The central obstacle to a cure is the remarkable persistence architecture of the hepatitis B virus. After the virus enters hepatocytes through the NTCP receptor, its relaxed circular DNA genome is repaired by host factors into covalently closed circular DNA, or cccDNA, a stable episomal minichromosome that persists in the nucleus and serves as the transcriptional template for all viral RNAs and proteins. In parallel, fragments of viral DNA integrate at random sites into the host genome as integrated DNA, or intDNA, which does not produce new virus but continuously secretes surface antigen and, in some cases, the oncogenic HBx protein, fueling immune tolerance and carcinogenesis. Current interferon-alpha and nucleos(t)ide analog therapies leave both reservoirs untouched, so viral rebound is common once treatment stops, and long-term nucleos(t)ide therapy carries risks such as nephrotoxicity.

The Groningen team frames the therapeutic landscape around two epigenetic strategies. The first, broad-acting epidrugs, globally modulate histone- and DNA-modifying enzymes; nine such agents are already FDA-approved in oncology. The second, epigenetic editing, fuses effector enzymes to programmable DNA-binding domains — zinc-finger proteins, transcription activator-like effectors, or deactivated CRISPR-Cas9 — to write repressive marks at specific viral loci. Because HBV gene activity on cccDNA is governed by histone acetylation and methylation, with HBx recruiting p300 to boost H3 and H4 acetylation while SETDB1-mediated H3K9me3 and PRMT1-mediated H4R3 methylation repress transcription, the viral minichromosome is an unusually tractable epigenetic target.

Among epidrugs, sirtuin 2 inhibitors have produced the most compelling results. The SIRT2 inhibitor AGK2 reduced HBV DNA, RNA, HBsAg and HBeAg by 15 to 30 percent in vitro and in HBV-transgenic mice without hepatotoxicity, and recent work showed the drug acts by recruiting repressive histone lysine methyltransferases to cccDNA, enriching H4K20me1, H3K27me3 and H3K9me3 while reducing RNA polymerase II occupancy. A more specific allosteric inhibitor, FLS-359, blocked the conversion of rcDNA into cccDNA entirely in primary human hepatocytes when given before infection, cutting cccDNA formation by more than half — though it was ineffective after infection was established, positioning it as a preventive rather than curative agent. Meanwhile, the DNA methyltransferase inhibitor 5-azacytidine reactivated interferon-stimulated genes and sensitized otherwise unresponsive cells to interferon-alpha, but it also raised NTCP expression and HBsAg levels, highlighting the double-edged nature of globally acting agents.

That lack of locus specificity is precisely what epigenetic editing is designed to fix, and the review catalogs five preclinical editing studies organized around two approaches. Indirect editing directs a Krüppel-associated box, or KRAB, repressor domain to key HBV regulatory elements; KRAB acts as a scaffold recruiting histone deacetylases, lysine methyltransferases and heterochromatin proteins to shut down transcription. Direct editing instead guides DNA methyltransferases such as DNMT3a or the bacterial M.SssI enzyme to CpG islands overlapping viral promoters, writing de novo methylation that can be mitotically inherited. Both routes reduced HBV RNA, DNA and antigens from cccDNA and, in some contexts, from intDNA, with far better specificity than epidrugs.

The zinc-finger platform, the most rigorously studied to date, illustrates both the promise and the limits. Zhao and colleagues built a six-finger ZFP-KRAB artificial transcription factor targeting the X gene enhancer that cut HBx RNA by roughly 60 percent in Hep3B cells, a line carrying integrated viral DNA. Luo and colleagues achieved the most durable result, with HBV DNA falling to about one-third of control levels in transgenic mice by day seven and remaining significantly lower at day 28, although HBsAg was unchanged. The direct approach by Xirong and colleagues, fusing DNMT3a to a ZFP targeting the X promoter CpG island, lowered HBsAg by 58 percent in mice and 90 percent in cells, with confirmed methylation at seven CpG sites — but the effect reversed by day 20, likely through passive or TET-mediated demethylation. A single TALE-based study by Bloom and colleagues delivered the fastest repression, cutting HBsAg and HBV RNA by 80 percent in Huh7 cells within 48 hours and by 80 to 95 percent in mice within five days, though attribution to KRAB was complicated by the absence of a TALE-only control.

The dCas9 platform, which replaces costly protein engineering with inexpensive single-guide RNAs, is now moving fastest toward patients. The only peer-reviewed dCas9 data for HBV come from a doctoral thesis by Rendón, who fused M.SssI to dCas9 and targeted the conserved CpG islands adjacent to the C and S gene promoters, achieving 2 to 15 percent increases in methylation that downregulated C and S gene transcription but faded within 48 hours. Nevertheless, two clinical trials of dCas9 epigenetic editing for hepatitis B are already underway — NCT06745973 and NCT06671093 — and parallel successes silencing HIV provirus with dCas9-KRAB, plus the demonstration that combined KRAB and DNA methyltransferase editors such as CRISPRoff produce durable, heritable silencing, provide strong proof of concept. Notably, dCas9 editing avoids the genotoxicity risk that shadows nuclease-based CRISPR-Cas9 approaches, whose guide RNAs can partially match human genomic sequences and whose double-strand breaks at the many, variable intDNA loci raise off-target mutagenesis concerns.

Industry has now taken up the challenge with three programs. Tune Therapeutics’ Tune-401, a liver-targeting lipid nanoparticle carrying a guide RNA to CGI2 and mRNA encoding a dCas9 fused to a methyltransferase and an undisclosed repressor, reported 99.99 percent repression of cccDNA-derived HBV RNA in primary human hepatocytes, strong repression of RNA and HBsAg from intDNA in Hep3B cells, epigenetic marks persisting through 275 rounds of cell division over 550 days, minimal off-target effects by RNA-seq, and a favorable safety profile in non-human primates — though these results have not yet been peer reviewed. A Phase 1b trial is recruiting in Hong Kong, Moldova and New Zealand. EpiGENIC’s Epi-003, a similar LNP platform, has entered a Phase 1 trial in China, and nChroma Bio’s CRMA-1001 has shown greater than 99 percent HBsAg reduction sustained for six months in preclinical models. A related Omega Therapeutics trial using LNP-delivered ZFP editors in liver cancer further validates liver-directed epigenetic delivery.

Significant hurdles remain before any of this translates into a functional cure. HBV genomes differ by more than 8 percent at the nucleotide level across genotypes, and the error-prone reverse transcriptase plus the chaotic nature of integration create sequence variability that could blunt editing efficacy, demanding bioinformatic surveillance and possibly personalized designs. Heterochromatin spreading from KRAB editors could silence neighboring host genes at unpredictable intDNA sites, while complete silencing of all antigen sources raises the paradox that eliminating HBsAg might allow exhausted immune cells to miss residual infection, arguing for integration with immune-reactivation strategies. Model limitations compound the problem: primary human hepatocytes rarely form intDNA, Hep3B lacks infection dynamics, and the field needs 3D organoid and humanized mouse systems that capture both reservoirs. Cost and equity loom as well, since most patients live in Southeast Asia and Sub-Saharan Africa while comparable gene therapies have been priced near two million dollars. Still, the authors conclude that with optimized effector combinations, LNP-based hit-and-run delivery — already shown to sustain liver gene silencing for over a year in mice and over 90 days in non-human primates in PCSK9 studies — and clinically relevant models, precision epigenetic editing could deliver what nucleos(t)ide analogs never could: a stable, reversible, non-genotoxic functional cure for chronic hepatitis B.

Subject of Research: Epigenetic editing strategies to silence hepatitis B virus cccDNA and integrated DNA for a functional cure of chronic hepatitis B

Article Title: The promise of epigenetic editing strategies in functionally curing chronic hepatitis B virus infections

Article References: The promise of epigenetic editing strategies in functionally curing chronic hepatitis B virus infections. (n.d.). https://doi.org/10.1186/s43682-025-00041-3

Image Credits: AI Generated

DOI: 10.1186/s43682-025-00041-3

Keywords: epigenetic editing, chronic hepatitis B, hepatitis B virus, cccDNA, integrated HBV DNA, dCas9, zinc-finger proteins, TALEs, KRAB, epidrugs, lipid nanoparticles, functional cure

The research, led by Yingxiang Wang, Cong Wang and Chenjiang You of South China Agricultural University together with Hong Ma of Pennsylvania State University, with Zhiyu Chen, Meiling Li and Shiqian Yang as co-first authors, focuses on two related proteins in the reference plant Arabidopsis thaliana known as C3H14 and C3H15. These proteins belong to the tandem CCCH zinc-finger family, a class of RNA-binding proteins characterised by cysteine and histidine residues that coordinate a zinc ion to form a compact finger-shaped domain capable of gripping specific RNA sequences. Earlier work had shown that C3H14 and C3H15 act with overlapping functions in secondary wall thickening and anther development, and that C3H15 contributes to thermotolerance through combined transcriptional and post-transcriptional regulation. The new study now assigns these proteins a role at the very heart of plant reproduction: the surveillance of meiocyte identity.

The team began by examining Arabidopsis plants in which both C3H14 and C3H15 had been disabled. Single mutants were largely fertile, reflecting the redundancy that has long been suspected for this gene pair, but the double mutant Atc3h14 Atc3h15 told a dramatically different story. Its meiocytes entered the meiotic programme but then arrested, unable to complete the specialised cell divisions that generate microspores. Centromere-specific fluorescence in situ hybridisation and immunostaining for meiotic markers such as gamma-H2AX, HEI10, SYN1 and ZYP1 revealed a prophase-like chromatin state that failed to progress normally. Crucially, the arrest was not the consequence of a failure in any single meiotic event: the researchers crossed the double mutant with defective versions of SPO11, DMC1, RAD51, SYN1 and PP2A regulatory subunits, and the sterile phenotype persisted regardless, indicating that the block sits upstream of the core meiotic machinery rather than within it.

To understand what had gone wrong at the molecular level, the investigators isolated meiocytes by laser capture and compared their transcriptomes with those of wild-type cells at the same developmental stage. The result was a molecular identity crisis. In the mutant meiocytes, messenger RNAs that should be confined to archesporial cells and to the tapetum accumulated ectopically and persistently. Genes required for initial sporogenous cell differentiation, including SPL/NZZ, AGAMOUS and EMS1/TPD1 signalling components, together with tapetum-preferential genes such as AMS, MYB80 and enzymes of sporopollenin biosynthesis, were all present at abnormal levels inside cells that were supposed to be dedicated exclusively to meiosis. RNA in situ hybridisation confirmed that transcripts like AtA6, AtZYP1a, AtDMC1 and the NADPH oxidase gene AtRBOHE were expressed ectopically or for too long in the mutant germ cells.

The consequences of this transcriptomic contamination proved lethal to the meiocytes. The ectopic expression of RBOH-type NADPH oxidases, which catalyse the production of reactive oxygen species, was followed by a measurable burst of ROS inside the mutant meiocytes, detected with the fluorescent probe H2DCF-DA. Soon after, TUNEL assays revealed the hallmarks of programmed cell death, and the dying meiocytes did so in synchrony with the tapetum, a tissue that normally dies on a precise schedule to release mature pollen. In other words, when the boundary between germline and soma blurred at the RNA level, both compartments perished together, leaving the anther sterile. Genetic analysis reinforced the point: individually overexpressing single misplaced genes in meiocytes, or removing them from the mutant background, did not rescue the phenotype, showing that the defect stems from a global failure of RNA homeostasis rather than the action of one rogue transcript.

Mechanistically, the study connected C3H14 and C3H15 directly to the cell’s mRNA disposal machinery. Immunoprecipitation followed by mass spectrometry identified over a thousand proteins that associate with C3H14 in meiocytes, and among the enriched categories were components of processing bodies, the cytoplasmic condensates where mRNAs destined for destruction are concentrated, and the conserved CCR4-NOT deadenylase complex, which trims the polyadenylated tail of mRNAs to initiate their decay. RNA electrophoretic mobility shift assays demonstrated that both C3H14 and a truncated form of C3H15 bind directly and specifically to AU-rich elements in the 3 prime untranslated regions of target mRNAs, including those of AGAMOUS, SPL/NZZ and RBOHF, and that this binding is abolished when the AU-rich motifs are mutated to cytosine-rich sequences. The picture that emerges is of a sequence-specific RNA-binding surveillance system that recruits general decay factors to eliminate transcripts inappropriate for the meiocyte state.

Perhaps the most compelling evidence for the importance of this mechanism is its evolutionary reach. Using phylogenetic analysis, the authors traced C3H14 and C3H15 homologues across the flowering plants, noting that independent whole-genome duplication events in dicots and monocots generated paired paralogues in both major lineages. When the team used CRISPR-Cas9 to knock out the homologous gene pairs in soybean, GmC3H15a and GmC3H15b, and in rice, OsC3H9 and OsC3H39, the resulting double mutants displayed the same devastating phenotype as the Arabidopsis originals: meiotic arrest, abnormal reactive oxygen species accumulation in meiocytes and complete male sterility. The conservation of this surveillance mechanism across such distantly related crop species suggests that post-transcriptional RNA elimination is not an idiosyncrasy of Arabidopsis but a fundamental requirement of meiocyte identity throughout the angiosperms.

The findings also resonate with parallel discoveries in animals. In mouse oocytes, the deadenylase CNOT6L couples the selective degradation of maternal transcripts to meiotic cell cycle progression, while the mRNA decay activator ZFP36L2 mediates chromatin modification and global transcriptional silencing in growing oocytes, and TTP-family proteins in mammals bind AU-rich elements to nucleate processing body formation. That plants and animals, whose last common ancestor was unicellular, have independently deployed RNA-binding zinc-finger proteins and deadenylase complexes to safeguard their germlines speaks to a deep logic of cell fate maintenance: keeping a differentiated state stable requires continuous, active removal of the molecular memory of alternative identities, not merely the activation of the appropriate transcriptional programme.

Beyond its conceptual significance, the work carries practical implications for agriculture. Male sterility is a cornerstone of hybrid seed production, and understanding the genetic switches that govern meiocyte identity offers new avenues for engineering controllable sterility systems in crops. The genes identified here, together with the AU-rich RNA elements they recognise and the decay complexes they recruit, constitute a potential toolkit for designing fertility systems that can be switched on or off at will. As global food demand rises and hybrid breeding expands to new species, mechanisms that were once studied purely out of curiosity about how a cell knows what it is may prove to be among the most valuable assets in the plant breeder’s arsenal. For now, C3H14 and C3H15 stand as elegant proof that in biology, knowing who you are sometimes means destroying every message that says otherwise.

Subject of Research: Post-transcriptional maintenance of meiocyte identity by the zinc-finger proteins C3H14 and C3H15 in flowering plants

Article Title: Zinc-finger proteins C3H14 and C3H15 maintain meiocyte identity in flowering plants

Article References: Chen, Z., Li, M., Yang, S., Li, K., Xin, Y., Liu, X., Guo, Y., Ma, H., You, C., Wang, C., & Wang, Y. (2026). Zinc-finger proteins C3H14 and C3H15 maintain meiocyte identity in flowering plants. Nature Plants. https://doi.org/10.1038/s41477-026-02401-w

Image Credits: AI Generated

DOI: 10.1038/s41477-026-02401-w

Keywords: meiocyte identity, C3H14, C3H15, zinc-finger proteins, mRNA decay, anther development, meiosis, tapetum, reactive oxygen species, programmed cell death, Arabidopsis thaliana, plant reproduction

RNA-binding proteins are pivotal to cellular function, mediating processes ranging from RNA splicing and transport to translation regulation. Central to their function is the ability to discriminate specific RNA sequences while maintaining flexible binding dynamics. Traditionally, this has been a perplexing challenge given the presence of both highly ordered domains and disordered regions within these proteins. The recent work focusing on the FUS protein provides critical insight into how these divergent structural elements cooperate to achieve precise yet adaptable RNA binding.

The FUS protein is particularly notable due to its association with neurodegenerative disorders and its multifaceted role in gene regulation. Its architecture comprises a single-stranded RNA-binding RanBP2-type zinc finger domain, a well-defined structural motif responsible for sequence-specific recognition, coupled with extensive intrinsically disordered regions that lack fixed conformation. Leveraging molecular dynamics simulations enhanced by advanced sampling techniques, the researchers meticulously characterized how these domains engage with RNA sequences.

The simulations targeted a short RNA oligonucleotide containing a known FUS target sequence, GGU, to dissect the nature of the protein-RNA interface. What emerged was a compelling model involving two binding modalities: one dominated by the ZnF domain alone and another, more energetically favorable state where the disordered regions augment the interaction. Surprisingly, while the ZnF domain offered sequence specificity, its affinity was relatively modest when acting in isolation.

Intrinsically disordered regions, far from being passive linkers, actively participate in binding by making mostly non-specific contacts governed by electrostatic attractions to the phosphate backbone of the RNA. These transient yet multivalent interactions were shown to lower the dissociation constant significantly, effectively doubling the protein’s affinity for the RNA molecule. Furthermore, the IDRs induce conformational distortions in the RNA backbone, serving to stabilize the overall complex beyond the contribution of ZnF alone.

The integration of these dual binding modes results in a protein-RNA complex with an affinity nearly tenfold greater than that of the zinc finger domain alone. This cooperative binding not only enhances stability but also imparts the adaptability necessary for proteins to function in the dynamic intracellular environment. The study’s authors suggest that this mechanism may extend to a broader family of nucleic acid-binding proteins that possess similar domain architectures.

Importantly, the team’s sequence analysis of various RNA-binding proteins with associated IDRs revealed that the dual-binding model is likely a widespread phenomenon, indicating an evolutionarily conserved strategy to balance specificity and flexibility. This challenges prior assumptions that disordered regions serve purely structural or regulatory roles without direct involvement in molecular recognition processes.

Professor Akio Kitao, leading the study, emphasizes the active role of IDRs, “Our data suggest that these disordered segments are integral to RNA recognition and binding, vastly influencing interaction kinetics and affinity. This redefines our conceptual framework of protein-RNA recognition beyond static domain-specific contacts.” Such insights underscore IDRs as crucial functional elements rather than mere passive connectors, reshaping their perceived importance in molecular biology.

From a methodological standpoint, the study showcases the power of combining molecular dynamics with enhanced sampling to overcome the sampling limitations typically encountered in simulating flexible protein regions. This approach enabled capturing transient, non-specific interactions and subtle conformational changes previously difficult to observe, highlighting the value of computational simulation in revealing biomolecular mechanisms at atomic resolution.

The implications of these findings are profound, offering potential targets for rational drug design focused on modulating protein-RNA interactions. Therapeutics that can mimic or disrupt these dual binding modes may precisely influence gene regulatory pathways implicated in diseases such as amyotrophic lateral sclerosis (ALS) and certain cancers, where FUS and similar proteins are pivotal players.

Looking forward, the researchers plan to extend their investigations to explore whether post-translational modifications of intrinsically disordered regions influence their RNA-binding characteristics. Such modifications could dynamically modulate protein activity and specificity, contributing further complexity to gene regulation and cellular response mechanisms in health and disease.

This pioneering study provides a vital new paradigm in understanding how RNA-binding proteins combine structured and unstructured domains to fine-tune interactions at the molecular level. It marks a significant stride towards deciphering the sophisticated language of nucleic acid recognition, with wide-reaching implications for molecular biology, bioengineering, and therapeutic development.

Subject of Research:
Article Title: RNA Binding Mechanism of the FUS Zinc Finger in Concert with Its Flanking Intrinsically Disordered Region
News Publication Date: August 11, 2025
Web References: https://doi.org/10.1021/acs.jcim.5c01059
References: Journal of Chemical Information and Modeling, Volume 65, Issue 15, August 11, 2025
Image Credits: Institute of Science Tokyo, Japan
Keywords: Molecular biology, RNA-binding proteins, Zinc fingers, Intrinsically disordered regions, Protein-RNA interaction, Molecular dynamics simulation, RNA structure, Protein conformation, Gene regulation, Biomolecular recognition, Neurodegenerative diseases, Computational modeling