells [192]. The first generation of LSD-1 inhibitors, which are of the tranylcypromine type, exhibited low potency and selectivity. After the search was shifted to focus on chemical structures that target the active site of the protein, selectivity was increased [193]. In vitro and in vivo studies using nanomolar concentrations of phenocopied, a tranylcypromine analog and inhibitor of LSD-1, demonstrated a pro-apoptotic effect in primary acute myeloid leukemia (AML) cells without affecting the repopulation potential of hematopoietic stem cells and progenitor cells [194]. These data suggest that LSD-1 is a key protein for the development of selective therapeutic targets for leukemia.

In the same manner, the inhibition of LSD-1 can increase the sensitivity of promyelocytic leukemia (APL) to the treatment with all-trans-retinoic acid (ATRA). Usually ATRA loses its activity in APL, and the cause of this alteration appears to be a reduction in the methylation of lysine 4 of histone H3. Under such conditions, the inhibition of LSD-1 can facilitate the activity of ATRA in APL cells, as it has been validated using the LSD-1 inhibitor trans-2-phenylcyclopropylamine [195].

In addition, iron- and α-ketoglutarate-dependent histone demethylases are also targets of important studies, allowing for the identification of inhibitors designed on the basis of the mechanism of the unique activity of these enzymes. Among the most important inhibitors, several molecules that are analogous to α-ketoglutarate have been described. These inhibitors range from molecules designed using N-oxalylglycine or N-oxalyltyrosine as a base molecule [25,162,196] (Figure 5b, upper panel) to natural products such as flavonoids and catechols that were identified as a result of screening libraries of pharmacologically active compounds [197] (Figure 5b, lower panel). Demethylase inhibitors for Jumonji-containing domain proteins usually compete with the 2-oxoglutarate factor and bind to the catalytic region containing iron. Compounds that possess high polarity, such as 2,4-pyridine-decarboxylases, inhibit the activity of Jumonji by interfering with its enzymatic activity, in the same way other 2-oxoglutarate-like molecules inhibit the HIF prolyl hydroxylase-1 oxygenases HPH1/EGLN2 and HPH2/EGLN1.Interestingly, two new series of jumonji demethylase inhibitors, 8-hydroxyquinolines (e.g., SID 85736331) and 2,2-bipyridines, are potent inhibitors with higher selectivity and better pharmacological properties than those exhibited by other inhibitors. The structures of these two new types of compounds make them promising therapeutic alternatives that could be pharmacologically enhanced for better penetration into the cell interior [198,199].

These results clearly highlight the potential of the histone demethylase superfamily members as therapeutic targets against the diverse neoplastic diseases that affect human beings. There is support for structural analyses to increase the understanding of the spatial enzyme-inhibitor interaction [99,200,201] and, hence, promote the specific molecular design of a new generation of epi-drugs targeting members of the amine oxidase superfamily [108,202,203] or those belonging to the oxygenase family [25,159,162].

Regulation of gene expression encompasses a complex network of molecular interactions that define the cellular phenotype of tissues and organisms. Alteration on the epigenetic balance can result in the emergence of terminal diseases, among which are the several forms of human cancer. Although, much work remains to be taken in the search for new strategies against the different types of cancer, in recent years, significant advances have been made in the identification and characterization of new potential targets for the development of specific therapies to treat multiple forms of cancer. Together with the continuous discovering of new macromolecules associated with human neoplasm, the recent discovery of histone demethylases has provided a new perspective in the search for anti-cancer therapeutic targets due to their unique enzymatic activity and their association with the cellular phenotype of a large number of human tumor lines.

Either as early markers or as anti-tumor candidates, histone demethylases currently represent a promising group of macromolecules that increase our understanding of malignity and the mysteries of their formation, development and evolutionary progress. LSD1 was the first demethylases cloned and perhaps the most promises protein as therapeutic target. Recently evidences showed that LSD1 played an important role in a broad spectrum of biological processes, including cell proliferation, adipogenesis, spermatogenesis, chromosome segregation and embryonic development. Furthermore, LSD1 also could promote progress of tumor by inhibiting the tumor suppressor activity of p53. To date, as a potential drug for discovering anti-tumor drugs, the medical significance of LSD1 inhibitors have been greatly appreciated. However, it is necessary to be attentive when analyzing and applying this new knowledge, as relevant targets as LSD-1 could be observed as an oncogene [31,41,42,43,44,45,46,47,48,49] or a tumor-suppressor [50,51], in an apparent situation-dependent biological role for this protein. These observations, enhanced the importance of histone demethylases for cellular and tissue phenotype maintenance as part of macro-complexes [51,114,127,152,169], must be taken into account to overcome any side effects facing their use as therapeutic targets [106]; thus highlighting the importance of improving our comprehension about biological and molecular role of histone demethylases, to be established as potential targets for a new generation of epigenetic drugs directed against their enzymatic activity. Various demethylases from Jumonji family are in preclinical studies for cancer including prostate cancer, glioblastoma, breast cancer, and diverse kind of leukemia.

In summary, this report is part of the future on the continuous fight against cancer; searching for new specific therapeutic strategies based on the understanding of the epigenetic regulators discussed herein and those different molecules rationally designed against these specific targets; the epi-drugs.