What is Dr. Zubair Khalid's research focus?

Dr. Zubair Khalid specializes in molecular virology, mRNA vaccine development, and computational biology, with a focus on avian pathogens like IBDV and Avian Reovirus.

Where is Dr. Zubair Khalid currently working?

Dr. Zubair Khalid is a Postdoctoral Research Associate at the University of Maryland (UMD), specifically within the Department of Animal and Avian Sciences.

Can Epigenetics Be Reversed? Mechanisms and Evidence

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Introduction to Epigenetics and Reversibility

Epigenetics refers to heritable, conditionally reversible changes in gene expression that occur without alterations to the underlying DNA sequence. The term encompasses a suite of molecular modifications—DNA methylation, post-translational histone modifications, chromatin remodeling, and non-coding RNA-mediated regulation—that collectively determine which genes are active in a given cell type. Unlike genetic mutations, which permanently alter the nucleotide sequence, epigenetic marks are dynamic by design. They must be established, maintained, and, crucially, removed in response to developmental cues and environmental signals.

The question of whether epigenetics can be reversed is therefore not merely academic. It sits at the intersection of developmental biology, regenerative medicine, and oncology. The short answer is yes: epigenetic marks can be actively reversed by dedicated enzymatic machinery, and this reversibility is exploited both naturally—during embryogenesis and cellular differentiation—and therapeutically, as in the treatment of hematological malignancies. However, the ease and completeness of reversal vary dramatically depending on the type of mark, the genomic context, and the differentiation state of the cell. Some marks are labile and turn over rapidly; others are remarkably stable and resist removal even under strong selective pressure.

To understand reversibility, one must first appreciate that epigenetic information is encoded biochemically. A methyl group attached to the fifth carbon of cytosine (5-methylcytosine, 5mC) in a CpG dinucleotide can be read by methyl-binding proteins that recruit repressive complexes. Histone tails protruding from the nucleosome core carry acetyl, methyl, phosphoryl, and ubiquitin groups that alter chromatin compaction and recruit effector proteins. These modifications are not permanent tattoos; they are chemical equilibria maintained by opposing enzyme families. When the writers (enzymes that add marks) are inhibited or the erasers (enzymes that remove marks) are activated, the equilibrium shifts, and the mark is lost. This fundamental biochemical principle underpins all epigenetic reversal. For a broader introduction to the field, see Epigenetics Explained.

Key Epigenetic Mechanisms and Their Reversibility

DNA Methylation and Demethylation

DNA methylation occurs predominantly at cytosine residues within CpG dinucleotides. The reaction is catalyzed by DNA methyltransferases (DNMTs): DNMT3A and DNMT3B establish new methylation patterns during development, while DNMT1 maintains existing patterns during DNA replication by copying methylation from the parental strand to the daughter strand. The product, 5mC, is recognized by methyl-CpG-binding domain (MBD) proteins such as MeCP2, which recruit histone deacetylases and other repressive factors to silence transcription.

For years, DNA methylation was considered the most stable epigenetic mark—and in many contexts, it is. Methylation at imprinted loci and on the inactive X chromosome persists for the lifetime of the organism. Yet even this "permanent" mark can be reversed. Passive demethylation occurs when DNMT1 is absent or inhibited during replication: the newly synthesized strand remains unmethylated, and after successive rounds of division, the mark is diluted out. Active demethylation, by contrast, is an enzyme-driven process that does not require cell division. The ten-eleven translocation (TET) family of enzymes—TET1, TET2, and TET3—oxidize 5mC to 5-hydroxymethylcytosine (5hmC), then to 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC). These oxidized derivatives are recognized by thymine DNA glycosylase (TDG), which excises the modified base, initiating base excision repair (BER) that restores an unmodified cytosine. This pathway is the primary route for active DNA demethylation in mammals.

Histone Modifications and Their Erasers

Histone modifications are more labile than DNA methylation by several orders of magnitude. The N-terminal tails of histones H3 and H4 protrude from the nucleosome and are subject to a dizzying array of post-translational modifications. Acetylation of lysine residues (e.g., H3K9ac, H3K27ac) neutralizes the positive charge of the histone tail, weakening its interaction with negatively charged DNA and promoting an open, transcriptionally permissive chromatin state. Methylation of lysines can be either activating or repressive depending on the residue and the degree of methylation: H3K4me3 marks active promoters, H3K36me3 marks the bodies of actively transcribed genes, and H3K27me3 and H3K9me3 are associated with facultative and constitutive heterochromatin, respectively.

Each modification has a dedicated enzymatic eraser. Histone deacetylases (HDACs) remove acetyl groups, restoring the positive charge and promoting chromatin compaction. The HDAC family comprises four classes: class I (HDAC1, 2, 3, 8), class II (HDAC4, 5, 6, 7, 9, 10), class III (sirtuins SIRT1–7, which require NAD+ as a cofactor), and class IV (HDAC11). Histone demethylases fall into two families: the lysine-specific demethylases (LSD1/KDM1A and LSD2/KDM1B), which remove mono- and dimethyl groups from H3K4 and H3K9 via a flavin adenine dinucleotide (FAD)-dependent oxidation, and the Jumonji C (JmjC) domain-containing demethylases (e.g., KDM2–KDM6 families), which use α-ketoglutarate and iron as cofactors and can remove all three methylation states. The existence of these erasers means that histone modifications are continuously turned over, with typical half-lives ranging from minutes to hours.

Chromatin Remodeling and Plasticity

Chromatin remodeling refers to the ATP-dependent movement or eviction of nucleosomes by SWI/SNF, ISWI, CHD, and INO80 family complexes. These remodelers slide histone octamers along DNA, expose or occlude transcription factor binding sites, and exchange histone variants (e.g., replacing H3 with H3.3 or H2A with H2A.Z). Unlike covalent modifications, remodeling is not a chemical mark per se, but it determines the accessibility of DNA to the transcriptional machinery. Because remodelers are recruited by sequence-specific transcription factors and by the modification state of histones, they are inherently reversible: when the recruiting signal disappears, the remodeler dissociates, and the nucleosome can reassume its default position. This plasticity is what allows a differentiated cell to change its transcriptional program in response to a new signal, and it is the basis for cellular reprogramming.

Enzymatic Players in Epigenetic Reversal

TET Enzymes and Active Demethylation

The TET enzymes are the principal mediators of active DNA demethylation. TET1 is enriched at CpG islands and promoters of actively transcribed genes, where it maintains a low-methylation state. TET2 is frequently mutated in myeloid malignancies, and its loss leads to hypermethylation and blocked differentiation. TET3 is highly expressed in oocytes and zygotes, where it mediates the widespread demethylation of the paternal genome shortly after fertilization.

The catalytic mechanism is a stepwise oxidation. TET enzymes use molecular oxygen, Fe(II), and α-ketoglutarate to oxidize the methyl group of 5mC, generating 5hmC. This first oxidation product is stable and can itself be read by specific proteins, but it can be further oxidized to 5fC and 5caC. The latter two are recognized by TDG, which cleaves the N-glycosidic bond, creating an abasic site. The base excision repair pathway then inserts an unmodified cytosine. The entire process—from 5mC to unmodified C—can occur within hours in actively demethylating cells. Importantly, 5hmC is not merely an intermediate; it is a stable mark in its own right, particularly in embryonic stem cells and neurons, where it may represent a distinct regulatory state.

Histone Deacetylases and Demethylases

The HDAC family removes acetyl groups from lysine residues on histones and numerous non-histone proteins. Class I HDACs are predominantly nuclear, ubiquitously expressed, and form the catalytic cores of the Sin3, NuRD, and CoREST complexes. Class II HDACs shuttle between nucleus and cytoplasm and show tissue-specific expression. Sirtuins (class III) are NAD+-dependent and link epigenetic regulation to cellular metabolism; SIRT1, for example, deacetylates H3K9ac and H3K14ac at promoters silenced during cellular stress.

Histone demethylases provide the counterpoint to histone methyltransferases. LSD1 removes H3K4me1/me2 but, when associated with the androgen receptor, can also demethylate H3K9me1/me2, acting as a coactivator. The JmjC demethylases are more versatile: KDM5A–D remove H3K4me2/me3, KDM6A/B remove H3K27me2/me3, and KDM4A–D remove H3K9me2/me3 and H3K36me2/me3. These enzymes are often components of large complexes that also contain chromatin remodelers and transcription factors, allowing coordinated reversal of multiple repressive marks at a given locus.

Evidence for Epigenetic Reversal in Development and Disease

Reprogramming in Early Embryo

The most dramatic natural example of epigenetic reversal occurs immediately after fertilization. The sperm genome arrives highly compacted, with protamines replacing most histones and dense DNA methylation at CpG islands. Within hours of fertilization, the paternal pronucleus undergoes active demethylation mediated by TET3, which oxidizes 5mC to 5hmC before the first cell division. The maternal genome, by contrast, is demethylated passively over subsequent cleavage divisions, as DNMT1 is excluded from the nucleus and methylation is diluted. By the blastocyst stage, global methylation levels have dropped from roughly 70–80% to approximately 20–30%. This erasure is essential for establishing totipotency—the ability of embryonic cells to give rise to all cell types—and for resetting imprinted genes in the germline.

Histone modifications are similarly reset. The paternal genome rapidly exchanges protamines for histones, and both parental genomes undergo changes in H3K4me3 and H3K27me3 that establish new regulatory domains. This wave of reprogramming demonstrates that even the most stable epigenetic marks can be erased when the cellular machinery is appropriately activated.

Cellular Reprogramming to Pluripotency

Somatic cells can be converted to induced pluripotent stem cells (iPSCs) by forced expression of the transcription factors OCT4, SOX2, KLF4, and MYC (the Yamanaka factors). This process requires extensive epigenetic remodeling: repressive marks at pluripotency genes (e.g., OCT4, NANOG) must be removed, and activating marks must be established. DNA methylation at these loci is lost through a combination of passive dilution and TET-mediated active demethylation. Histone H3K27me3 at silenced developmental genes is removed by KDM6A/B, while H3K4me3 is deposited at newly activated promoters. The efficiency of reprogramming is low—typically 0.1–1% of starting cells—but the fact that it occurs at all is direct evidence that the epigenetic state of a differentiated cell is not fixed. The process is stochastic and takes 1–2 weeks, with epigenetic changes preceding stable gene expression changes.

Epigenetic Changes in Cancer and Their Reversal

Cancer cells display profound epigenetic abnormalities: global hypomethylation (particularly at repetitive elements and oncogene promoters) coexists with focal hypermethylation at tumor suppressor gene promoters. For example, the CDKN2A locus (encoding p16INK4A) is silenced by promoter hypermethylation in many cancers, and MLH1 is hypermethylated in a subset of colorectal cancers with microsatellite instability. These changes are not irreversible. Treatment with nucleoside analog inhibitors of DNMTs, such as 5-azacitidine and decitabine, leads to passive demethylation and re-expression of silenced tumor suppressors. These drugs are incorporated into DNA during replication, where they covalently trap DNMT1, depleting the enzyme and causing genome-wide loss of methylation. Clinical responses in myelodysplastic syndromes and acute myeloid leukemia demonstrate that reversing aberrant methylation can restore normal differentiation and produce durable remissions. For a deeper discussion of how epigenetic marks influence human biology, see Epigenetics in Humans.

Methods Used to Study Epigenetic Reversal

CRISPR and Epigenome Editing

The fusion of catalytically dead Cas9 (dCas9) to epigenetic effector domains allows targeted reversal of marks at specific loci. dCas9-TET1 can demethylate a chosen promoter, while dCas9 fused to the catalytic domain of the histone demethylase LSD1 or the histone acetyltransferase p300 can alter histone modifications at defined regions. These tools have been used to demonstrate causality: demethylating the promoter of the fetal hemoglobin gene (HBG1/2) in adult erythroid cells reactivates its expression, providing a potential therapeutic strategy for sickle cell disease and β-thalassemia. Similarly, dCas9-p300-mediated acetylation at enhancers can activate otherwise silent genes. The specificity of these approaches depends on the guide RNA sequence and on the local chromatin context; not all loci are equally accessible to the editing machinery.

Pharmacological Inhibitors

Small-molecule inhibitors of epigenetic enzymes are the most widely used tools for studying reversal. 5-Azacitidine and decitabine inhibit DNMTs at micromolar concentrations (typically 0.1–10 µM in cell culture). HDAC inhibitors such as trichostatin A (TSA), suberoylanilide hydroxamic acid (SAHA/vorinostat), and sodium butyrate inhibit class I and II HDACs at nanomolar to millimolar concentrations, respectively. These drugs cause global changes in acetylation and can reactivate silenced genes, but they lack locus specificity. The JmjC demethylase inhibitor GSK-J1 selectively blocks H3K27me3 demethylation, and the LSD1 inhibitor tranylcypromine (an antidepressant) is being repurposed for cancer therapy. When using these drugs experimentally, it is critical to determine the optimal concentration and exposure time for the cell type in question, as prolonged treatment can cause cytotoxicity and off-target effects.

Reprogramming Assays

Cellular reprogramming assays are functional tests of epigenetic reversibility. The gold standard is the generation of iPSCs from somatic cells, as described above. A more rapid assay uses cell fusion: fusing a somatic cell with an embryonic stem cell leads to reprogramming of the somatic nucleus within 24–48 hours, as judged by reactivation of pluripotency markers. Another approach is transdifferentiation, where fibroblasts are directly converted to neurons, hepatocytes, or cardiomyocytes by expression of lineage-specific transcription factors. These assays measure not just the removal of a single mark but the coordinated reversal of an entire epigenetic program. They are technically demanding and require careful validation, including bisulfite sequencing for DNA methylation, chromatin immunoprecipitation (ChIP) for histone modifications, and RNA-seq for gene expression.

Therapeutic Implications of Reversing Epigenetics

The clinical success of epigenetic drugs has validated the concept that reversing aberrant marks can be therapeutic. Azacitidine and decitabine are approved for myelodysplastic syndromes and acute myeloid leukemia, where they improve survival and delay leukemic progression. HDAC inhibitors, including vorinostat and romidepsin, are approved for cutaneous T-cell lymphoma. These drugs are thought to work by reactivating silenced tumor suppressors, inducing differentiation, and promoting apoptosis. Combination regimens—for example, a DNMT inhibitor plus an HDAC inhibitor—show synergistic activity in preclinical models, likely because demethylation exposes promoters that then require histone acetylation for full activation.

Beyond oncology, epigenetic reversal is being explored in neurological and psychiatric disorders. Valproic acid, a weak HDAC inhibitor, has been used for decades as a mood stabilizer. In animal models, HDAC inhibitors can reverse learning deficits associated with environmental enrichment or pharmacological treatment, and they have been shown to reactivate genes silenced by early-life stress. The field of Epigenetics Psychology examines how such environmental exposures shape the epigenome and whether these changes can be therapeutically reversed.

The most promising emerging application is in hemoglobinopathies. Pharmacological inhibition of the epigenetic silencing of fetal hemoglobin—using agents that disrupt the interaction between the repressor BCL11A and its genomic targets—can increase HbF levels and ameliorate disease severity. CRISPR-based epigenome editing to demethylate the HBG promoters is in preclinical development. These approaches illustrate a general principle: if a disease is caused by an aberrant epigenetic mark, removing that mark may restore normal function.

Limitations and Challenges in Reversing Epigenetics

Despite the proof of principle, several obstacles limit the reversal of epigenetic marks. First, not all marks are equally reversible. DNA methylation at heterochromatic regions, particularly at pericentromeric satellites and imprinted loci, is highly stable. The maintenance methyltransferase DNMT1 is processive and efficient, and the heterochromatic environment may exclude TET enzymes. Even in iPSCs, some loci retain "epigenetic memory" of their tissue of origin, with residual methylation at genes that were silenced in the donor cell.

Second, reversal of one mark may be insufficient if other marks are redundant. A promoter silenced by both DNA methylation and H3K27me3 may require removal of both marks for reactivation. The order of removal matters: demethylation before histone mark removal may be ineffective if the locus remains in a repressive chromatin state.

Third, global reversal is not the same as targeted reversal. Pharmacological inhibitors affect all loci, causing widespread gene dysregulation. Treatment with 5-azacitidine reactivates not only tumor suppressors but also oncogenes, transposable elements, and endogenous retroviruses. The resulting genomic instability and activation of innate immune sensing pathways can be beneficial (as in the viral mimicry response) or detrimental.

Fourth, the kinetics of reversal are slow. DNA demethylation via TET and BER takes hours to days, and full reprogramming of a somatic cell takes weeks. During this window, cells may be vulnerable to apoptosis or transformation. Finally, the epigenome is not the only determinant of cell state. Transcription factor networks, signaling pathways, and the three-dimensional organization of the genome in topologically associating domains (TADs) all contribute to gene regulation. Reversing a single mark may not be sufficient to overcome a stable transcriptional program.

Common Pitfalls and Misconceptions

Students often make several errors when reasoning about epigenetic reversal. The most common is assuming that all epigenetic changes are reversible with equal ease. This is false. DNA methylation at CpG islands in promoters is more readily reversed than methylation in gene bodies or at repeat elements. Histone modifications are generally more dynamic than DNA methylation, but H3K9me3 at pericentric heterochromatin is exceptionally stable.

A second misconception is that reversing an epigenetic mark always restores a healthy state. In cancer, demethylating agents can reactivate oncogenes or cause genomic instability. In development, erasing imprints would be catastrophic. Reversal is a tool, not a panacea.

A third error is conflating "reversible" with "reversed in every cell." Epigenetic reversal is often stochastic and incomplete. Even after successful iPSC reprogramming, residual methylation at some loci persists. In a population of cells, some may fully reverse a mark while others do not, leading to heterogeneous outcomes.

A fourth pitfall is ignoring the distinction between active and passive demethylation. Passive demethylation requires DNA replication and is therefore only relevant in dividing cells. Post-mitotic neurons, for example, cannot passively dilute methylation; they rely exclusively on TET-mediated active demethylation. Experimental designs must account for the proliferative state of the cells under study.

Finally, students sometimes assume that epigenetic marks are the cause rather than the consequence of gene expression changes. In many cases, transcription factor binding precedes and directs epigenetic remodeling. Reversing the mark without addressing the upstream regulator may not produce a stable change in gene expression. For a discussion of how epigenetic changes arise in response to environmental stimuli, see Change Epigenetics.

Summary and Practical Takeaways

Epigenetics can be reversed, but the extent and ease of reversal depend on the specific mark, the genomic location, and the cellular context. DNA methylation is reversed by TET-mediated oxidation followed by base excision repair, or passively by failure of maintenance methylation during replication. Histone acetylation is reversed by HDACs, and histone methylation by LSD and JmjC demethylases. Chromatin remodeling is inherently reversible through ATP-dependent remodelers. Natural examples of reversal include zygotic reprogramming, iPSC generation, and the response of cancer cells to demethylating agents. Therapeutically, DNMT and HDAC inhibitors are approved drugs, and targeted epigenome editing is advancing rapidly. However, not all marks are equally reversible, and global reversal carries risks of off-target effects and genomic instability.

For students preparing for exams, the key points to remember are: (1) the names and mechanisms of the eraser enzymes; (2) the distinction between active and passive demethylation; (3) the evidence for reversal in development and reprogramming; (4) the clinical applications of epigenetic drugs; and (5) the limitations that prevent universal, complete reversal. Understanding these principles will allow you to evaluate new claims about epigenetic therapies critically. The question "can epigenetics be reversed?" has a nuanced answer: yes, but with qualifications. The field of Epigenetics Important continues to refine our understanding of when and how reversal is possible.

Frequently Asked Questions

Can epigenetics be reversed?

Yes. Epigenetic marks are enzymatically reversible. DNA methylation can be removed by TET enzymes through active demethylation or diluted passively during replication. Histone acetylation is removed by HDACs, and histone methylation by LSD and JmjC demethylases. Natural examples include embryonic reprogramming and iPSC generation, and pharmacological agents that inhibit epigenetic writers are used clinically to reverse aberrant marks in cancer.

How can epigenetic changes be reversed?

Epigenetic changes are reversed by three general mechanisms: (1) active enzymatic removal, as when TET enzymes oxidize 5mC or HDACs remove acetyl groups; (2) passive dilution, where a mark is not maintained during DNA replication and is progressively lost over cell divisions; and (3) replacement, where modified histones are exchanged for unmodified variants by chromatin remodelers. Experimentally, reversal can be induced with pharmacological inhibitors of writers or with targeted epigenome editing using dCas9 fusion proteins.

Are all epigenetic modifications reversible?

No. While most modifications have dedicated eraser enzymes, some marks are exceptionally stable. DNA methylation at imprinted loci and at pericentromeric heterochromatin persists for the lifetime of the organism. H3K9me3 at constitutive heterochromatin is also highly stable. The reversibility of a mark depends on the accessibility of the locus to eraser enzymes, the presence of protective binding proteins, and the proliferative state of the cell.

Can lifestyle changes reverse epigenetic changes?

Evidence from human and animal studies suggests that diet, exercise, and stress reduction can alter the epigenome, but the effects are generally modest and locus-specific. For example, physical exercise has been associated with changes in DNA methylation at genes involved in metabolism and inflammation. However, lifestyle interventions do not produce the global, rapid reversal seen with pharmacological agents. The field is young, and causal relationships between specific lifestyle factors and specific epigenetic changes remain difficult to establish. For more on this topic, see Epigenetics Trauma for a discussion of stress-related epigenetic changes.

What enzymes reverse DNA methylation?

The TET family—TET1, TET2, and TET3—catalyze the oxidation of 5mC to 5hmC, 5fC, and 5caC. These oxidized bases are then excised by thymine DNA glycosylase (TDG), and the resulting abasic site is repaired by the base excision repair pathway to restore unmodified cytosine. Passive demethylation occurs when DNMT1 is absent or inhibited, allowing methylation to be lost during DNA replication.

Can epigenetic changes be inherited?

Yes. Epigenetic marks can be transmitted from parent to offspring, both through mitotic inheritance (from a cell to its daughter cells) and, in some cases, through meiotic inheritance across generations. Imprinted genes are classic examples of transgenerational epigenetic inheritance. However, most epigenetic marks are reset during gametogenesis and early embryogenesis, so the extent of transgenerational inheritance in mammals remains an active area of research. See Epigenetics Inherited for a detailed treatment.

Is epigenetic reversal a potential cancer treatment?

Yes, and it is already in clinical use. The DNMT inhibitors azacitidine and decitabine are approved for myelodysplastic syndromes and acute myeloid leukemia. HDAC inhibitors such as vorinostat and romidepsin are approved for cutaneous T-cell lymphoma. These drugs reverse aberrant silencing of tumor suppressor genes and induce differentiation or apoptosis of cancer cells. However, they are not curative in most patients, and resistance develops. Combination strategies and targeted epigenome editing are being developed to improve efficacy and reduce off-target effects.

Key Takeaways

Epigenetic marks are biochemically reversible through dedicated enzymatic pathways, including TET-mediated DNA demethylation and HDAC-mediated histone deacetylation.
Active DNA demethylation proceeds through oxidation of 5mC to 5hmC, 5fC, and 5caC, followed by TDG-mediated base excision and BER.
Passive demethylation occurs when DNMT1 maintenance methylation fails during DNA replication, requiring cell division.
Histone modifications are generally more dynamic than DNA methylation, with erasers such as HDACs and JmjC demethylases acting on timescales of minutes to hours.
Natural epigenetic reversal occurs during zygotic reprogramming, iPSC generation, and in response to cellular differentiation signals.
Pharmacological inhibitors of DNMTs and HDACs are approved cancer therapies, demonstrating clinical utility of epigenetic reversal.
Not all marks are equally reversible; imprinted loci, heterochromatin, and some CpG islands resist demethylation, and global reversal carries risks of off-target gene activation and genomic instability.