Polycomb repression operates through two cooperating complexes. PRC1 decorates histone H2A with a single ubiquitin molecule at lysine 119, producing the H2Aub mark, while PRC2 trimethylates histone H3 at lysine 27, producing H3K27me3. Together these marks silence genes that would otherwise trigger premature differentiation. In mouse embryos, previous work had shown that H2Aub and H3K27me3 can be decoupled during preimplantation development, and that maternal H3K27me3 can guide certain imprinted expression patterns. But human embryos posed a puzzle: H3K27me3 is largely erased across the genome after fertilization and only later restored. If the repressive mark disappears, how does the cell know where to put it back?

To answer that question, the team needed to measure H2Aub in samples that are extraordinarily scarce, human sperm, fully grown oocytes, two-cell embryos, eight-cell embryos, blastocysts and embryonic stem cells. They adapted a targeted profiling approach called Stacc-seq, an in situ method derived from CUT&Tag that can map chromatin marks from as few as a handful of cells. The authors validated the technique extensively, showing that profiles from 500, 50 or even 20 cultured cells closely matched conventional ChIP-seq, and confirming in engineered mouse and human stem cells that the signal vanished when PRC1 was acutely degraded. Immunofluorescence on human sperm independently confirmed that H2Aub is present in the paternal genome.

The resulting atlas shows that H2Aub does not simply follow H3K27me3. Instead, it emerges remarkably early, appearing in GC-rich regions of the genome as soon as the two-cell stage, right around the time the embryonic genome switches on. From there it progressively accumulates at both active and inactive genes, and by the blastocyst stage it concentrates strongly at canonical Polycomb target genes, the developmental regulators that must stay silent until the embryo commits to its first lineages. Notably, H2Aub in humans tracks GC content and CpG density far more closely than it does in mice, where the mark is largely confined to inactive genes, revealing a species difference in how Polycomb complexes are recruited to chromatin.

The relationship between the two Polycomb marks changes across development. In oocytes and two-cell embryos, H3K27me3 is only weakly correlated with H2Aub, consistent with the genome-wide erasure of the H3K27me3 landscape after fertilization. By the blastocyst stage, however, the two marks colocalize precisely at inactive Polycomb target genes that lack the active mark H3K4me3. This suggested a causal hypothesis: rather than inheriting H3K27me3 from the gametes, human embryos may rebuild it on a scaffold laid down by PRC1 through H2Aub.

To test that idea, the researchers removed H2Aub in human embryos by overexpressing PR-DUB, the BAP1-containing complex that erases the ubiquitin mark, using microinjection of three-pronuclear zygotes donated through IVF programs. Single-embryo RNA sequencing at day 3 and day 4 showed that losing H2Aub impaired embryonic development and caused moderate derepression of Polycomb target genes, particularly those that are normally silent. More striking was what happened to H3K27me3: its reestablishment at Polycomb targets was abolished, while the mark was instead deposited ectopically at inappropriate regions, including gene-poor, partially methylated domains of the genome. In other words, without the H2Aub scaffold, PRC2 not only failed to find its proper targets but painted the wrong parts of the genome.

The team extended the analysis to human embryonic stem cells using an inducible protein-degradation system that removes PRC1 within hours. Acute loss of H2Aub again collapsed H3K27me3 at Polycomb targets and generated ectopic H3K27me3 domains. The experiments also uncovered a second layer of control: DNA methylation. When the researchers inhibited the maintenance methyltransferase DNMT1, additional ectopic H3K27me3 appeared, and combining DNMT1 inhibition with PRC1 degradation produced the strongest mislocalization. H2Aub and DNA methylation therefore cooperate to keep H3K27me3 confined to its correct domains, a partnership that echoes the known interplay between Polycomb complexes and DNA methylation valleys in pluripotent cells.

The study also clarified how PRC1 itself is recruited in human cells. The variant PRC1 machinery, including components such as RYBP, KDM2B and CBX proteins, binds both the GC-rich active genes and the classical Polycomb targets. Knocking down RYBP in stem cells preferentially affected the active GC-rich genes, while catalytic loss of PRC1 hit both classes, with the strongest transcriptional consequences at silenced Polycomb targets. This dual role, dampening highly active GC-rich promoters while enforcing repression at developmental genes, helps explain why H2Aub accumulates at active genes in human embryos without silencing them outright.

Taken together, the findings sketch a stepwise model for the earliest establishment of Polycomb repression in humans. H2Aub appears first, seeded at GC-rich DNA in two-cell embryos, expands across active and inactive genes, and culminates at Polycomb targets in the blastocyst, where it guides PRC2 to redeposit H3K27me3 on a genome that had been wiped nearly clean. Because the experiments were performed in human gametes and embryos, the work fills a gap that mouse studies could not, and it carries practical weight for assisted reproduction, since errors in this reprogramming are thought to contribute to early developmental failure. The datasets have been deposited in public archives, and the authors suggest that the H2Aub-dependent mechanism they describe represents the foundational step by which human embryos rebuild one of their most important layers of gene control.

Subject of Research: H2Aub-dependent reestablishment of Polycomb repression in human preimplantation embryos

Article Title: Establishing Polycomb regulatory landscapes in human early development

Article References: Hu, X., Zhang, C., Liu, B., Lai, F., Ma, S., Hou, Z., Shao, Q., Liu, L., Lu, X., Yu, G., Liu, H., Yang, Z., Wang, N., Wu, X., Lin, Z., Wang, L., Shan, Y., Pan, G., Chen, D., … Xie, W. (2026). Establishing Polycomb regulatory landscapes in human early development. Nature Genetics. https://doi.org/10.1038/s41588-026-02764-y

Image Credits: AI Generated

DOI: 10.1038/s41588-026-02764-y

Keywords: Polycomb, H2AK119ub, H3K27me3, PRC1, PRC2, human embryos, epigenetics, blastocyst, histone modification, DNA methylation, embryonic stem cells, Stacc-seq

The review, led by Professor Kefei Yuan of West China Hospital, Sichuan University, and colleagues, was published online in the Chinese Medical Journal on July 30, 2026. It systematically assembles the latest research on how epigenetic regulation and metabolic reprogramming interact in HCC, organizing a sprawling literature into a coherent framework. “A thorough understanding of how epigenetic modifications and metabolic reprogramming interact in HCC is critical for elucidating the mechanisms underlying hepatocarcinogenesis and developing new treatment approaches against HCC,” explains Prof. Yuan, the corresponding author of the review. “In this article, we systematically introduce the molecular basis of each epigenetic modification, then summarize the latest research progress on epigenetic regulation of metabolism in HCC, trying to link these two frontier fields.”

The central insight of the review is that the relationship between metabolism and epigenetics is bidirectional. Metabolic pathways generate small molecules that serve as substrates or cofactors for epigenetic enzymes. Acetyl-CoA, produced from glucose and lipid metabolism, fuels histone acetylation. S-adenosylmethionine (SAM), a product of one-carbon amino acid metabolism, supplies the methyl groups that decorate DNA and histones. Lactate, the end product of the Warburg effect—the phenomenon in which cancer cells preferentially ferment glucose even in the presence of oxygen—has emerged as a signaling molecule in its own right, driving a newly recognized histone modification called lactylation. In this way, the metabolic state of a cell is written directly into its chromatin, the packaged form of DNA that determines which genes are accessible for transcription.

The reverse direction is equally consequential. Epigenetic modifications alter the expression of metabolic enzymes, reinforcing and amplifying the metabolic phenotype of the tumor. Aberrant methylation of promoters controlling metabolic genes can silence or activate entire pathways, shifting cells toward enhanced glycolysis, increased fatty acid synthesis, or reprogrammed amino acid handling. This feedback loop—metabolites shaping epigenetic marks, epigenetic marks shaping metabolic enzymes—creates a self-reinforcing circuit that locks HCC cells into their malignant state and contributes to the construction of a tumor-promoting microenvironment. The review organizes this crosstalk around five major types of epigenetic modification, each with distinct connections to metabolism.

The first is DNA methylation, which shows bidirectional crosstalk with multiple metabolic pathways. On one side, aberrant methylation of metabolic gene promoters drives changes in glycolysis, lipid synthesis, and amino acid metabolism, effectively rewriting the tumor’s metabolic instruction manual. On the other side, metabolic intermediates such as SAM, derived from one-carbon metabolism, provide the raw material for DNA methylation itself, closing the loop. This means that nutritional state, enzyme availability, and flux through one-carbon metabolism can all influence which genes are silenced or expressed, and those expression changes in turn reshape metabolism further.

The second is RNA methylation, particularly N6-methyladenosine (m6A), the most abundant internal modification of messenger RNA in mammalian cells. The review describes how m6A fine-tunes the expression of metabolic enzymes through a cast of regulatory proteins: “writers” such as METTL3 that install the mark, “erasers” such as FTO that remove it, and “readers” such as YTHDFs and IGF2BPs that interpret it and determine the fate of the modified transcript. This dynamic system controls key metabolic processes in HCC, including glycolysis, lipid synthesis, and amino acid metabolism, allowing tumor cells to adjust enzyme production rapidly in response to shifting conditions without altering the underlying DNA sequence.

The third category is non-coding RNAs, which form a multi-layered regulatory network. MicroRNAs directly target the messenger RNAs of metabolic enzymes, dampening or eliminating their translation. Circular RNAs act as sponges that sequester microRNAs, relieving their inhibitory effects, and some can even encode functional peptides. Long non-coding RNAs operate through yet other mechanisms: they interact with proteins, remodel chromatin, or stabilize metabolic enzymes, thereby influencing metabolic pathways at multiple levels simultaneously. The sheer diversity of these RNA species means that a single metabolic pathway in HCC may be regulated by dozens of non-coding RNAs acting in concert, creating redundancy that can frustrate single-target therapies but also offering multiple points of intervention.

The fourth is histone modification, the family of chemical tags on chromatin proteins that includes methylation, acetylation, lactylation, phosphorylation, and ubiquitination. These modifications integrate metabolic signals directly into transcriptional programs. Histone acetylation depends on acetyl-CoA derived from glucose and lipid metabolism, meaning that a cell burning sugar at high rates is chemically primed to open up its genes. Histone methylation depends on SAM from amino acid metabolism, linking nitrogen and one-carbon metabolism to gene expression. Perhaps most striking is lactylation, an emerging modification in which lactate—the very molecule produced by the Warburg effect—is attached to histones, directly coupling glycolytic flux to gene regulation. The review also flags RNA editing as a fifth mechanism, predominantly adenosine-to-inosine (A-to-I) editing mediated by the enzyme ADAR1, which contributes to metabolic reprogramming by altering key genes involved in polyamine metabolism and oxidative stress responses.

What elevates this review beyond a descriptive catalog is its therapeutic framing. The epigenetic-metabolism axis, the authors argue, represents a promising target for drug development. Preclinical studies have shown that targeting DNA methyltransferases, histone modifiers, or non-coding RNAs can reverse metabolic reprogramming and suppress tumor growth. Inhibitors of DNMT1, a DNA methyltransferase, and of histone methyltransferases have demonstrated anti-tumor effects in HCC models. Meanwhile, nanoparticle-mediated delivery of small interfering RNAs targeting metabolic enzymes has shown efficacy in reducing tumor burden in preclinical settings. “Given that epigenetic mechanisms drive metabolic reprogramming essential for HCC survival and drug resistance, targeting this axis theoretically could overcome chemoresistance,” notes Prof. Yuan. “However, promising preclinical candidates have not yet advanced to clinical trials, underscoring the need for further interdisciplinary research.”

The authors are candid about the gaps that remain. The specific effects of metabolites such as cholesterol, ethanol, and alpha-ketoglutarate on DNA methylation require further exploration, a question with direct relevance given that alcohol exposure and lipid dysregulation are major risk factors for liver disease. The relationship between histone modifications and amino acid metabolism remains unclear, as does the role of ketone body metabolism in histone acetylation and the involvement of histone demethylases in metabolic reprogramming. These unresolved questions define the research agenda for the field. What is already clear, however, is that the epigenetic-metabolic crosstalk in HCC is not a side effect of cancer but a central organizing principle of it—a bidirectional network in which every metabolic choice a tumor cell makes is recorded in its chromatin, and every epigenetic mark reshapes its metabolism. Untangling that loop, the review suggests, may be the key to therapies that do not merely attack liver cancer cells but dismantle the regulatory logic that sustains them.

Subject of Research: The bidirectional crosstalk between epigenetic regulation and metabolic reprogramming in hepatocellular carcinoma

Article Title: New review in Chinese Medical Journal illuminates the crosstalk between epigenetics and metabolism in hepatocellular carcinoma

Article References: New review in Chinese Medical Journal illuminates the crosstalk between epigenetics and metabolism in hepatocellular carcinoma. Original publication (Bibliographic details unavailable.)

Image Credits: AI Generated

DOI: Not provided

Keywords: hepatocellular carcinoma, epigenetics, metabolic reprogramming, DNA methylation, RNA methylation, m6A, non-coding RNAs, histone modification, lactylation, Warburg effect, cancer metabolism, Chinese Medical Journal

The scale of the underlying literature is striking. The team systematically searched PubMed and Web of Science for studies reporting food-derived or dietary compounds with documented effects on epigenetic pathways, filtering out reviews, meta-analyses, non-English texts, and articles lacking essential metadata. The final corpus comprises 1,963 primary research articles spanning from 1991 through August 2026, with publication counts climbing steeply in recent years. The year 2025 alone contributed 211 articles, the highest annual total in the dataset, followed by 2022 with 153 and 2024 with 150. This temporal pattern mirrors the broader surge of interest in nutriepigenomics, the study of how nutrition modulates the chemical tags and chromatin structures that govern gene expression.

Building such a database by hand would have taken years of painstaking curation, so the Mexican team, led by Aylin Del Moral-Morales and José L. Medina-Franco of the DIFACQUIM Research Group, turned to artificial intelligence. They used a large language model, the Claude API accessed through the LangChain framework, to read each article and extract structured pairs of compound names and food sources. The model was instructed to adopt the analytical stance of a medicinal chemist, annotating every dietary, beverage, or food-derived compound explicitly described as a modulator of DNA methylation, histone modification, or non-coding RNA pathways. Crucially, the temperature parameter was set to zero to suppress output variability and minimize hallucinations, and the prompts explicitly excluded vague group designations such as vitamins or polyphenols in favor of identifiable molecular entities.

Raw extraction from scientific text is only the beginning; the real challenge is making thousands of heterogeneous terms speak the same language. Food source names varied wildly across the literature, so the researchers embedded every unique dietary source string as a dense numerical vector using the all-MiniLM-L6-v2 sentence-transformer model. Agglomerative hierarchical clustering with a cosine distance threshold of 0.3 then grouped lexically and semantically similar terms, and the shortest member of each cluster became the canonical label. This semantic harmonization collapsed the messy vocabulary of the literature into 585 standardized food and beverage source terms, from tea and grapes to turmeric, soy, citrus, olive, broccoli, beef, milk, egg, and fish oil.

Compound identity received equally rigorous treatment. Every extracted compound name was mapped to the PubChem database through the PubChemPy interface, retrieving compound identifiers, IUPAC names, InChIKeys, SMILES structure strings, and curated synonyms. When a query returned multiple matches, the first valid PubChem CID was adopted as the reference. The structures then passed through RDKit, a cheminformatics toolkit that stripped salts down to their primary fragments, neutralized formal charges, canonicalized tautomers, and converted everything into a single canonical SMILES form. Records lacking a valid CID, an annotated mechanism, or a compound name were systematically excluded, ensuring that every entry in the final database is chemically unambiguous and interoperable with other resources.

To add mechanistic depth, the team integrated quantitative binding data from BindingDB, a curated knowledgebase of protein-small molecule interactions. Only records involving human proteins were retained, and each interaction had to include a numeric affinity measure in nanomolar units, such as IC50, Ki, Kd, or EC50. The target proteins were then cross-referenced against the EpiFactors database, which classifies epigenetic regulators by molecular function, so that only proteins with validated roles in epigenetic regulation survived the filter. Each binding record was annotated with functional categories such as writer, reader, or eraser of epigenetic marks, allowing researchers to trace a dietary molecule all the way from the dinner table to a specific enzymatic pocket on a chromatin-modifying enzyme.

The resulting numbers reveal where the field’s attention has concentrated. Among the 653 compounds, folate, choline, L-methionine, resveratrol, and vitamin B12 top the citation rankings, each linked to more than 100 unique literature references. Curcumin, butyrate, bisphenol A, betaine, and sulforaphane round out the most-studied list. The distribution spans micronutrients, bioactive phytochemicals such as quercetin, genistein, and epigallocatechin gallate, and even synthetic or environmental agents like valproic acid. The strong emphasis on methyl donors, B vitamins, and polyphenols reflects their well-established mechanistic ties to DNA methylation, histone modification, and histone deacetylase inhibition. On the food side, tea leads the diversity ranking with 46 unique epigenetically active compounds, ahead of grapes, turmeric, and soy, and plant-derived sources generally outpace animal-derived ones in chemical variety.

The dataset also exposes gaps that could shape future research. Of the 43 distinct epigenetic mechanisms cataloged, DNA methylation and HDAC inhibition dominate the record counts, while RNA-related modifications such as RNA methylation, demethylation, and alternative splicing remain thinly represented. Among the 80 protein targets, histone-modifying writers and erasers are the most prevalent functional classes, with 29 and 20 factors respectively, whereas Polycomb group proteins and DNA modification cofactors count only one or two members each. Binding affinity data cover 101 unique compound-target pairs, with IC50 measurements accounting for over 90 percent of the 2,877 binding records, while Ki and Kd values are comparatively rare. These imbalances offer a roadmap for underexplored territory, particularly non-histone targets and less prevalent modification types.

What sets NutriEpiDB apart from earlier compilations is its commitment to machine readability and responsible artificial intelligence use. The dataset is FAIR2-certified, meeting findability, accessibility, interoperability, and reusability standards extended with AI-readiness and responsible AI criteria. All tables are provided in open formats including CSV, Parquet, and JSON, structured according to the MLCommons Croissant standard, and annotated with schema.org and W3C PROV-O provenance metadata that trace every variable back to its data generation step. The resource is released under the Open Data Commons Attribution License, permitting unrestricted reuse with attribution, and is accessible through an interactive data portal alongside a complete data package containing raw and processed data, transformation scripts, and methodological documentation.

The authors are candid about the limitations. Because the database was built primarily from article titles and abstracts, population descriptors such as ethnicity, genetic background, and country of recruitment are largely unavailable, and the English-language restriction plus reliance on PubMed and Web of Science may underrepresent certain geographic and cultural perspectives. Selection bias toward well-studied targets like histone deacetylases is evident, and compounds lacking well-defined PubChem structures, including some metabolites and obscure phytonutrients, are necessarily absent. Most binding data derive from in vitro assays, and the authors explicitly caution that the compounds and food sources listed should not be interpreted as dietary or therapeutic recommendations, since safety, tolerability, and dose-dependent effects must be assessed individually. Still, with semi-automated update pipelines planned and interoperability with PubChem, EpiFactors, UniProt, and BindingDB maintained as those repositories evolve, NutriEpiDB is positioned to become foundational infrastructure for hypothesis generation, machine learning model development, and evidence synthesis at the intersection of nutrition, epigenetics, and computational biology.

Subject of Research: A curated database of dietary compounds with epigenetic targets built from systematic literature mining and AI-assisted extraction

Article Title: NutriEpiDB: a database for dietary compounds with epigenetic targets

Article References: Aylin, D. M.-M., & Medina-Franco, J. L. (2026). NutriEpiDB: a database for dietary compounds with epigenetic targets. Frontiers in Drug Discovery, 6, Article 1946880. https://doi.org/10.3389/fddsv.2026.1946880

Image Credits: AI Generated

DOI: 10.3389/fddsv.2026.1946880

Keywords: nutriepigenomics, epigenetics, dietary compounds, database, large language models, DNA methylation, histone modification, binding affinity, PubChem, nutrigenomics, FAIR data, cheminformatics

The intellectual roots of the idea reach back further than most readers might expect. As early as 1931, researchers hypothesized that bacillus Calmette–Guérin (BCG) vaccination could reduce neonatal mortality through nonspecific stimulation of the immune system, an effect that could not be explained by protection against tuberculosis alone. The concept of trained immunity was formally articulated in the twenty-first century, when studies showed that monocytes exposed to certain microbial components responded more vigorously to a second, unrelated challenge. Interest surged again during the COVID-19 pandemic, when clinical observations revived the question of whether BCG vaccination might confer broad, heterologous protection against viral infections. The new review, led by Liqi Liao, Yong-Xian Cheng and Xuhan Liu, places these threads into a single conceptual framework spanning cellular basis, inducing factors and the full disease spectrum.

At the cellular level, trained immunity is not confined to circulating monocytes and macrophages. The review emphasizes that hematopoietic stem and progenitor cells in the bone marrow can themselves be trained, generating myeloid progeny with an enhanced inflammatory repertoire long after the initial stimulus has cleared. BCG vaccination in humans has been shown to reprogram the hematopoietic progenitor compartment, and animal studies demonstrate that trained hematopoietic stem cells generate protective innate immunity against tuberculosis. Tissue-resident macrophages, natural killer cells, innate lymphoid-like cells and dendritic cells have all been shown to display memory-like behavior under appropriate stimulation. Even organisms lacking adaptive immunity, from invertebrates to plants, exhibit forms of innate immune memory, suggesting that this is an ancient defense strategy rather than a mammalian novelty.

The triggers of trained immunity are diverse. Pathogen-associated molecular patterns such as β-glucan from fungal cell walls and lipopolysaccharide from bacterial membranes can induce the trained state, as can damage-associated molecular patterns released by injured tissue. Oxidized low-density lipoproteins, oxidized phospholipids carried on lipoprotein(a), and even catecholamines released during stress have been shown to train monocytes toward a proinflammatory phenotype. The Western diet, rich in fat and sugar, triggers NLRP3-dependent innate immune reprogramming, effectively training the immune system toward chronic metaflammation. This breadth of inducers explains why trained immunity is now implicated far beyond infectious disease, in atherosclerosis, metabolic syndrome and cancer.

Mechanistically, the review centers on what its authors call the epigenetic-metabolic axis. Upon stimulation, innate immune cells activate the mTOR–HIF-1α signaling cascade, driving a shift from oxidative phosphorylation toward aerobic glycolysis, a metabolic signature once considered exclusive to cancer cells. This metabolic rewiring is not merely fuel for the cell; it directly shapes the epigenetic landscape. Accumulated glycolytic intermediates and altered flux through the tricarboxylic acid cycle influence the availability of substrates and cofactors for chromatin-modifying enzymes, linking cellular energetics to gene regulation in a feedback loop that can persist for weeks to months.

Two histone modifications emerge as central players. Trimethylation of histone H3 at lysine 4 (H3K4me3) and acetylation at lysine 27 (H3K27ac) mark promoters and enhancers of inflammatory genes, keeping them in a poised state that allows rapid transcription upon re-exposure to a stimulus. The mevalonate and cholesterol biosynthesis pathways provide a second metabolic route into this program: activation of these pathways in trained monocytes drives both histone modifications and long-term functional changes. One-carbon metabolism, which supplies S-adenosylmethionine for methylation reactions, and NAD+-dependent sirtuins further couple metabolic state to chromatin structure. Long noncoding RNAs positioned near immune genes and three-dimensional chromatin architecture, including topologically associated domains, add additional layers of regulation that keep trained cells ready for action.

The protective face of trained immunity is best illustrated by BCG. Vaccination induces NOD2-dependent epigenetic reprogramming of monocytes, producing enhanced cytokine responses to unrelated pathogens. Randomized trials, including the ACTIVATE study in elderly individuals, showed that BCG vaccination increased the time to first infection, and experimental challenge studies demonstrated reduced viremia after yellow fever vaccination in BCG-vaccinated volunteers. Animal work extends the principle: intravenous BCG protects mice against lethal SARS-CoV-2 challenge, mucosal BCG and MTBVAC vaccination induce stronger trained immunity than standard intradermal routes, and trained alveolar macrophages in the lung confer prolonged antibacterial and antitumor protection. β-glucan-induced trained immunity has been shown to protect against Leishmania infection and to promote antitumor granulopoiesis, while trained macrophages can even confer heterologous protection against gram-negative bacteria after Staphylococcus aureus exposure.

Yet the same machinery can turn against the host. In atherosclerosis, oxidized low-density lipoprotein trains monocytes and macrophages to produce sustained proinflammatory cytokines and form foam cells, embedding a proinflammatory memory in the arterial wall. Myocardial infarction itself drives trained immunity of monocytes, accelerating subsequent atherosclerosis, and maladaptive training of myelopoiesis has been shown to link inflammatory comorbidities. In certain tumor microenvironments, trained or lipid-associated macrophages facilitate cancer progression rather than restraint. Chronic interleukin-1 exposure pushes hematopoietic stem cells toward precocious myeloid differentiation at the expense of self-renewal, a shift that can undermine long-term immune resilience. The review stresses that the outcome of trained immunity is strictly context-dependent: the identical epigenetic-metabolic program that accelerates pathogen clearance in acute infection can fuel disease progression when the stimulus is chronic.

This duality opens a therapeutic frontier. Host-directed therapies could deliberately induce trained immunity to boost vaccine efficacy or enhance antimicrobial defense, using agents such as β-glucan particles, bacterial outer membrane vesicles, or nanobiologics that deliver trained-immunity-promoting signals. Conversely, dampening pathological training could treat cardiovascular disease, with the mevalonate pathway, mTOR signaling and specific histone-modifying enzymes such as lysine methyltransferases and demethylases serving as druggable nodes. In oncology, strategies that retrain pro-metastatic macrophages or combine trained-immunity-inducing nanoparticles with checkpoint inhibitors have already shown promise in preclinical models. Compounds such as oroxylin A, which promotes LC3-associated phagocytosis in trained macrophages, hint at pharmacological routes to harness the process.

The authors argue that single-cell RNA sequencing, multi-omics profiling and systems approaches will be essential to translate these insights into precision immunology, identifying which patients stand to benefit from immune training and which are at risk from it. Recent multi-omics work on BCG vaccination has already revealed epigenetic cell states that predict the strength of the trained response in individuals, a first step toward personalized vaccination strategies. As the field matures, trained immunity is shifting from a provocative concept to a practical framework: one that reframes immune memory as a continuum spanning innate and adaptive systems, and one whose manipulation, the review concludes, could reshape the prevention and treatment of infectious, inflammatory and oncological disease alike.

Subject of Research: The epigenetic and metabolic mechanisms of trained immunity and their dual roles in infection, chronic inflammation and cancer

Article Title: The epigenetic-metabolic axis of trained immunity: dual roles in disease pathogenesis and intervention

Article References: Liao, L., Zhai, Z., Zhong, L., Shi, C., Yao, S., Tao, L., Cheng, Y.-X., & Liu, X. (2026). The epigenetic-metabolic axis of trained immunity: dual roles in disease pathogenesis and intervention. Journal of Molecular Medicine, 104(1), Article 121. https://doi.org/10.1007/s00109-026-02729-y

Image Credits: AI Generated

DOI: 10.1007/s00109-026-02729-y

Keywords: trained immunity, epigenetic reprogramming, metabolic rewiring, BCG vaccine, innate immune memory, mTOR-HIF-1α signaling, histone modification, mevalonate pathway, atherosclerosis, cancer immunotherapy, monocytes, hematopoietic stem cells

Stress granules are membraneless structures that suddenly appear inside cells when they face hostile conditions such as heat shock, oxidative damage, or nutrient deprivation. When translation of messenger RNA stalls, the stalled mRNA molecules, together with RNA-binding proteins like G3BP1 and TIA1, condense into these granules, effectively putting nonessential protein production on hold until conditions improve. In the short term, this is a survival strategy. But when stress granules persist or accumulate abnormally, they have been implicated in neurodegenerative disease and chronic inflammatory damage. Whether they play a role in diabetic injury of the pancreas had not been systematically explored until now.

The research team, led by corresponding author Dongbo Xue, began by mining single-cell RNA sequencing data from human pancreatic tissue affected by diabetes mellitus. This technology allows scientists to profile gene expression in thousands of individual cells at once, revealing how each cell population changes under disease conditions. Their analysis showed that under diabetic conditions, the proportion of acinar cells in the pancreatic tissue decreased, a sign that these cells are being lost or damaged. At the same time, the fraction of acinar cells that scored positive for stress granule related gene signatures increased, indicating that the surviving cells were under heightened stress.

To understand what was happening at the chromatin level, the investigators integrated chromatin immunoprecipitation sequencing data for H3K4me1, a histone mark in which a methyl group is added to the fourth lysine residue of histone H3 in a monomethylated form. H3K4me1 is classically associated with poised or active enhancer regions of the genome, and it can elevate transcription of nearby genes. The team found that the level of this modification was elevated in diabetes, and that the pattern of H3K4me1 enrichment corresponded to the changes in stress granule related gene expression seen in the acinar cell subtypes.

The critical question was whether this correlation reflected a genuine causal mechanism. To test it, the researchers turned to in vitro experiments using acinar cells exposed to a high glucose environment, mimicking the hyperglycemic conditions of diabetes. Western blotting, a technique that detects specific proteins by their size and antibody binding, showed that high glucose promoted the phosphorylation of eIF2α, a translation initiation factor whose phosphorylation is a classic trigger of the integrated stress response. When eIF2α is phosphorylated, protein synthesis is throttled back, and the stalled translation complexes become the raw material for stress granule assembly.

Consistent with this, the high glucose treatment increased the expression of G3BP1 and TIA1, two core scaffolding proteins of stress granules, and immunofluorescence imaging confirmed enhanced aggregation of stress granules within the treated cells. Immunofluorescence uses fluorescently labeled antibodies to light up specific molecules inside cells, allowing researchers to see directly where and how abundantly the granules form. The visual evidence matched the biochemical data: cells bathed in high glucose were visibly studded with stress granules compared with their normally cultured counterparts.

The epigenetic link was then tested directly. When the researchers inhibited H3K4me1, the expression of G3BP1 and TIA1 fell, and stress granule formation decreased accordingly. This result, confirmed by chromatin immunoprecipitation followed by quantitative PCR, a method that measures how much of a specific histone mark sits on particular gene promoters, establishes H3K4me1 as an upstream driver rather than a mere bystander. In other words, the hyperglycemic environment appears to rewire the epigenetic landscape of acinar cells, and that rewiring licenses the assembly of stress granules by boosting the production of the proteins that build them.

The findings carry several important implications. First, they identify pancreatic acinar cells as active participants in diabetic pathology, not just passive bystanders to the better known destruction of insulin producing beta cells. Acinar cells are the most abundant cell type in the pancreas and are responsible for secreting the enzymes that digest food in the small intestine. Their loss or dysfunction in diabetes could contribute to the exocrine pancreatic complications, including digestive problems and pancreatitis like inflammation, that are increasingly recognized in diabetic patients. Second, the study connects two fields that rarely intersect: chromatin biology and RNA stress physiology, suggesting that epigenetic marks may serve as master switches controlling how cells handle stress at the level of protein synthesis.

There is also a therapeutic dimension. If elevated H3K4me1 is what drives excessive stress granule formation in acinar cells, then pharmacological tools that modulate the enzymes responsible for writing or erasing this histone mark could, in principle, blunt the stress response and protect the pancreas from diabetic damage. The authors caution that their work is at an early stage, relying on human single cell data analysis and cell culture experiments rather than direct intervention in patients, and the study was published as an early version subject to further editorial refinement. Animal experiments in the study were conducted under approval from the Ethics Committee of the First Affiliated Hospital of Harbin Medical University and in compliance with ARRIVE 2.0 reporting guidelines, and the work was funded by the National Natural Science Foundation of China.

For the broader research community, the study opens a rich vein of follow up questions. Which specific genes are regulated by the elevated H3K4me1 in acinar cells, and are the same mechanisms at work in other tissues damaged by diabetes, such as the kidney, retina, and peripheral nerves? Do stress granules in acinar cells eventually dissolve and restore normal translation, or do they persist and seed pathological protein aggregation over years of hyperglycemia? And could stress granule signatures detected in pancreatic tissue or circulating biomarkers serve as early warning signs of acinar cell injury before overt disease appears? As the prevalence of diabetes continues to climb worldwide, understanding how high glucose rewrites the chemical instructions inside our cells, one methyl group at a time, may prove essential to protecting organs far beyond the insulin producing beta cells that have long dominated the conversation.

Subject of Research: Epigenetic regulation of stress granule formation in pancreatic acinar cells under diabetic hyperglycemia

Article Title: Elevated H3K4me1 in pancreatic acinar cells mediates stress granule formation in response to hyperglycemia in diabetes

Article References: He, Y., Li, L., Du, Z., Zheng, Y., Meng, Z., Liu, X., Xie, Z., Wang, H., Hao, C., & Xue, D. (2026). Elevated H3K4me1 in pancreatic acinar cells mediates stress granule formation in response to hyperglycemia in diabetes. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08888-3

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08888-3

Keywords: diabetes mellitus, pancreatic acinar cells, stress granules, H3K4me1, histone modification, epigenetics, hyperglycemia, G3BP1, TIA1, eIF2α phosphorylation, single-cell RNA sequencing, ChIP-seq

The scale of the problem justifies the effort. Sudden cardiac death is characterized by high incidence, low survival rates, and, crucially, it often presents as the very first manifestation of an underlying cardiovascular disease. Studies cited in the review, including a 2025 analysis in the Journal of the American Heart Association of sudden cardiac death mortality in the United States between 1999 and 2022, and a 2025 study in JACC: Clinical Electrophysiology showing that sudden cardiac death is frequently the first sign of cardiovascular disease, underline how little opportunity clinicians have to intervene before the fatal arrhythmia strikes. Out-of-hospital survival rates remain low, and risk stratification tools, including electrocardiographic markers and clinical scores, capture only part of the danger. That gap has pushed researchers toward the genome and, more recently, the epigenome, in search of signals that identify vulnerable hearts before they fail.

The genetic story begins with common variation. The review summarizes single nucleotide polymorphisms and insertion/deletion variants associated with sudden cardiac death, discovered through two complementary strategies: candidate-gene studies, which examine genes chosen for their known roles in cardiac function, and genome-wide association studies, which scan the entire genome without prior assumptions. A recurring theme is that risk variants cluster in genes involved in cardiovascular function, signal transduction, and cellular metabolism. Among the candidates highlighted are variants in NOS1AP, CASQ2 and GPD1L associated with sudden death risk in coronary artery disease; polymorphisms in angiotensin-converting enzyme-related pathways; a TGFBR2 polymorphism linked to cardiac arrest risk; common variation in fatty acid metabolic genes; and a missense variant in the neuregulin 1 gene that, remarkably, has been associated with both schizophrenia and sudden cardiac death, hinting at shared biology between neural and cardiac electrical systems.

Insertion/deletion polymorphisms form a second, technically distinct layer of variation. These variants, in which short stretches of DNA are present or absent, can alter gene dosage, disrupt regulatory sequences, or change protein structure. The review catalogues a series of such variants reported largely in Chinese populations, including indels in CTH, HSPA1B, the 3′ untranslated region of COL1A2, STIM1, COX10, NPC1, DSG2, LTBP4, the MIR155HG locus, the 3’UTR of STAT5A, and the alpha2-adrenergic receptor gene, the latter identified in Finnish cohorts as a genetic risk factor for sudden cardiac death. Several of these variants sit in regulatory regions rather than coding sequences, which means their functional impact likely operates through altered gene expression, a mechanism that connects naturally to the epigenetic half of the review.

Genome-wide association studies have added their own landmarks. A protective locus at GPC5 was identified in 2010; a susceptibility locus at chromosome 2q24.2 was mapped in individuals of European ancestry in 2011; and subsequent efforts, including a comprehensive 2018 evaluation of the genetic architecture of sudden cardiac arrest published in the European Heart Journal, have refined the picture. Yet the review is candid about a persistent problem: missing heritability. Known variants explain only a fraction of the familial clustering and population-level risk that epidemiologists observe, a challenge familiar across complex-disease genetics since the influential 2009 Nature paper by Manolio and colleagues. Rare variants, gene-gene interactions, gene-environment interactions, and epigenetic modifications are all candidate explanations for the unexplained portion, and the Moscow team argues that only large-scale, integrative, multi-ethnic studies combining genomic, epigenomic and clinical data with functional validation will close the gap.

That argument leads directly to epigenetics, the layer of molecular regulation that changes gene activity without altering the DNA sequence itself. The review focuses on three principal mechanisms: microRNAs, DNA methylation, and histone modifications. MicroRNAs are short non-coding RNAs that bind target messenger RNAs and suppress their translation or trigger their degradation. In the cardiovascular system they are pervasive regulators, controlling everything from cardiac conduction to fibrosis. The classic example, miR-1, regulates the arrhythmogenic potential of the heart by targeting the gap junction gene GJA1 and the potassium channel gene KCNJ2, directly linking microRNA levels to the electrical stability of the myocardium. Other microRNAs, including members of the miR-29, miR-133 and miR-30 families, modulate cardiac fibrosis and matrix remodeling, processes that create the structural substrate for lethal arrhythmias.

What makes microRNAs especially attractive for sudden cardiac death research is their measurability in body fluids. Tissue-based microRNAs recovered from post-mortem heart samples and circulating microRNAs detected in plasma both act as mechanistic regulators of the disease process and have emerged as exploratory biomarker candidates. Studies have reported that miR-1, miR-499 and miR-208 are sensitive markers for diagnosing sudden death due to early acute myocardial infarction; that miR-3113-5p, miR-223-3p, miR-133a-3p and miR-499a-5p can discriminate sudden cardiac death; and that plasma extracellular vesicle microRNA-208b-3p and microRNA-143-3p show promise as predictive biomarkers in acute coronary syndrome. Circulating microRNAs have even been examined in relation to sudden death risk in patients with coronary heart disease and to outcomes after cardiac arrest. The forensic dimension is equally striking: quantitative analysis of miR-1, miR-133a and miR-26a in heart tissue and whole blood has been proposed as a tool to help pathologists differentiate causes of death when autopsies are inconclusive.

The review, however, refuses to oversell these findings. It notes that inconsistent directions of association, methodological heterogeneity, and limited independent validation currently preclude the clinical application of microRNA biomarkers. Different studies measure different microRNA panels in different sample types with different normalization strategies, and few results have been replicated in independent cohorts. DNA methylation studies of sudden cardiac death are even scarcer, but they suggest potential roles for imprinted genes and oxidative stress pathways. One notable thread involves Prader-Willi syndrome, an imprinting disorder in which cardiac causes of death are overrepresented; recent work shows that deficiency of the necdin gene, lost in the syndrome, attenuates cardiac contractility, and that loss of Snord116 protects cardiomyocyte kinetics during ischemic stress, providing mechanistic links between imprinted loci and cardiac vulnerability. Differential methylation in the GSTT1 regulatory region has also been reported in sudden unexplained death, pointing toward detoxification and oxidative stress as methylation-sensitive pathways relevant to arrhythmic risk.

Histone modifications and chromatin remodeling represent the most underexplored but most promising frontier. Histone acetylation and deacetylation, mediated by histone acetyltransferases and histone deacetylases, reshape chromatin accessibility and thereby tune the expression of entire gene programs. Experimental work has shown that class I HDACs regulate angiotensin II-dependent cardiac fibrosis, that the stress-activated kinase MKK7 governs the epigenetics of cardiac repolarization and arrhythmia prevention, and that HDAC3 and HDAC5 influence cardiac conduction disease and mitochondrial function through their control of transcription factors such as Nkx2.5, MEF2 and PGC-1alpha. A 2025 experimental study even used the DNA damage marker gamma-H2AX to detect DNA damage in heart tissue from sudden cardiac death victims, suggesting that chromatin-level injury signatures may be readable in post-mortem material. These findings remain at the level of mechanism rather than validated clinical markers, but they sketch a plausible route by which environmental stressors, from ischemia to toxic exposures, become permanently inscribed in cardiac chromatin and translate into electrical instability.

The overall message of the review is one of cautious momentum. The genetic and epigenetic landscape of non-familial sudden cardiac death is no longer blank: common and rare variants, regulatory indels, circulating microRNAs, methylation patterns and chromatin modifiers all contribute pieces to the puzzle. But the pieces do not yet assemble into a predictive model that a cardiologist could apply at the bedside, and the authors are explicit that translating these molecular insights into clinical practice will require large-scale, integrative and multi-ethnic studies that combine genomic, epigenomic and clinical data with functional validation. Until then, the review serves as both a map of what is known and a checklist of what remains undone, and it makes a compelling case that the missing heritability of sudden cardiac death may be hiding, at least in part, not in the sequence of the genome but in the chemical marks written upon it.

Subject of Research: Genetic and epigenetic determinants of non-familial sudden cardiac death

Article Title: Genetic and epigenetic landscape of non-familial sudden cardiac death

Article References: Natalia, B., Natalia, M., Maxim, K., Olga, F., & Ivan, K. (2026). Genetic and epigenetic landscape of non-familial sudden cardiac death. Molecular Biology Reports, 53(1), Article 1630. https://doi.org/10.1007/s11033-026-12821-8

Image Credits: AI Generated

DOI: 10.1007/s11033-026-12821-8

Keywords: sudden cardiac death, genetic variants, single nucleotide polymorphism, insertion-deletion polymorphism, genome-wide association study, epigenetics, microRNA, DNA methylation, histone modification, chromatin remodeling, missing heritability, cardiac arrhythmia

The technical foundation of this story lies in how the brain handles carbon and energy. Neurons are voracious consumers of glucose, and under aerobic conditions glycolysis converts glucose to pyruvate, which is then imported into mitochondria and oxidized by the tricarboxylic acid cycle. When oxygen is scarce or demand outpaces oxidative capacity, lactate dehydrogenase converts pyruvate to lactate, regenerating NAD+ so glycolysis can continue. In the brain, this is not merely an emergency fallback. The astrocyte-neuron lactate shuttle describes a division of labor in which astrocytes, which take up glucose and glutamate from the synaptic environment, export lactate through monocarboxylate transporters for neurons to oxidize as fuel. A complementary intracellular lactate shuttle moves lactate into mitochondria via mitochondrial monocarboxylate transporters, where it can be reconverted to pyruvate and fed into oxidative phosphorylation. The result is a metabolically flexible network in which lactate flows between cell types and compartments as both fuel and signal.

What transformed lactate from fuel to regulator was the discovery in 2019 that lactate can covalently modify lysine residues on histones—the proteins around which DNA is wound—in a reaction termed lactylation. Histone lactylation requires lactyl-coenzyme A as the acyl-donor, and the addition of a lactyl group to lysine residues alters chromatin accessibility, generally opening DNA and promoting gene transcription. This places lactylation alongside acetylation and methylation as an epigenetic mark, but with a crucial difference: its abundance tracks directly with metabolic state. When glycolysis is high and lactate accumulates, lactylation rises; when lactate falls, the mark recedes. Metabolism is thereby coupled to gene expression through a single chemical handle, and the review emphasizes that this coupling is now being mapped in detail across the nervous system.

In neural stem cells and neural progenitor cells, histone lactylation appears to be a molecular throttle on the decision to proliferate and differentiate. The review describes evidence that elevated lactate promotes the expression of genes associated with neural stem cell proliferation and neuronal maturation, including activity-regulated cytoskeletal-associated protein and brain-derived neurotrophic factor, through lactylation-driven epigenetic activation. In the adult brain, neurogenesis persists in two niches: the subgranular zone of the dentate gyrus in the hippocampus and the subventricular zone lining the lateral ventricles. Adult hippocampal neurogenesis, in particular, underpins learning, memory, and mood regulation, and markers such as doublecortin flag the newborn neurons it produces. By linking glycolytic flux to the chromatin state of neural stem cells, lactylation provides a mechanism by which the metabolic environment of the niche—oxygen tension, vascular supply, inflammatory tone—can be translated into instructions about whether progenitors stay quiescent, divide, or commit to a neuronal fate.

Lactylation is not confined to histones. Non-histone proteins, including enzymes, structural proteins, and signaling molecules, carry lysine lactylation marks that can alter their activity, stability, localization, or interaction partners. The review highlights that non-histone lactylation regulates specific protein functions in ways that extend the modification’s influence well beyond gene regulation, though the precise functional consequences at individual sites remain an active area of investigation. Because lactylation, like acetylation, adds a bulky, negatively charged group to lysine, it can compete with or mimic other modifications, creating a dense regulatory cross-talk in which metabolic state reshapes the entire post-translational modification landscape of a cell.

The protective face of lactate in the nervous system rests on three pillars described in the review. First, lactate is an efficient energy substrate for neurons, and supporting neuronal metabolism can buffer cells against the energetic stress that accompanies aging and disease. Second, lactate signaling promotes neurotrophic effects: lactate can stimulate vascular endothelial growth factor A production, supporting the vascular niche that nourishes neural progenitors, and it engages hydroxycarboxylic acid receptor 1, a G-protein-coupled receptor that mediates anti-inflammatory signaling. Third, lactate can dampen neuroinflammation. In models of activated immune cells, elevated lactate has been shown to shift macrophage and microglial behavior toward less inflammatory phenotypes, in part through lactylation-dependent and HCAR1-dependent pathways. In the context of neurodegeneration, where chronic inflammation and mitochondrial dysfunction are hallmarks, these mechanisms suggest that boosting lactate availability or lactylation could, in principle, slow neuronal loss.

The toxic face emerges when the same chemistry is applied in the wrong context. The review is explicit that the switch between protection and harm depends on concentration, cell type, pathological setting, and the specific lactylation site involved. Excessive lactate accumulation accompanies the acidic, hypoxic conditions of injury and chronic disease, where it can exacerbate oxidative damage and support inflammatory programs. In neurodegenerative diseases characterized by protein aggregation—amyloid-beta and tau in Alzheimer’s disease, alpha-synuclein in Parkinson’s disease, huntingtin in Huntington’s disease—altered post-translational modification landscapes may influence how proteins misfold, aggregate, and evade clearance. Lactylation of proteins involved in aggregation or in the machinery that clears misfolded proteins could tip the balance toward pathology. Similarly, in amyotrophic lateral sclerosis, where motor neuron degeneration is coupled to defects in the motor neuron lactate shuttle and disrupted energy metabolism, the direction of lactate flux may determine whether supporting cells nourish or starve the vulnerable neurons they surround.

The disease-specific evidence reviewed by the authors paints a picture of context-dependence rather than simple causation. In Alzheimer’s disease, impaired glucose metabolism and reduced lactate handling are observed alongside amyloid precursor protein pathology, and altered lactate dynamics in the cerebrospinal fluid reflect broader metabolic decline; yet lactate’s neurotrophic and anti-inflammatory actions suggest potential benefit if delivered appropriately. In Parkinson’s disease, dopaminergic neurons in the substantia nigra are energetically fragile, with high mitochondrial demand and susceptibility to reactive oxygen species, making metabolic support attractive but also raising the stakes of any intervention that perturbs redox balance. In Huntington’s disease, mutant huntingtin disrupts transcription and metabolism broadly, and lactylation-dependent gene regulation could either compensate for or compound these defects. The unifying insight is that lactate is not simply good or bad; it is a dose-, timing-, and site-dependent variable that must be read in the context of each disease stage and cellular environment.

Therapeutic implications follow directly from this framing, and the review evaluates several strategies for manipulating lactate metabolism and lactylation. Approaches include modulating lactate dehydrogenase activity, targeting monocarboxylate transporters to redirect lactate flux between cells, pharmacologically shifting metabolism with agents such as dichloroacetate that favor mitochondrial oxidation over glycolysis, and exploiting exercise—high-intensity interval training elevates circulating lactate and has been associated with enhanced neurotrophic signaling and adult hippocampal neurogenesis. Enzymes that write and erase lysine lactylation marks are also candidate drug targets, analogous to the histone acetyltransferases and deacetylases that have yielded epigenetic therapies in oncology. The challenge, as the authors stress, is achieving the right modification at the right site in the right cells: systemic manipulation of lactate would simultaneously affect neurons, astrocytes, microglia, endothelial cells, and peripheral immune cells, each of which may respond differently.

The review’s conclusion is a call for precision rather than enthusiasm. Lactate and protein lactylation exhibit a genuine regulatory duality in neurogenesis and neurodegenerative disease, and the factors that determine which face prevails—concentration, cell type, pathological context, and lactylation site—are only beginning to be mapped. Site-specific mapping of the lactylome in human disease tissue, tools for cell-type-selective manipulation of lactylation, and biomarkers that report on lactylation state in living patients are among the needs the authors identify for clinical translation. If those gaps close, the molecule once dismissed as metabolic exhaust may become a lever for regenerating the aging and diseased brain—one lysine residue at a time.

Subject of Research: Roles of lactate metabolism and lysine lactylation in adult neurogenesis and neurodegenerative disease

Article Title: Roles of lactate and protein lactylation in neurogenesis and neurodegenerative disease

Article References: Zhou, X., Han, J., Cui, X., Wu, S., Lu, Z., & Zong, S. (2026). Roles of lactate and protein lactylation in neurogenesis and neurodegenerative disease. Journal of Translational Medicine, 24(1), Article 1210. https://doi.org/10.1186/s12967-026-08940-2

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08940-2

Keywords: lactate, protein lactylation, neurogenesis, neural stem cells, epigenetics, Alzheimer's disease, Parkinson's disease, Huntington's disease, amyotrophic lateral sclerosis, astrocyte-neuron lactate shuttle, histone modification, neuroinflammation

Polycomb group proteins first announced themselves in fruit flies, where researchers screening for mutants noticed animals with an extra sex comb, a structure normally restricted to the front legs of males. At least sixteen of these genes were eventually identified as repressors of the homeotic, or Hox, genes, the ancient genetic address system that tells every cell where it sits along the body axis. The discovery revealed a profound biological principle: development depends not only on turning genes on, but on keeping the wrong genes firmly off. That principle has been conserved across hundreds of millions of years of evolution, and in vertebrates Polycomb proteins now appear central to cellular differentiation, proliferation, tissue integrity, homeostasis and stem cell renewal.

The molecular machinery is elegantly modular. Polycomb proteins assemble into two major multi-protein complexes, Polycomb Repressive Complex 1 and Polycomb Repressive Complex 2, each built around a catalytic core plus a variable set of accessory subunits. The catalytic heart of PRC1 is a heterodimer of RING1A or RING1B with one of six PCGF family members, and it stamps histone H2A with a single ubiquitin molecule at lysine 119, a mark abbreviated H2AK119ub. PRC2, whose catalytic engine is the methyltransferase EZH1 or EZH2 working alongside EED, SUZ12 and RBBP4/7, deposits di- and trimethyl marks on lysine 27 of histone H3, known as H3K27me2 and H3K27me3. The two complexes are deeply interdependent. The H3K27me3 mark laid down by PRC2 is recognised by the chromobox subunit of canonical PRC1, which is thereby recruited to the same target sites to spread H2AK119ub and lock genes into a repressed state. Variant PRC1 can operate in the opposite direction, ubiquitinating H2A first and pulling PRC2 in behind it, creating a self-reinforcing loop of silencing.

Studying these complexes in mammals is brutally difficult because embryos lacking Polycomb function typically die around gastrulation, long before organs form. This is precisely where the zebrafish, Danio rerio, earns its reputation as one of developmental biology’s most powerful model organisms. Introduced as a genetic system by George Streisinger in the 1970s, zebrafish combine optical clarity, rapid development, sexual maturity within twelve weeks of fertilisation, ex utero embryonic growth that allows manipulation from the single-cell stage, and clutches of hundreds of offspring per mating. Crucially, maternal gene products sustain the embryo up to roughly the 1000-cell stage, far longer than in mice, giving researchers an extended window to probe early gene regulation. Zebrafish embryos also survive early development even with severe cardiovascular defects, because they obtain oxygen by passive diffusion rather than relying solely on blood circulation, something no mammal can do.

The fish genome adds another layer of opportunity. Around 320 million years ago, the teleost lineage underwent a whole-genome duplication event, generating pairs of duplicate genes called ohnologs, of which roughly 15 to 20 percent were retained after rediploidization. In the zebrafish genome, thirteen Polycomb subunits exist as duplicated pairs, although some, including RING1A, PCGF2, PCGF3, RBBP7 and the accessory subunit HDAC2, were lost during rediploidization. Despite these losses, the diversity and complexity of Polycomb complexes remain intact, and researchers can now dissect their roles using an arsenal of techniques ranging from classical ENU mutagenesis and morpholino knockdowns to zinc-finger nucleases, TALENs and CRISPR/Cas9 gene editing.

When it comes to the heart, the review paints a picture of two complexes with distinct but complementary timing. Embryos lacking either PRC1 or PRC2 develop the striking ‘heart-strings’ phenotype: after 48 hours post-fertilisation, the heart fails to loop and instead remains a string-like tube rather than forming its two chambers, the atrium and ventricle. Loss of rnf2, the sole zebrafish homolog of PRC1’s enzymatic subunit, produces pericardial edema and this stringy heart. Yet early cardiac markers such as nkx2.5, tbx5 and hoxb5b appear at normal levels in the lateral plate mesoderm of rnf2 mutants, indicating that the heart field is specified correctly and that PRC1 acts later in the programme. RNA sequencing of individual mutant hearts by Chrispijn and colleagues showed that T-box transcription factors tbx2a, tbx2b, tbx3a and tbx5 become upregulated over time, while structural genes such as vmhc, myl7, myh6 and nppa are downregulated, pointing to a role for PRC1 in the structural maturation of the organ.

The finer details are remarkable. In homozygous rnf2(f5) mutants, genes involved in smooth muscle and skeletal muscle development, which normally participate in transforming the atrioventricular canal between the two chambers, become dysregulated. Transmission electron microscopy reveals misaligned cardiac sarcomeres, with closely packed myofibrils and abnormally spaced I-bands and Z-discs. Calcium imaging with the fluorescent dye fluo-4 AM shows a weakened calcium signal, and the result is a heart that contracts feebly. The expression of atrioventricular canal markers such as alcama, vcana and bmp4 becomes expanded and diffuse rather than tightly constricted. Together these data establish that PRC1 is essential for cardiac sarcomere assembly, the maintenance of contraction, and the proper constriction of the atrioventricular canal and its valves, largely by repressing inappropriate non-cardiac sarcomere genes. A second non-canonical PRC1 component, mga, adds another twist: morpholino knockdown of mgaa blocks heart tube looping, and the defect can be rescued by co-reducing gata4, whose transcript levels rise two- to four-fold in the mutant, showing that variant PRC1 fine-tunes this key regulator.

PRC2, by contrast, acts earlier. Maternal-zygotic ezh2 mutants, in which the methyltransferase is eliminated from both maternal and zygotic sources, show reduced numbers of nkx2.5-expressing cardiac progenitors, partially reduced chamber markers, and cardiomyocytes that wander away from the cardiac tube, leaving a small, tubular heart that never loops. The atrioventricular canal fails to form properly, possibly because of ectopic expression of has2, a gene that promotes cardiac cell migration. Chromatin immunoprecipitation sequencing shows that both ezh2 and the H3K27me3 mark are absent from heart-development genes such as tbx3a, tbx5 and isl1, and PRC1 peaks are missing from the same loci, consistent with PRC2 recruiting PRC1 to its targets. Intriguingly, ezh1 mutants show no heart defects despite a modest reduction in H3K27me3, and ezh1;ezh2 double mutants resemble ezh2 single mutants, demonstrating that ezh1 cannot compensate for ezh2 in the heart. Pharmacological inhibitors of EZH2, including GSK126, DZNep and PF-06726304 acetate, all induce heart edema, reinforcing the complex’s central role.

The pectoral fin story is equally compelling, and arguably more evolutionarily provocative. Zebrafish pectoral fins are the evolutionary relatives of tetrapod forelimbs, and although a fin and an arm look nothing alike, the underlying regulatory logic is deeply conserved. Retinoic acid, synthesised by the enzyme aldh1a2, initiates the programme by activating wnt2b and tbx5 in the lateral plate mesoderm; tbx5 then triggers the Fgf cascade that drives outgrowth, while shh, hox genes and the apical ectodermal ridge pattern the bud along its anterior-posterior, dorsal-ventral and proximal-distal axes. When PRC1 or PRC2 function is lost, pectoral fins simply fail to form. In rnf2 mutants, early fin markers such as tbx5 and hand2 appear on schedule but vanish from the fin mesenchyme by 40 hours post-fertilisation, the Fgf targets fgf24 and fgf10 shrink to a small patch, and apical ectodermal ridge markers such as dlx2a, fgf8 and versican are absent, so fin initiation occurs but outgrowth collapses.

The epigenetic mechanism behind the missing fins involves a runaway retinoic acid signal. In rnf2 mutants, aldh1a2 is overexpressed and spreads beyond its normal posterior domain, while cyp26a1, which degrades retinoic acid, is reduced, producing excess retinoic acid that the embryo tries to metabolise by upregulating genes such as dhrs3. Hox genes normally restricted to the anterior body, including hoxc6a and hoxc8a, are ectopically expressed in the posterior and throughout the brain, and fin-specific hox expression of hoxa9b, hoxc8a and hoxd9a is lost while their axial domains expand. Patterning markers such as shh, msxc, eng1a and wnt7a are absent or reduced. Even partially suppressing retinoic acid with DEAB restores tbx5 and hand2 only partially, insufficient to restart fin outgrowth. In maternal-zygotic ezh2 mutants the same picture emerges: tbx5 is entirely missing, the hoxab cluster expands beyond its normal domain, and shh is absent from the fin bud. Because PcG-deficient zebrafish survive gastrulation where mice do not, this model opens a unique window onto how epigenetic silencing sculpts organs, and the authors argue that understanding these molecular pathways could ultimately illuminate congenital heart defects and limb abnormalities in humans, and point toward new therapeutic targets in regenerative medicine.

Subject of Research: Polycomb group protein-mediated epigenetic regulation of heart and pectoral fin development in zebrafish

Article Title: Epigenetic regulation in zebrafish development: the roles of polycomb group proteins in heart and pectoral fin development

Article References: Kavade, P. S., Parab, S. S., Capone, V., Carannante, D., Ambrosino, C., Altucci, L., & Carafa, V. (2024). Epigenetic regulation in zebrafish development: the roles of polycomb group proteins in heart and pectoral fin development. Epigenetics Communications, 4(1), Article 7. https://doi.org/10.1186/s43682-024-00030-y

Image Credits: AI Generated

DOI: 10.1186/s43682-024-00030-y

Keywords: Polycomb group proteins, PRC1, PRC2, zebrafish, epigenetics, heart development, pectoral fin development, Hox genes, histone modification, retinoic acid signaling, EZH2, developmental biology

Reactive oxygen species, or ROS, include free radicals such as superoxide and hydroxyl radicals as well as non-radical species like hydrogen peroxide and singlet oxygen. They arise from endogenous sources, chiefly mitochondrial metabolism, and from exogenous exposures including ultraviolet radiation, pollutants and drugs. In healthy cells, ROS act as essential signaling molecules, but when their production outpaces the cell’s antioxidant defenses, oxidative stress ensues. The review emphasizes that this imbalance is a hallmark of cancer cells, which typically generate more localized ROS than their normal counterparts due to elevated metabolic activity, mitochondrial dysfunction, oncogene activation and increased activity of oxidases, cyclooxygenases and lipoxygenases.

The consequences of excess ROS reach far beyond simple DNA breakage. Hydroxyl radicals can attack purines and pyrimidines, producing base modifications such as 8-oxo-7,8-dihydroguanine, one of the most reliable biomarkers of oxidative DNA damage. In normal human cells, 8-oxoGua adducts form at a rate of roughly one per 100,000 guanine residues, but in smokers this rate rises by more than 30 percent, and patients with lung, breast or prostate cancer show levels approximately 50 percent higher than healthy individuals. When repair mechanisms such as 8-oxoguanine glycosylase fail under persistent oxidative stress, mutations accumulate in critical genes including the RAS oncogene and the TP53 tumor suppressor, fueling genomic instability and malignant transformation.

Perhaps the most compelling section of the review concerns the epigenetic consequences of oxidative stress. ROS can drive both global DNA hypomethylation and site-specific hypermethylation of tumor suppressor gene promoters, a pattern reversal that is characteristic of cancer cells. The mechanistic links are intricate: oxidative stress alters the activity of DNA methyltransferases and histone modifiers, while depletion of glutathione reduces levels of S-adenosylmethionine, the universal methyl group donor required by DNA methyltransferases and histone methyltransferases. The oxidized base lesion 8-hydroxydeoxyguanosine interferes with methyltransferase binding to nearby cytosines, and hydroxylation of 5-methylcytosine to 5-hydroxymethylcytosine can reduce cytosine methylation by as much as 90 percent, according to work by Valinluck and Sowers cited in the review.

The authors also describe how hydrogen peroxide exposure recruits DNMT1 to damaged chromatin, where it joins complexes containing DNMT3B, sirtuins and members of the polycomb repressive complex 4, leading to silencing marks such as H3K27me3 and loss of active marks like H3K4me3 and H4K16ac. In kidney epithelial cells, chronic oxidative stress converted normal cells into cancerous ones, altering the expression of DNMT1, DNMT3a, MBD4, HDAC1, HMT1 and HAT1, and this malignant phenotype was partially reversed by treatment with the demethylating agent 5-aza-2′-deoxycytidine, underscoring the reversibility of the epigenetic machinery. Similar ROS-driven methylation changes have been documented for tumor suppressor genes such as RUNX3 in colorectal cancer and p16 in the progression from Barrett’s esophagus to esophageal adenocarcinoma.

ROS also orchestrate inflammation and immune evasion, two processes intimately entwined with epigenetics. Elevated ROS activates signaling cascades including NF-kB, MAPK, STAT3 and PI3K/AKT, promoting uncontrolled proliferation, angiogenesis and epithelial-mesenchymal transition, the program by which cancer cells acquire migratory and invasive properties. At the same time, ROS suppresses T cells and natural killer cells and skews macrophages toward an M2 phenotype that supports tumor vascularization. The Nrf2-KEAP1 pathway emerges as a double-edged sword: in normal cells, Nrf2 activation induces phase II detoxification and antioxidant enzymes, but in established cancers, sustained Nrf2 activation confers drug resistance and cellular immortality.

This duality is precisely where polyphenols display their therapeutic versatility. Compounds such as sulforaphane from cruciferous vegetables, curcumin from turmeric, epigallocatechin-3-gallate from green tea, genistein from soy, quercetin from onions and apples, and resveratrol from grapes and berries can each act as either antioxidant or pro-oxidant depending on context and concentration. As antioxidants, they activate Nrf2, promote its nuclear localization and stimulate the synthesis of glutathione S-transferase, heme oxygenase 1, NADPH quinone oxidoreductase, catalase and superoxide dismutase, thereby neutralizing ROS before they can damage the epigenome. As pro-oxidants, they raise ROS levels in drug-resistant cancer cells, inhibit Nrf2 and trigger apoptosis, sensitizing tumors to chemotherapy.

The epigenetic footprints of individual polyphenols are strikingly specific. Sulforaphane inhibited DNMT1 and DNMT3a in breast cancer cells, downregulated the telomerase reverse transcriptase hTERT, and remodeled chromatin at the hTERT promoter with increased H3K9 acetylation and decreased H3K9 and H3K27 trimethylation. Curcumin reversed hypermethylation of the glutathione S-transferase pi 1 promoter in MCF-7 breast cancer cells, re-expressed the hypermethylated BRCA1 gene through TET1-mediated demethylation, upregulated miR-15a and miR-16 to suppress the anti-apoptotic protein Bcl-2, and even reversed multidrug resistance in colorectal cancer cells by increasing ROS and diminishing Nrf2 expression. Genistein demethylated and reactivated p16INK4a and BTG3, while EGCG reactivated the silenced tumor suppressors p16INK4a, p21CIP/WAF1 and Wnt inhibitory factor 1 through promoter demethylation and histone acetylation changes.

The review is candid about the obstacles standing between laboratory promise and clinical practice. Most of the evidence comes from in vitro models, and animal studies must precede any prescription in clinical settings. Polyphenols suffer from poor water solubility and low oral bioavailability, although bio-based nanocarriers, including polymeric nanoparticles, nanocrystals, cyclodextrins, nano-caseins and electrospun nanofibers, are being developed to overcome these limitations. Dosing also demands caution: low concentrations of quercetin reduced ROS and antagonized the pro-oxidant mechanism of cisplatin in ovarian cancer cells, potentially undermining treatment, while high doses of green tea catechins have been associated with hepatotoxicity and worsened colon carcinogenesis in experimental animals. A study by Salvador and colleagues further showed that ROS-generating therapeutic agents can be more toxic to normal cells than to p53-negative tumors, highlighting the need to select agents and doses that spare healthy tissue.

Even with these caveats, the synthesis offered by Raina and colleagues is a compelling case that the epigenetic damage wrought by oxidative stress is not a one-way street. Because epigenetic modifications are inherently reversible, and because polyphenols can simultaneously scavenge ROS, modulate Nrf2, dampen NF-kB-driven inflammation and recalibrate DNA methyltransferases, histone acetyltransferases and histone deacetylases, these dietary compounds occupy a uniquely favorable position in the search for chemopreventive and chemosensitizing agents. The authors call for further research into the biological processes underlying ROS-induced epigenetic alterations, the interaction between polyphenols and the gut microbiome, and the precise concentrations required for anti-cancer effects. If those questions can be answered, the humble molecules found in broccoli, turmeric, green tea and grapes may yet earn a place alongside conventional therapies in the oncologist’s arsenal.

Subject of Research: ROS-induced epigenetic alterations in carcinogenesis and their reversal by dietary polyphenols

Article Title: Polyphenols reverse ROS induced epigenetic alterations in the process of carcinogenesis

Article References: Raina, R., Bhatt, R., Hussain, A., Afroze, N., & Hussain, A. (2025). Polyphenols reverse ROS induced epigenetic alterations in the process of carcinogenesis. Epigenetics Communications, 5(1), Article 5. https://doi.org/10.1186/s43682-025-00034-2

Image Credits: AI Generated

DOI: 10.1186/s43682-025-00034-2

Keywords: reactive oxygen species, polyphenols, epigenetics, carcinogenesis, Nrf2 pathway, DNA methylation, histone modification, sulforaphane, curcumin, EGCG, chemoprevention, oxidative stress

The classic experimental system for studying plant regeneration is the two-step tissue culture protocol. Explants, small pieces of plant tissue, are first placed on a callus induction medium containing a balanced ratio of the plant hormones auxin and cytokinin. Under these conditions, a subset of cells re-enters the cell cycle and proliferates into callus, a disorganized mass of cells that resembles lateral root primordia more than any mature tissue. When the callus is then transferred to a shoot induction medium rich in cytokinin, some cells adopt shoot identity and form a shoot apical meristem, the self-renewing stem cell hub that generates leaves, stems, and eventually flowers. Understanding the molecular switches that govern each of these transitions has become one of the central questions in plant developmental biology.

At the heart of the process lies hormonal control. Auxin, arguably the most influential player, drives dedifferentiation and callus formation, and its signaling operates through transport inhibitor response 1 receptors and auxin response factors that directly regulate reprogramming genes. Work in Arabidopsis thaliana has shown that auxin upregulates LEAFY COTYLEDON 1 and 2, transcription factors central to somatic embryogenesis. Studies in rice further revealed that local auxin gradients determine whether regenerating tissues form roots or shoots: high auxin with low cytokinin favors roots, while the opposite combination promotes shoots. PIN-FORMED transport proteins channel auxin directionally to specific sites, initiating de novo root organogenesis with remarkable spatial precision.

Cytokinin acts as auxin’s counterweight. It sustains cell division and, at high concentrations, promotes shoot primordia formation, while lower levels favor roots or somatic embryos. Cytokinin signaling flows through the Arabidopsis response regulator family, and type-B regulators such as ARR1, ARR2, ARR10, and ARR12 bind directly to the promoter of WUSCHEL, the master gene that maintains stem cell activity in the shoot apical meristem. Cytokinin also suppresses auxin accumulation through the YUCCA biosynthetic pathway, and remarkably, it reaches into the epigenetic machinery by modulating histone acetyltransferases, thereby making meristem maintenance genes more accessible for transcription. Mutations in type-B response regulators impair shoot regeneration, whereas overexpressing ARR12 enhances it, and WUSCHEL overexpression can rescue defective regeneration in double mutants, underscoring the centrality of this regulatory module.

Other hormones fine-tune the process. Gibberellins do not initiate reprogramming but sustain proliferation and maturation once cells have been redirected, promoting the elongation of somatic embryos into functional plantlets. Brassinosteroids, acting through the BRI1 receptor, synergize with auxin and cytokinin to boost cell division in callus and support organ formation. Abscisic acid, the stress hormone, plays the antagonist: elevated levels suppress cell division and organogenesis, inducing a dormancy-like state that prevents cells from reverting to totipotency under unfavorable conditions. ABA and cytokinin act antagonistically in shoot regeneration, illustrating how plants calibrate their regenerative responses against environmental risk.

Beneath the hormonal layer sits an elaborate gene regulatory network. Callus formation begins with the division of xylem pole pericycle cells, a process sharing molecular logic with lateral root initiation. The ALF4 protein initiates pericycle division, and root meristem genes including WOX5, WOX11, SCARECROW, PLETHORA 1 and 2, and SHORT ROOT become strongly upregulated. Mutant analyses confirm their importance: plt3plt5plt7 mutants lose shoot progenitor regeneration, and lbd16 mutants show reduced expression of WOX5 and PLETHORA genes with diminished shoot regeneration. On shoot induction medium, CUC1 and CUC2 transcription factors mark pre-meristematic zones, recruit PIN1 polarization, and activate SHOOT MERISTEMLESS to define future shoot progenitors, while ESR1 and ESR2 act upstream to promote the process. A transient root stem cell niche, marked by WOX5 expression, appears to be a prerequisite for the subsequent switch to shoot identity.

Small RNAs add another layer of temporal and spatial control. The miR156-SPL module explains why regeneration efficiency declines with plant age: juvenile plants with high miR156 repress SPL transcripts, keeping cytokinin responses active, whereas older plants accumulate SPL proteins that suppress type-B response regulators and reduce regeneration. miR165/166 promotes shoot formation, and the ARGONAUTE10 protein sequesters these microRNAs, with ago10 mutants showing elevated miR165/166 and enhanced regeneration. Meanwhile, a miR160-ARF10-ARR15 cascade fine-tunes the early balance between auxin and cytokinin signaling during callus formation, demonstrating how microRNA circuits integrate hormonal cues across developmental stages.

Perhaps the most striking insight from the review is that reprogramming is governed less by genetic change than by epigenetic rewiring. Global DNA hypomethylation is a hallmark of callus formation across species. In Arabidopsis, MET1 maintains CG methylation, and its loss increases WUSCHEL transcription and accelerates shoot formation; met1 mutants produce more shoots. Yet local hypermethylation also matters, silencing organ identity genes such as GSTU10, MAPK12, and BXL1 to help establish an undifferentiated cell mass. Histone marks are equally dynamic: H3K9ac and global H3 acetylation rise during callus formation, activating sugar metabolism, peptide signaling, and hormone transport genes while repressing photosynthesis programs. The demethylase JMJ30 removes repressive H3K9me3 marks from LBD16 and LBD29 promoters to enable callus formation, while ATX4 deposits activating H3K4me3 at shoot identity genes. Loss of the deacetylase HDA19 suppresses shoot regeneration, showing that both writing and erasing these marks is essential.

Proteomic studies, though still limited in number, reveal consistent patterns. Embryogenic cells accumulate heat-shock proteins as molecular chaperones, elevated glutathione-S-transferases that detoxify auxin, and cytoskeletal components such as tubulin and actin that establish cell polarity during division. Carbohydrate metabolism enzymes, photosynthetic proteins that appear as embryos mature, and nuclear import factors like importin all shift in abundance, reflecting the massive metabolic and architectural reorganization that accompanies the acquisition of pluripotency. Reactive oxygen species homeostasis intersects with auxin signaling during dedifferentiation, linking stress physiology directly to developmental potential.

New technologies are transforming the field’s resolution. Single-cell RNA sequencing has mapped the cellular composition of Arabidopsis and rice roots, revealed hormone signaling variation across cell clusters, and, when combined with ATAC-seq, identified WUSCHEL and DORNROESCHEN as crucial factors for regenerating mesophyll cells. CRISPR-based activation has been used to enhance somatic embryogenesis in tomato by targeting the SlWRKY29 gene, and CRISPR-dCas9 systems now allow precise transcriptional tuning of master regulators. Even artificial intelligence has entered the tissue culture lab: machine learning models, including neural networks and gradient boosting algorithms, have successfully predicted optimal media compositions for callus induction in carrot, micropropagation of lavender, and callogenesis in saffron, cutting both cost and experimental time. Together, these tools point toward a future in which regeneration efficiency can be engineered rather than stumbled upon, with implications for crop improvement, the propagation of recalcitrant and endangered species, and a deeper understanding of how any cell, given the right molecular instructions, can be persuaded to start over.

Subject of Research: Molecular mechanisms of cell fate reprogramming during de novo organogenesis in plants

Article Title: Understanding the molecular landscape of de novo organogenesis: insights into cell fate reprogramming

Article References: Nath, J., Kumari, A., Joshi, S., Gusain, S., Kumari, K., Yadav, S. K., & Joshi, R. (2025). Understanding the molecular landscape of de novo organogenesis: insights into cell fate reprogramming. Discover Biotechnology, 2(1), Article 27. https://doi.org/10.1007/s44340-025-00035-w

Image Credits: AI Generated

DOI: 10.1007/s44340-025-00035-w

Keywords: de novo organogenesis, cell fate reprogramming, auxin, cytokinin, WUSCHEL, epigenetics, DNA methylation, histone modification, callus, single-cell RNA sequencing, CRISPR, plant regeneration