Abstract
Lymphocyte differentiation and identity are controlled by signals in the microenvironment that ultimately mediate gene expression in the nucleus. Although much focus has centered on the strategic and often unique roles transcription factors play within lymphocyte subsets, it is increasingly clear that another level of molecular regulation is crucial for regulating gene expression programs. In particular, epigenetic regulation is critical for appropriately regulated temporal and cell-type–specific gene expression during immune responses. As such, mutations in epigenetic modifiers are linked with lymphomagenesis. Furthermore, certain infections can remodel the epigenome in host cells, either through the microenvironment or by directly co-opting host epigenetic mechanisms, leading to inappropriate gene expression and/or ineffective cellular behavior. This review will focus on how histone modifications and DNA methylation, and the enzymes that regulate the epigenome, underpin lymphocyte differentiation and function in health and disease.
Introduction
Immune cells rely on intrinsic regulators to mediate cell fate decisions in response to environmental signals. Lymphocytes participating in immune responses create effectors tailored for the infecting pathogen as well as form immune memory populations that will provide immunity to subsequent infections. They are able to do this by activating or repressing gene expression programs depending on the cell fate and/or function required. Different signals in the microenvironment will induce the expression of specific transcription factors that regulate gene programs that may be unique to a cell subset or functional attribute. Focusing in on lymphocytes, transcription factors regulate lineage specification in the bone marrow, Th cell subset diversity to different pathogens, and effector versus memory T cell differentiation, as well as B cell differentiation and Ab diversity. However, there are some fate decisions and lymphocyte subsets, such as memory B cells, that are not associated with an identifiable transcription factor (1, 2). Thus, recent research focus has shifted to understanding how controlling the access of transcriptional machinery to genes by modulating chromatin structure regulates gene expression.
Epigenetic regulation is an important regulator of lymphocyte fate decisions during development in primary lymphoid organs as well as differentiation in the periphery. There are three broad types of epigenetic regulators: DNA methylation, microRNA, and histone modifications. Together with transcriptional regulators, this combinatorial process is important for both effective cellular responses to infection and prohibiting inappropriate gene expression programs that could lead to autoimmunity or cancerous outcomes. This review will mainly focus on histone modifications but will also note some crucial roles of DNA methyltransferases (DNMTs).
Types of epigenetic modifiers: histone-modifying complexes
Chromatin accessibility can be modulated through histone posttranslational modifications performed by enzymes that alter the N-terminal tail of histones (3). These enzymes can deposit or remove posttranslational modifications that include methylation, acetylation, sumoylation, and ubiquitylation. There are two broad groups of histone-modifying complexes—the Trithorax and Polycomb group proteins. These chromatin modifiers are critical for stable activation and repression of gene expression, respectively. For example, the Trithorax protein MLL2 can form complexes to trimethylate lysine 4 of the histone 3 (H3K4), permitting transcription (4, 5), whereas the two Polycomb repressive complexes work together to repress gene expression through di/trimethylation of H3K27 (PRC2) and monoubiquitination of histone H2AK119 (PRC1) (6). Histone posttranslational modifications are dynamic and reversible processes which include the enzymatic activity of demethylases and acetyltransferases. There is a wide range of histone lysine demethylases (KDMs) (7), although the unique roles of each individual KDM in regulating immune responses are only beginning to be revealed.
Types of epigenetic modifiers: DNMTs
The level of chromatin compaction and the transcriptional status are also determined by the presence of covalent methyl groups on specific regions of the DNA. This phenomenon is orchestrated and maintained by DNMTs (8), a highly conserved family of proteins that influences the accessibility of transcription factors to their responsive elements in the DNA, thus regulating gene transcription (9). In mammals, the three major members of the DNMT family are DNMT1, DNMT3A, and DNMT3B. DNMT1 maintains methylation patterns, whereas DNMT3A and DNMT3B are critical for de novo DNA methylation (10).
Epigenetic modifiers regulate cell fate decisions during immune responses
Epigenetics plays a major role during lymphocyte development and in mediating cell differentiation during immune responses. Distinct patterns of histone marks and DNA methylation have been shown to be crucially and intimately involved in the maintenance of T cell identity and B cell development, in mediating differentiation of immune cells in the periphery, and in the functional capacities of T cell subsets (11–18).
T cells.
De novo DNA methylation carried out by both DNMT1 and DNMT3 shapes T cell responses, playing an essential role in their homeostasis and in regulating T cell lineage-specific gene expression during an immune response (15, 16). Activated Ag-specific CD8 T cells exhibit a rapid demethylation at the IL-2 and IFN-γ loci and significant histone acetylation at the IFN-γ promoter and enhancer regions. These epigenetic modifications occur in the effector stage of the response and are maintained through immune memory development (17, 18). Similarly, deposition of histone marks regulates the potential of cells to become either memory or effector T cells. In particular, PRC2 is important for depositing H3K27me3 on promemory genes to repress their expression in effector T cells. In contrast, the loss of PRC2 does not result in a severe loss of memory T cells (19). Furthermore, the functional role of genes in effector T cells can be predicted by the epigenetic signature of bivalent genes in naive T cells (14). Bivalent genes are characterized by the presence of both an activating mark, typically H3K4me3, and a repressive mark, such as H3K27me3, at the same loci. The gene is primed for transcription but is not active, poising it to be expressed upon activation and/or differentiation. Although the idea of bivalency is controversial, these studies have used sequential chromatin immunoprecipitation to determine the presence of permissive and repressive modifications at the same loci. Thus, CD8 T cell fate choices are shaped by the action of DNMTs and the critical balance between permissive and repressive histone modifications, regulated by the deposition and removal of H3K4me3 and H3K27me3 throughout an immune response to infection. These layers of regulatory mechanisms allow T cells to clear an infection while maintaining plasticity within the memory population to respond effectively to a secondary infection.
B cells.
T-dependent B cell immune responses generate three main populations: 1) an extrafollicular plasmablast population that provides a wave of low-affinity Ab early in the response; 2) formation of germinal centers, transient anatomical structures located in secondary lymphoid organs and involved in the generation of high-affinity immune memory (20); and 3) the formation of immune memory, comprising of long-lived plasma cells and memory B cells (20). Differentiation into each of these fates requires large changes in transcriptional programs, and these changes have recently been shown to be mediated epigenetically. B cell differentiation during an immune response is associated with changes of the DNA methylome as well as H3K4me3 and H3K27me3 modifications. Activation of naive B cells and differentiation into germinal center B cells is accompanied by a marked demethylation of the genome (21), and PRC2 (EZH2 in particular) has been shown to be critical for both germinal center formation (22–24) and Ab production (25). An example of the critical regulation by both histone modifications and DNA methylation during B cell differentiation is that of activation-induced cytidine deaminase (AID), an enzyme that is essential for somatic hypermutation and class-switch recombination within germinal center B cells (26). Methylation of the Aicda promoter (the gene encoding AID), together with H3K27me3 deposition and low levels of H3 acetylation, corresponds to its repression in naive B cells (23, 27). Upon encounter with the Ag, naive B cells acquire an activated phenotype with decreased DNA methylation and H3K27me3, coupled with enriched H3K4me3 modifications, at the Aicda locus (28). Similar mechanisms can be applied for other genes involved in B cell development, including Bcl6 for germinal center formation (29), Prdm1 for plasma cell differentiation (30), and Pax5/Bcl2 in the memory compartment (31).
Unlike CD8 T cells, much less is known about the epigenetic modulations required to produce memory B cells or the role they play during secondary responses. Memory B cell subsets defined by either Ig isotype or cell surface markers (32, 33) have been linked with mediating either rapid differentiation into plasmablasts during secondary responses or the continual persistence of the memory population. Deletion of the H3K9 acetyltransferase MOZ resulted in a difference in the formation of these subsets, and thus an alteration in the ability to form secondary high-affinity plasmablasts (11). This study was an initial step in identifying a role for histone modifiers in memory B cell function, but much more work is required to fully understand the molecular underpinnings of fate choices made by memory B cells. Although epigenetic regulation of immune cell differentiation is still a relatively nascent field, it is clear that dynamic temporal modulation of permissive and repressive marks is critical for effective B and T cell function during immune responses. Furthermore, disruption of this delicate balance can lead to dysfunctional cellular behavior.
Together, these enzymes are dynamically modulated throughout development and during immune responses to mediate multitudinous cell functions and differentiation steps important for normal biology, but are also emerging as important clinical targets for numerous and diverse immune-mediate disorders. In particular, modifications that promote aberrant gene expression are features of lymphomagenesis, subversion of appropriately regulated immune responses by infectious agents, and a selection of B or T cell–driven autoimmune diseases. We will detail the current understanding of the epigenetic mechanisms underlying lymphocyte dysregulation in these disorders below, focusing in on key examples. As the number of cancers in which epigenetics has taken a prominent role in research is numerous, we will focus on B cell lymphomas in the following section.
Epigenetic regulation of lymphoma development and epigenetic targets for therapeutics
B cell non-Hodgkin lymphomas (B-NHLs) comprise a heterogeneous group of tumors arising from different stages of mature B lymphocytes. This diverse group of malignancies usually develops in the lymph nodes but can occur in almost any tissue and shows extremely variable clinical courses, ranging from very indolent to tightly aggressive (34, 35). The pathogenesis of B-NHL is associated with distinct genetic lesions within germinal center B cells as carrier of hypermutated IgV genes (36). Chromosomal translocations and aberrant somatic hypermutation represent the main mechanisms of genetic lesions that contribute to lymphomagenesis (37). The most common germinal center–derived lymphomas are follicular lymphoma (FL) and diffuse large B cell lymphoma (DLBCL), with a 25–40% of incidence rate among all B cell malignancies (38). With the introduction of next-generation sequencing, the genetic landscape of these complex malignancies has rapidly been unraveled in recent years. Multiple studies have demonstrated that the heterogeneity of the epigenome is a feature of many B cell lymphomas (Fig. 1). Epigenetic alterations in lymphomas are predominantly driven by mutations in chromatin-modifying enzymes that lead to changes in promoter accessibility for transcription factors or protein complexes involved in DNA methylation. Perturbed DNA methylation patterns ultimately induce the dysregulation of tumor suppressor genes, leading to sustained proliferation and tumor progression or creating an oncogenic form of a normal gene (39). Recent advances in the characterization of the mutational landscape of FL and DLBCL have revealed dysregulation of genes that encode for epigenetic modifiers, such as MLL2, CREEB, EZH2, and EP300 (40, 41). The methyltransferase EZH2 is often constitutively activated in germinal center–derived non-Hodgkin lymphomas owing to gain-of-function mutations within the catalytic SET domain, most notably affecting tyrosine 646 (Y646) (42). This mutation is mainly found in a monoallelic state, cooperating with the wild-type counterpart. Functional studies revealed that the Y646-mutated protein has a higher affinity for H3K27me2 than the wild-type protein, leading to a widespread redistribution of the H3K27me3 marks. The substantial relocation of methyl groups leads to transcriptional changes at putative PRC2 target genes, including AID. AID is crucial for germinal center reactions, and its dysregulation effects somatic hypermutation, class-switch recombination, and germinal center size and may promote lymphomagenesis (23, 43). Additional human genetic studies revealed that mutations affecting other PRC2 subunits (i.e., EED, SUZ12, and JARID2) cause hematopoietic diseases including myelodysplastic syndrome, T cell acute lymphoblastic leukemia, and B cell lymphoma (44, 45). Further detailed studies are required to determine their exact functions in lymphoma development. Similarly, PRC1 components such as RING1A, RING1B, and BMI1 have been implicated in lymphomas such as FL, DLBCL, and Burkitt lymphoma (46). For example, the recruitment of the canonical PRC1 containing BMI1/PCGF4 plays a central role in repressing the transcription of target genes such as the CDKN2A locus (encoding for p16Ink4a and p19Arf), which is critical for maintaining cellular self-renewal capacity (47). Overexpression of BMI1 has been linked to a variety of malignancies, including DLBCL (48); however, its specific role in the genesis of lymphomas is not fully understood. In summary, a number of PRC2 and PRC1 members have been associated with B cell–derived lymphomas. Although the role of EZH2 in germinal center formation has been studied in detail, mechanistic insight into the roles of other PRC2 components, as well as the PRC1, during immune responses, and the implications of their dysregulation, warrants further investigation.
Epigenetic dysregulation in B cell–driven lymphomas. Deregulation of epigenetic modulators gives rise to distinct types of germinal center–derived lymphomas. The yellow box refers to genetic alterations that contribute to the origin of germinal center B cell–like DLBCL, notably induced by EZH2 gain-of-function (yellow star), BMI1 overexpression (green star), and MLL2 inactivation (red star). These mutations, either individually or in combination, lead to the repression of multiple tumor suppression genes (blue line in the DNA strand) through the deposition of methyl or ubiquitin groups (H3K27me3, red stems; H2AK119ub, purple stems) and the subsequent development of the malignancy. In contrast, FL is prominently associated with MLL2 inactivation (red star) and chromosomal recombination with the Mll gene, which originates MLL fusion proteins (black dot). Addition of methyl groups to the histone 3 (H3K4me3, orange stems) activates the expression of its target genes. AF9 recruits Dot1L modifiers that deposits activating methyl groups to H3K79me2 (blue stems). This dual mechanism leads to the hyperactivation of target genes, such as oncogenes, that are involved in the genesis of FL.
Epigenetic dysregulation in B cell–driven lymphomas. Deregulation of epigenetic modulators gives rise to distinct types of germinal center–derived lymphomas. The yellow box refers to genetic alterations that contribute to the origin of germinal center B cell–like DLBCL, notably induced by EZH2 gain-of-function (yellow star), BMI1 overexpression (green star), and MLL2 inactivation (red star). These mutations, either individually or in combination, lead to the repression of multiple tumor suppression genes (blue line in the DNA strand) through the deposition of methyl or ubiquitin groups (H3K27me3, red stems; H2AK119ub, purple stems) and the subsequent development of the malignancy. In contrast, FL is prominently associated with MLL2 inactivation (red star) and chromosomal recombination with the Mll gene, which originates MLL fusion proteins (black dot). Addition of methyl groups to the histone 3 (H3K4me3, orange stems) activates the expression of its target genes. AF9 recruits Dot1L modifiers that deposits activating methyl groups to H3K79me2 (blue stems). This dual mechanism leads to the hyperactivation of target genes, such as oncogenes, that are involved in the genesis of FL.
MLL proteins induce an open chromatin conformation, allowing for activation of target genes through the deposition of methyl groups on H3K4 (4, 5). In contrast to EZH2, deletion of MLL2/4 results in enlarged germinal centers (4). Chromosomal translocation and other rearrangements of Mll genes in frame with multiple partner genes produce novel functional mixed lineage leukemia–fusion proteins (MLL-FPs) with oncogenic features, causing very aggressive outcomes with poor prognosis in MLL-rearranged acute lymphoblastic leukemia (49). A number of MLL-FPs recruit the H3K79 methyltransferase disruptor of telomeric silencing 1-like (Dot1L), which induces aberrant gene expression (50). However, patients carrying MLL-FPs have very few cooperating mutations, making MLL-FP–driven lymphoma ideal for animal modeling to test promising new epigenetic drugs (51, 52). Finally, the dysregulation of KDMs and acetyltransferases can also be causative of B cell malignancies and has recently been reviewed in detail elsewhere (53).
Perturbation of DNA methylation profiles has also been characterized during the development of lymphoma and leukemia. Mutations in DNMT3A expression that result in a loss of methyltransferase activity are highly linked with both T-acute lymphoblastic leukemia (54) and acute myeloid leukemia (55). B cell lymphomas exhibit DNA methylation heterogeneity. The methyltransferase DNMT1, which plays a role in germinal center B cells, is significantly increased in Burkitt lymphoma primary tumor samples compared with healthy lymph node tissues (56, 57) and in DLBCL patient cases (58, 59). Furthermore, MLL-rearranged acute lymphoblastic leukemias also exhibit loss of DNA methylation at enhancers of critical transcription factors (60), highlighting the importance of understanding the multiple layers of epigenetic regulation in lymphoma and leukemia formation and in developing treatments that target either histone or DNA methylation (recently reviewed in Ref. 61).
The progress in understanding lymphoma pathogenesis has led to new treatments of these diseases (Table I). The current treatment for B cell–driven lymphoma is based on highly toxic regimens such as chemotherapy, radiation, and the use of the CD20-specific Ab rituximab (62). New epigenetic therapeutic approaches are undergoing clinical testing for the treatment of FL and germinal center B cell–DLBCL, including EZH2, fusion protein, and histone deacetylase inhibitors (53). Several EZH2-selective inhibitors have been developed in recent years. The pharmacological inhibition of EZH2 activity can effectively prevent the proliferation of the EZH2 mutant cells in FL and DLBCL, both in vitro and in xenograft mouse models, and is currently being tested in phase I/II clinical trial (63). GSK126 (64) and EPZ-6438 (65) are selective EZH2 inhibitors that can bind to the EZH2 wild-type and Y641 mutant, leading to diminished H3K27me3 level and upregulation of gene expression. Simultaneous inhibition of EZH2 and another H3K27me3 methyltransferase EZH1 is an advanced strategy for targeting accumulated H3K27me3 because EZH1 may compensate for EZH2 within PRC2 (66). Other potential strategies being proposed include targeted disruption of the EZH2–EED complex (67) and the allosteric PRC2 inhibition targeting the H3K27me3 binding pocket of EED (68). The therapeutic approaches thus far for MLL-FP–driven transformation consist in targeting the fusion partners (AF4, AF9, AF10, ENL, ELL), which are part of either the super elongation complex or the DOT1L-containing complex (50). One of the most promising approaches, currently being evaluated in clinical trials, entailed the usage of small molecules capable to inhibit DOT1L (52), with the consequent loss of aberrant H3K79 methylation on target genes critical for the transforming activity of MLL-FPs (69).
Epigenetic Target . | Drug(s) . | Disease(s) . | Phase . | Reference(s) . |
---|---|---|---|---|
EZH2 | GSK126 | FL, DLBCL | I/II | (64, 65) |
EPZ-6438 | ||||
EZH1/2 | UNC1999 | Multiple myeloma, EZH2-based cancer | I | (66) |
EZH2-EED | SAH-EZH2 | MLL–AF9 leukemia DLBCL | — | (67) |
EED226 | (68) | |||
MLL-FP | EPZ015666 (PRMT5 inhibitor) | MLL-rearranged leukemia | — | (114) |
Dot1L inhibitors | (50) |
Epigenetic Target . | Drug(s) . | Disease(s) . | Phase . | Reference(s) . |
---|---|---|---|---|
EZH2 | GSK126 | FL, DLBCL | I/II | (64, 65) |
EPZ-6438 | ||||
EZH1/2 | UNC1999 | Multiple myeloma, EZH2-based cancer | I | (66) |
EZH2-EED | SAH-EZH2 | MLL–AF9 leukemia DLBCL | — | (67) |
EED226 | (68) | |||
MLL-FP | EPZ015666 (PRMT5 inhibitor) | MLL-rearranged leukemia | — | (114) |
Dot1L inhibitors | (50) |
Pathogen-induced epigenetic changes within lymphocytes that result in ineffective immune responses
Evolution has shaped our immune system to control a multitude of diverse pathogens, tailoring specific responses depending on the microorganism, whether of viral, bacterial, or parasitic origin. The plasticity of the immune responses is in part attributable to dynamic and reversible processes that regulate distinct gene expression through the methylation and acetylation of histones and promoters. Pathogens can directly and indirectly use these mechanisms to preserve their survival, and as such can evade clearance by suppressing an effective host immune system. The following sections will provide an example each of viral, bacterial, and parasitic infectious agents that indirectly lead to ineffective clearance of the pathogen through epigenomic changes that in turn influence the microenvironment and cell behavior. We will then detail the molecular basis for lymphocyte exhaustion, the direct hijacking of host epigenetic mechanisms for pathogen survival, and the overlap between pathogen-induced epigenetic changes and tumorigenesis.
Epigenetic modulation of proinflammatory genes is induced during infection, and it is fundamental to mount effective immune responses toward them. However, sustained changes in epigenetic modifications can favor the survival of infecting pathogens (Fig. 2). Influenza virus induces a typical acute viral infection during which the host secretes a variety of cytokines and chemokines, including type I IFN genes, chemokines (CXCL14, CCL25, CXCL6), and ILs (IL-13, IL-17C, IL-4R) (70), that eventually elicit neutralizing Abs to resolve the infection. Changes in DNA methylation levels are associated with expression of these proinflammatory genes upon influenza challenge with annual epidemic strains. In contrast to a resolving immune response to seasonal influenza strains, the highly pathogenic avian H5N1 influenza virus exhibits a dramatic decrease in promoter DNA methylation levels, inducing the hypercytokinemia prominent in pandemic pathogenesis (71, 72). Another example of epigenetic regulation of proinflammatory responses has been reported during Gram-negative bacterial infections. LPS, the major molecule expressed on the outer membrane of these bacteria, induces a signaling cascade in a TLR4-dependent manner that contributes to chromatin remodeling at inflammatory genes. In particular, an elegant study showed that LPS induces IL-12 transcription by nucleosome repositioning at the IL-12 promoter, concomitant with the presence of the permissive modifications H3K4me3 and H3K36me3 (73). These two examples demonstrate how changes in the host epigenome are correlated to modulation of the pathogen-induced microenvironment. There are also a number of other pathogens, such as Leishmania, that mediate epigenomic silencing of proinflammatory genes (74, 75). Mechanistic insight into the epigenetic modifiers involved in remodeling the epigenome to tailor an immune response depending on the type of infectious pathogen may reveal new therapeutic approaches for infections that have evaded effective vaccine design or treatment.
Immune response modulation during infections. Inflammatory environments are required in mounting specific immune responses to eliminate infecting pathogens and are regulated by epigenetic modulators. Upon influenza virus infection (left side of the figure), the cells from the innate immune system trigger intracellular signals through TLR7 that drive the deregulation of specific DNMTs (DNMT1, 3A, 3B). The limited methyltransferase activity of these DNMTs allows the expression of multiple proinflammatory genes, which in turn amplify these signals and activate a specific acquired immune response against influenza viruses. Gram-negative bacterial infections produce high amount of LPS, which is recognized by the transmembrane TLR4. Intracellular signaling increases the acetylation, phosphorylation, and activating epigenetic marks (e.g., H3K4me3, H3K36me3) of distinct proinflammatory genes. Similarly, malarial Ags are recognized by different classes of TLRs. The signal cascade following the binding induces an increasing level of acetylation and methylation (H3K4me3) of target cytokines genes, regulating their expression. This forms a feedback loop with increased production of IFN-γ.
Immune response modulation during infections. Inflammatory environments are required in mounting specific immune responses to eliminate infecting pathogens and are regulated by epigenetic modulators. Upon influenza virus infection (left side of the figure), the cells from the innate immune system trigger intracellular signals through TLR7 that drive the deregulation of specific DNMTs (DNMT1, 3A, 3B). The limited methyltransferase activity of these DNMTs allows the expression of multiple proinflammatory genes, which in turn amplify these signals and activate a specific acquired immune response against influenza viruses. Gram-negative bacterial infections produce high amount of LPS, which is recognized by the transmembrane TLR4. Intracellular signaling increases the acetylation, phosphorylation, and activating epigenetic marks (e.g., H3K4me3, H3K36me3) of distinct proinflammatory genes. Similarly, malarial Ags are recognized by different classes of TLRs. The signal cascade following the binding induces an increasing level of acetylation and methylation (H3K4me3) of target cytokines genes, regulating their expression. This forms a feedback loop with increased production of IFN-γ.
One such pathogen is the malaria-causing Plasmodium parasite. Similar to viral and bacterial pathogens, Plasmodium parasites also induce a robust innate immune response in the exposed host (76, 77). The considerable production of IFN-γ during the Plasmodium falciparum infection eventually suppresses germinal center B cell response and antimalarial humoral immunity (78, 79), which may explain the persistence of long-term chronic infections in some patients (80) (Fig. 3). Freshly isolated monocytes and lymphocytes from different African ethnic groups naturally infected with P. falciparum revealed that this hyperresponsiveness is correlated with increased H3K4me3 marks at important proinflammatory promoters, such as IFN, TNF, IL6, IL18, and TGFB1 genes, compared with healthy controls (81). This altered cytokine microenvironment influences the adaptive immune response of the infected host, leading to the chronic features of this infection.
Epigenetic regulation of chronic infection. (A) During an acute response to viral infection, naive T cells respond to the proinflammatory microenvironment to become effector or memory subsets. In a chronic lymphocytic choriomeningitis infection, effector T cells acquire an exhausted phenotype through epigenetic modifications that lead to prolonged expression of genes known to be involved in the chronic infection pathogenesis, such as Pdcd1. H3K4me3 marks lead to open chromatin regions. (B) In parasitic infection, germinal centers are disrupted by prolonged exposure to IFN-γ produced by Th cells. The downstream signal cascade induces the upregulation of T-bet, which in turn represses the expression of the transcription factor Bcl6, critical for germinal center reactions. As a result of germinal center impairment, Ab class-switching and affinity maturation are suppressed, which is correlated to a less effective response to parasitic infection.
Epigenetic regulation of chronic infection. (A) During an acute response to viral infection, naive T cells respond to the proinflammatory microenvironment to become effector or memory subsets. In a chronic lymphocytic choriomeningitis infection, effector T cells acquire an exhausted phenotype through epigenetic modifications that lead to prolonged expression of genes known to be involved in the chronic infection pathogenesis, such as Pdcd1. H3K4me3 marks lead to open chromatin regions. (B) In parasitic infection, germinal centers are disrupted by prolonged exposure to IFN-γ produced by Th cells. The downstream signal cascade induces the upregulation of T-bet, which in turn represses the expression of the transcription factor Bcl6, critical for germinal center reactions. As a result of germinal center impairment, Ab class-switching and affinity maturation are suppressed, which is correlated to a less effective response to parasitic infection.
Antigenic persistence during chronic infections can ultimately result in T and B cell exhaustion (82, 83). Exhausted T cells have been studied in detail, as they are critical therapeutic targets for treating chronic viral infections and for cancer treatment. Inhibitory receptors are upregulated on exhausted T cells (84). During infection, the upregulation of these receptors correlates to inhibition of T cell function that would normally help clear the virus (85). Tumors bind to these receptors on T cells to inhibit their function and in essence protect tumor cells from deletion. Inhibitory receptor blockade to reinvigorate T cells has therefore been a major step forward in clinical interventions. However, the response to blockade therapy has been variable, and thus, research has focused on understanding the molecular basis for this obstacle and whether exhaustion can be reversed, potentially through targeting specific epigenetic modifiers (86–89). Analysis of chromatin accessibility using Assay for Transposase-Accessible Chromatin using sequencing profiles has revealed that almost 50% of chromatin-accessible regions were differently present in exhausted CD8+ T cells during chronic lymphocytic choriomeningitis infection or in tumor models compared with functional CD8+ T cells (87, 88, 90, 91) (Fig. 3). These data also showed an increased chromatin accessibility at the enhancer site of the Pdcd1 gene (encoding for programmed cell death protein 1; PD-1), the most well-established surface receptor promoting this state of impaired effector functions (90). Exhaustion-specific epigenetic profiles in the mouse are conserved in Ag-specific exhausted human T cells in HIV-1 infection (91, 92). Recent data suggest that there are two distinct stages during the formation of exhausted cells. In one, there is little molecular change, and thus the cells can still revert back to an effector stage. However, the second stage results in epigenetic changes that render the T cells in a fixed state, in that checkpoint blockade may initially reactivate the cells but their chromatin landscape remains the same, and hence withdrawal of treatment leads to reversal of T cells back to an exhausted phenotype. Thus, researchers are targeting the initial stages of exhaustion (86, 88, 93); for instance, blocking DNA methylation has recently been shown to be a promising complementary intervention to checkpoint blockade (86). Much less is known about whether atypical memory B cells, a population of cells prominent in HIV and malaria patients (94, 95) that have been suggested to show an exhaustion phenotype (94), also have changes in their chromatin structure compared with classical memory B cells. Given the unknown origin and function of atypical memory B cells, determining the epigenetic landscape may lead to new insights into how this population is formed and whether it has a positive or detrimental role in the humoral immune response. In summary, determining how epigenetic alterations are induced by chronic Ag exposure, and in particular the enzymes that mediate these modifications, will open up new ways of therapeutic intervention to prevent their exacerbation.
Finally, pathogens can also evade immunosurveillance by directly co-opting epigenetic mechanisms. For example, the carboxy terminal of the nonstructural protein 1 (NS1) in the influenza A strain H3N2 is crucial for counteracting the antiviral response by mimicking the histone H3 tail. The human PAF1 complex is required for the efficient transcription of genes, including those involved in the antiviral response against influenza. NS1 is able to mimic the H3 tail, hijacking the PAF1 complex by directly binding it and thus inducing the suppression of hPAF1C-mediated transcriptional elongation. This interaction subsequently counteracts the binding of RNA polymerase II to its target genes, including numerous antiviral and proinflammatory genes (96, 97). This eventually leads to the deregulation of acquired immune responses, prolonging the viral infection within the host (98). Viruses such as EBV and HIV prevent recognition by host immunosurveillance by remaining latent inside host cells through epigenetic modification of their genome to mimic their host genome (99, 100). The latency membrane protein 1 establishes EBV latency and induces tumorigenesis by promoting epigenetic alterations of histones and resulting in the chromatin accessibility of two established latency membrane protein 1–target proto-oncogenes, c-Fos and EGR1 (101). The overexpression of these oncogenes eventually results in the development of tumors associated with EBV infection such as Burkitt lymphoma, nasopharyngeal carcinoma, and some gastric cancer (102–104). Revealing these overlapping mechanisms demonstrated how some epigenetic alterations are shared in chronic infection and in tumorigenesis. A deeper understanding of these processes potentially represents a fundamental milestone for innovative treatments for these diseases, which still lack of an appropriate cure.
Using epigenetic modifiers to fight viral, bacterial, and parasitic infections
Emerging classes of pharmacological agents targeting epigenetic modifiers are now being used as a new line of attack against difficult to clear viral, bacterial, and parasitic infectious agents (Table II). For HIV patients, the current antiviral therapy (highly active antiretroviral therapy) does not eradicate latent reservoirs of the virus within memory CD4+ T cells (105). HIV-1 latency is finely regulated by the epigenetic landscape of the provirus through deacetylation and methylation of histones (106, 107). Thus, administering highly active antiretroviral therapy with epigenetic inhibitors (listed in Table II) underpins a new promising strategy to fight HIV infection, termed “shock and kill” therapy. The therapy manipulates histone posttranslational modifications to reactivate the provirus, followed by the killing of infected cells with conventional antiviral therapy. Inhibition of epigenetic modifiers also opened up new frontiers in the treatment of other viral infections, such as EBV and influenza (Table II). Recent studies on the epigenetics of plasmodium parasites (108) have also created new opportunities for therapeutics. For instance, inhibitors of histone deacetylases and histone lysine methyltransferases (Table II) have been tested in different models of malarial infection [reviewed in (108)] and shown to interfere with parasite growth and survival. Thus, investigation into epigenetic modifiers is critical for not only understanding disease pathogenesis but also to develop new therapeutic strategies for multiple different infectious diseases.
Organism . | Category . | Drug(s) . | Target(s) . | Reference(s) . |
---|---|---|---|---|
HIV | HKMTi | GSK-343, EPZ-6438 | EZH2/EED | (115) |
DNMTi | Phytoalexin resveratrol | DNMT1 | (116) | |
PRMTi | Aryl pyrazoles | PRMT6 | (117) | |
EBV | HDACi | 5-Aza | BZLF1 | (118) |
3-deazaneplanocin A | ||||
Influenza | HDACi | Ortho-aryl | HDAC8 | (119) |
N-hydroxycinnamides | (120) | |||
P. falciparum | HDACi | Apicidin | Class I/II HDACs | (121) |
HKMTi | BIX-01294 | G9a | (122) | |
TM2-115 |
Organism . | Category . | Drug(s) . | Target(s) . | Reference(s) . |
---|---|---|---|---|
HIV | HKMTi | GSK-343, EPZ-6438 | EZH2/EED | (115) |
DNMTi | Phytoalexin resveratrol | DNMT1 | (116) | |
PRMTi | Aryl pyrazoles | PRMT6 | (117) | |
EBV | HDACi | 5-Aza | BZLF1 | (118) |
3-deazaneplanocin A | ||||
Influenza | HDACi | Ortho-aryl | HDAC8 | (119) |
N-hydroxycinnamides | (120) | |||
P. falciparum | HDACi | Apicidin | Class I/II HDACs | (121) |
HKMTi | BIX-01294 | G9a | (122) | |
TM2-115 |
DNMTi, DNMT inhibitor; HDACi, histone deacetylase inhibitor; HKMTi, histone lysine methyltransferase inhibitor; PRMTi, protein arginine methyltransferase inhibitor.
Conclusions
Epigenetic regulation is important for normal immune cell function, and its dysregulation is correlated to disordered lymphocyte function in disease. Histone modifiers, including PRCs and MLL proteins, are fundamental for appropriate regulation of gene expression programs that shape immune responses to different pathogens or malignancies. Advanced genomic sequencing is unveiling how bivalency plays a central role in immune responses. Indeed, proinflammatory genes have both permissive and repressive marks (H3K4me3 and H3K27me3, respectively) which allow them to be poised for expression during an immune response. However, these modifications can also alter the immune response to favor the survival of the pathogen over the ability of the host immune system in mounting an effective response that will clear the invading agents. Thus, revealing the connection between different types of environmental signals, and the dynamic modulation of complex members that result in changes in the epigenome, would be a strong avenue for the field to pursue.
Many of the intrinsic mechanisms regulating immune responses during lymphoid malignancies and chronic infections discussed in the review show a high degree of similarity. The epigenetic regulation of PD-1 is a crucial example because its expression is upregulated in both cancer and chronic infection causing the T cell exhaustion phenotype. Ab blockade of PD1 activity reduces immunosuppression, resulting in induction of tumor regression and decrease in HIV/SIV load. Similarly, the proinflammatory environment in certain autoimmune disorders is dysregulated because of the chronic presentation of self-antigens. Patients with systemic lupus erythematosus and Sjögren syndrome display increased levels of type I IFN–induced genes as a consequence of the hypomethylated regions in the type I IFN cluster (109), and epigenetic regulators have been linked with autoinflammatory feedback loops (110), site-specific super-enhancer activity (111), and hyperactivity of lymphocyte responses due to modulation of H3K27me3 (112). Of interest, previous studies have shown a consistent risk of B-NHL development among patients with Sjögren syndrome, facilitated by the high and long-standing inflammatory environment in these patients (113). Therefore, similar epigenetic mechanisms may underpin disruption of immune function and signaling across a number of immune-related diseases. For this reason, future studies should be focused in unveiling further overlapping mechanisms and in defining the intracellular targets of these pathways, which may provide a potential mutual strategy to treat these diseases.
The reversible editing of the epigenome during an immune response, and thereby a specific approach targeting cell function by regulating gene expression without altering the DNA, represents an exciting field of investigation, paving the way for novel potential therapeutic interventions. Hundreds of molecules with inhibiting activity have been tested on dysregulated or mutated histone modifiers and DNMTs, showing remarkable success rates in the relapse of some tumors, with few of them currently enlisted in clinical trials. Despite the great advances in this field, further studies on each subunit of the histone modifier complexes in disease need to be done. Although EZH2 of the PRC2 has been extensively investigated in lymphoma models, other subunits of the PRC2, as well as the PRC1 complex, lack detailed information. Advances in knowledge of the causative links between the microenvironment, chromatin remodeling, and immune function may lead to major breakthroughs in the design of therapeutic approaches for infectious diseases currently without effective vaccines and in the prevention of relapse in cancer patients.
Footnotes
This work is supported by National Health and Medical Research Council Career Development Fellowship 1108066 (to K.L.G.-J.).
Abbreviations used in this article:
- AID
activation-induced cytidine deaminase
- B-NHL
B cell non-Hodgkin lymphoma
- DLBCL
diffuse large B cell lymphoma
- DNMT
DNA methyltransferase
- DotT1L
disruptor of telomeric silencing 1-like
- FL
follicular lymphoma
- KDM
lysine demethylase
- MLL-FP
mixed lineage leukemia–fusion protein.
References
Disclosures
The authors have no financial conflicts of interest.