Enhancers are regulatory DNA elements that can be located thousands or even hundreds of thousands of bases away from the promoters of the genes they regulate. Their activity depends in part on the formation of three-dimensional chromatin loops that allow enhancer-bound regulatory factors to contact promoter-bound transcription machinery. Although researchers have known that many enhancers produce short-lived transcripts called enhancer RNAs, or eRNAs, the precise role of these molecules has remained difficult to define. Promoter regions also generate nascent RNAs as transcription begins, creating an opportunity for RNA molecules from the two regulatory regions to interact. The new study proposes that hnRNPK recognizes and organizes these emerging transcripts, converting RNA–RNA contacts into a stable molecular bridge between enhancer and promoter.

Rather than functioning only as a conventional RNA-binding factor, hnRNPK appears to provide an architectural framework for transcriptional regulation. The researchers found that it binds preferentially to nascent RNAs produced from active enhancers and promoters. These RNAs are generated locally while the relevant chromatin regions are being transcribed, placing them close to the DNA sites that need to communicate. By associating with both classes of transcript, hnRNPK may concentrate enhancer- and promoter-derived RNAs in the same molecular environment. Such proximity could stabilize interactions between the RNAs and help fold the intervening chromatin into a loop. The model offers a mechanistic explanation for how transient RNA molecules can influence a much larger and more persistent structure in the genome.

The study further indicates that hnRNPK can assemble into phase-separated condensates. These are dynamic, concentrated compartments formed when proteins and nucleic acids cluster through numerous weak interactions rather than through a single permanent molecular bond. Phase separation is increasingly recognized as a way for cells to organize biochemical reactions without enclosing them in a membrane. In the case of hnRNPK, the condensates were described as containing cavities that encapsulate RNA polymerase II. This arrangement could create a specialized transcriptional environment in which regulatory RNAs, chromatin-associated factors and the polymerase are brought together at high local concentrations, making productive transcription more likely.

A particularly important feature of the proposed mechanism involves the RPB3 subunit of RNA polymerase II. The investigators report that hnRNPK interacts with Pol II through RPB3, providing a direct protein-based route for incorporating the transcriptional enzyme into hnRNPK condensates. Once Pol II is concentrated within these assemblies, enhancer-associated complexes may be able to influence its delivery to nearby or physically connected promoters. The researchers suggest that hnRNPK dimerization could contribute to this process: one portion of an hnRNPK assembly could engage enhancer-derived RNA and regulatory factors, while another connects with promoter-associated RNA and Pol II. In this framework, enhancer–promoter looping is not merely a passive consequence of chromatin folding but an active process coordinated by RNA-binding proteins.

This mechanism also helps explain why enhancer transcription may be functionally important even when eRNAs are rapidly degraded and do not encode proteins. Their role may depend less on their final abundance than on their production at the correct genomic location and time. As these RNAs emerge from the DNA, they can serve as temporary molecular signals or scaffolds for proteins such as hnRNPK. The resulting assemblies could integrate several signals at once, including the identity of the enhancer, the activity state of the promoter and the availability of Pol II. By linking these signals, hnRNPK may help ensure that developmental genes are activated only when the appropriate regulatory elements are engaged.

The biological significance of the findings was tested using a mutation in hnRNPK associated with Au–Kline syndrome, a rare developmental disorder. The mutation, designated c.953+1dupG, altered the physical properties of hnRNPK condensates. Instead of remaining liquid-like and dynamic, the mutant condensates became more gel-like. Liquid-like condensates can continuously exchange their molecular components with the surrounding nucleus, allowing them to assemble, reorganize and dissolve as transcriptional requirements change. A gel-like state may restrict this exchange, trapping components or preventing the rapid rearrangements required for enhancer–promoter communication. The results suggest that the material state of a condensate is not a cosmetic property but a critical determinant of gene regulation.

Knock-in mice carrying the disease-associated mutation developed developmental defects, providing evidence that the molecular changes have consequences at the level of the organism. Fibroblasts derived from these animals showed reduced enhancer–promoter looping and diminished recruitment of Pol II to the promoters of key developmental genes. These observations connect the mutation’s effects across several scales: it changes the physical behavior of hnRNPK condensates, weakens three-dimensional contacts between regulatory DNA elements, reduces transcriptional machinery at gene promoters and ultimately contributes to developmental abnormalities. The findings are consistent with the idea that many congenital disorders may result not only from loss of a protein’s biochemical activity but also from changes in the dynamics and material properties of nuclear assemblies.

The work presents hnRNPK as a general structural regulator of gene expression rather than a factor restricted to a small set of specialized genes. By mediating interactions between enhancer and promoter RNAs, organizing phase-separated condensates and engaging Pol II through RPB3, hnRNPK may provide a common platform for communication across the genome. The proposed model also broadens the view of noncoding transcription: RNAs produced from regulatory DNA may act as active architectural components, even when they do not persist as stable cellular molecules. Further research will be needed to determine how hnRNPK selects particular RNA sequences, how its condensates are regulated by signaling pathways and whether similar defects in condensate dynamics contribute to other human diseases. For now, the study identifies a molecular link between nascent RNA, chromatin looping and transcriptional activation, revealing how the genome’s distant control elements can work together with remarkable precision.

Subject of Research: hnRNPK-mediated enhancer–promoter looping, RNA–RNA interactions, phase-separated condensates and RNA polymerase II recruitment

Article Title: hnRNPK condensates facilitate enhancer–promoter looping and RNA polymerase II recruitment

Article References: Ye, R., Zhao, H., Chen, J. et al. hnRNPK condensates facilitate enhancer–promoter looping and RNA polymerase II recruitment. Nat Genet (2026). https://doi.org/10.1038/s41588-026-02710-y

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41588-026-02710-y

Keywords: hnRNPK, enhancer RNA, promoter RNA, enhancer–promoter looping, RNA polymerase II, phase separation, condensates, chromatin architecture, Au–Kline syndrome, gene regulation

NF-κB transcription factors form dimers from five key subunits—RelA (p65), RelB, c-Rel, NF-κB1 (p50/p105), and NF-κB2 (p52/p100)—allowing functional versatility in gene regulation. The dynamic interplay between these subunits and AR proteins orchestrates the fine-tuning of downstream transcription, a process now known to be heavily influenced by the ankyrin repeat domains acting as versatile protein–protein interaction modules. This structural motif appears instrumental in mediating not only inhibitory control but also chromatin remodeling and transcriptional specificity.

The IκB family, long recognized for its inhibitory regulation of NF-κB, is itself a fertile ground of AR domain-containing proteins. These encompass precursor proteins like p100 (IκBδ) and p105 (IκBγ), classical cytoplasmic inhibitors—IκBα, IκBβ, and IκBε—and the more recently appreciated nuclear IκBs, including Bcl-3, IκBζ, IκBNS, and IκBη. Each harbors six to eight ankyrin repeats that directly engage NF-κB dimers, thereby orchestrating nuanced regulatory outcomes. Intriguingly, these interactions transcend mere sequestration, as nuclear IκBs participate actively in transcriptional complexes to either repress or promote gene expression.

Among nuclear IκBs, IκBζ forms a transcriptionally active complex with p50 and p52 NF-κB subunits on specific target genes such as Lcn2, employing a critical aspartate residue within its first ankyrin repeat for the interaction. The nuanced recognition of specific NF-κB subunits highlights the precision of AR-mediated binding, suggesting architectural adaptability encoded within these motifs. Bcl-3 further exemplifies the multifaceted nature of these interactions, stabilizing p50 homodimers on DNA and preventing their ubiquitination, thereby modulating inflammatory gene expression. Structural studies reveal that Bcl-3 extensively contacts ARs 1, 6, and 7 of p50, underscoring the spatial specificity inherent in AR domain engagements.

IκBη extends this paradigm, utilizing its eight ankyrin repeats to bind p50, a process integral to its nuclear localization and function. This emphasizes that ARs not only mediate protein–protein interactions but can also influence subcellular distribution, offering a dual regulatory axis in transcriptional control. Collectively, these insights redefine nuclear IκBs from passive inhibitors to active transcriptional co-regulators, intricately sculpting NF-κB-driven gene expression.

Beyond classical NF-κB regulators, the oncogenic AR protein p28GANK shines as a compelling antagonist of NF-κB activity. Overexpressed in hepatocellular carcinoma, p28GANK contains seven ankyrin repeats structurally reminiscent of IκBs and exerts profound effects on NF-κB RelA (p65) subunit activity. Contrasting mechanistic reports converge on its ability to bind RelA via these repeats, suppressing its transcriptional activity through different molecular routes. One pathway involves modulation of RelA acetylation levels by recruiting the deacetylase SIRT1, dampening transcription without affecting nuclear translocation or DNA binding. Alternatively, other evidence indicates p28GANK enforces cytoplasmic retention of RelA by exporting it through a CRM-1-dependent pathway, effectively sequestering NF-κB from chromatin. This duality underscores the functional versatility provided by the ankyrin repeat scaffold in modulating key oncogenic signaling molecules.

The ASPP family, encompassing ASPP1, ASPP2, and the inhibitory iASPP, further exemplify AR-domain-mediated regulation at the interface of apoptosis and inflammation. Characterized by their C-terminal proline-rich region, four ankyrin repeats, and SH3 domain, these proteins utilize their ANK-SH3 composite to engage the p65 subunit of NF-κB. Through these interactions, ASPP2 can interface with NF-κB pathways, integrating apoptotic control with inflammatory signaling, a nexus vital for cellular fate decisions in stress and disease contexts. The inhibitory iASPP likewise binds p65, indicating that modulation by AR proteins spans activation to suppression within NF-κB-driven transcription.

Intriguingly, the NF-κB family itself is autoregulatory through its ankyrin repeats. The ubiquitin ligase KPC1 targets the AR domain of NF-κB precursor p105, enhancing its ubiquitination and limiting proteasomal processing into p50. This regulatory mechanism influences the balance of NF-κB dimers and downstream gene expression, impacting tumor suppressor expression and immune cell recruitment. The capacity of AR domains within NF-κB proteins to attract ubiquitin ligases reflects a sophisticated self-modulatory feedback controlling signaling amplitude and duration.

Turning attention to Notch signaling, the Notch intracellular domain (NICD) features its own cluster of seven ankyrin repeats essential for transcriptional activation. NICD interacts directly with the transcription factor RBPJ via its AR domain, initiating expression of key downstream genes such as Dll4, establishing positive feedback loops that underpin cell fate determination during development and angiogenesis. This interaction is finely modulated by another AR domain-containing protein, GIT1, which competes with NICD for RBPJ binding, inhibiting the Dll4-Notch1 axis in stalk cells. Such competition preserves cellular heterogeneity and supports angiogenic sprouting, highlighting AR domains as dynamic modules regulating signal flux beyond simple activation.

Notably, the gene NRARP, itself a Notch target, encodes a protein comprising three ankyrin repeats that extend the NICD ankyrin repeat stack upon forming a tripartite complex with NICD1 and RBPJ. This extension acts as a negative feedback loop, tempering Notch signaling output and illustrating how AR domain architecture can shape transcription factor complex conformation and function. This mechanistic insight into NRARP’s role completes a feedback circuit integral for fine-tuning vascular development.

This emerging paradigm underscores ankyrin repeats as modular units of regulation transcending canonical structural roles. Their presence across diverse proteins—ranging from classical inhibitors like IκBs to oncoproteins like p28GANK, and signaling mediators like NICD and NRARP—demonstrates a conserved evolutionary strategy to exploit repeat motifs for dynamic protein interactions, subcellular localization, and transcriptional control. Such versatility grants AR-containing proteins the ability to govern multiple signaling pathways simultaneously, making them prime candidates for therapeutic targeting in inflammation, cancer, and developmental disorders.

Given the ubiquity and functional diversity of ankyrin repeats, future research will undoubtedly uncover novel AR-containing players and mechanisms in epigenetic and transcriptional regulation. Structural biology combined with systems-level analysis of AR-mediated interactomes promises to reveal comprehensive networks that govern cellular identity and response, providing unprecedented opportunities to manipulate these pathways in disease intervention. The exquisite specificity and adaptability of AR domains offer templates for designing small molecules or biologics that modulate protein–protein interactions currently deemed undruggable.

As our understanding expands, the convergent roles of ankyrin repeat proteins in both NF-κB and Notch signaling pathways underscore the integrative nature of cellular signaling hubs. They act not only as structural motifs but as finely tuned regulatory elements that determine the specificity, timing, and magnitude of transcriptional responses. This knowledge pivots ankyrin repeats from peripheral structural components to central regulatory nodes with broad impact on health and disease.

In summary, the intricate dance of ankyrin repeat-containing proteins in modulating transcription factors like NF-κB and NICD reveals a landscape of complex protein interaction networks vital for cellular regulation. Their modulation of gene expression networks implicates these AR modules as keystones in the balance between homeostasis and pathology. Unlocking their mechanistic secrets heralds a new chapter in molecular biology, where precise control over these repeat domains might pave the way for novel therapies across a spectrum of inflammatory, oncogenic, and developmental diseases.

—

Subject of Research: Ankyrin repeat-containing proteins and their roles in epigenetic and transcriptional regulation.

Article Title: The role of ankyrin repeat-containing proteins in epigenetic and transcriptional regulation.

Article References: Wu, M., Zhao, Y., Yang, J. et al. The role of ankyrin repeat-containing proteins in epigenetic and transcriptional regulation. Cell Death Discov. 11, 232 (2025). https://doi.org/10.1038/s41420-025-02519-4

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41420-025-02519-4