Engineering chromatin loops to control cell fate: LoopID reveals catalytic-independent functions of epigenetic regulators

Abstract

Enhancer–promoter (E-P) interactions are central to cell-type-specific transcriptional programs, yet the molecular machinery that establishes and maintains these loops has remained poorly defined. A recent study by Jiang et al, published in Nature Genetics, presents a series of transformative discoveries that redefine our understanding of E-P interactions and their role in gene regulation and cell fate determination. The research introduces LoopID, a chromatin-interaction-based proteomic platform that, for the first time, enables systematic identification of protein components, termed the “looposome,” localized specifically at chromatin looping anchors. Using LoopID, they profile the “looposome” in mouse embryonic stem cells (ESCs) and uncover an unexpected, catalytic-independent role for the histone demethylase JMJD2 (KDM4) in organizing chromatin architecture through phase-separated condensates. Beyond mechanism, the study demonstrates that targeted assembly of JMJD2 condensates at defined genomic loci can engineer E-P interactions driving cellular reprogramming toward pluripotent and two-cell-like states. Together, these findings represent not only a major technical advance but also a conceptual leap—establish LoopID as a foundational technology for dissecting chromatin structure, introduce a new conceptual framework for epigenetic regulators as structural organizers, and provide a powerful strategy to manipulate cell fate by rewiring three-dimensional (3D) genome architecture.

Relevance to Cell Transplantation and Regenerative Medicine

The ability to engineer specific enhancer–promoter (E-P) interactions and control gene expression programs offers a powerful tool for cellular reprogramming. Manipulating the phase separation of proteins like JMJD2 could generate specific cell types efficiently, overcoming issues of donor scarcity and immune rejection in cell-based therapies. LoopID also advances the characterization of chromatin loop machinery in different cell types and disease states, potentially identifying new therapeutic targets.

Gene expression regulation is fundamental to biological processes, and E-P loops are crucial for orchestrating cell-type-specific transcriptional programs. These loops bring enhancers and promoters into close proximity, facilitating transcription. Three-dimensional (3D) genome organization is now recognized as a fundamental layer of gene regulation^1,2. Enhancers communicate with their associated promoters through chromatin loops, enabling precise spatial control of transcription ³ . While architectural proteins such as CTCF, cohesin, and YY1 have been implicated in loop formation and stabilization^4,5, E-P interactions are crowded molecular environments populated by transcription factors, coactivators, and chromatin regulators ⁶ . Yet, despite intense interest, the field has lacked a method capable of directly identifying the protein composition at specific chromatin loop anchors. ChIP-MS and other locus-centric methods cannot tell whether a protein sits at a chromatin loop anchor or merely lies nearby on the same fiber ⁷ . As a result, the protein machinery that assembles at bona fide chromatin loops, the looposome, has remained largely hypothetical.

The study under discussion resolves this long-standing limitation through a conceptual and technical advance: LoopID, the first proteomics approach explicitly built around chromatin loop structure rather than linear genomic position ⁸ .

A Novel Proteomic Method, LoopID

LoopID integrates CRISPR-dCas9 targeting with split-TurboID ⁹ proximity labeling to selectively biotinylate proteins at interacting chromatin loci ⁸ . By tethering complementary halves of TurboID to enhancer and promoter anchors using dCas9 systems, enzymatic activity is reconstituted only when the two loci are brought into spatial proximity by chromatin looping. This design confers three decisive advantages. First, LoopID achieves interaction specificity, labeling proteins only at bona fide E-P contacts. Second, it exhibits a high signal-to-noise ratio. Third, LoopID enables quantitative comparison of looposome composition under different cellular or genetic perturbations.

Applying LoopID to the Nanog and Oct4 E-P interactions in mouse embryonic stem cells (ESCs) ¹⁰ , the authors generate the first comprehensive catalog of looposome proteins. This dataset includes canonical architectural factors, transcriptional regulators, and, strikingly, a large cohort of epigenetic regulators (ERs) whose roles in chromatin structure had not been fully appreciated.

Collectively, LoopID emerges as a foundational technology for interrogating the molecular logic of 3D genome organization.

Epigenetic Modifiers as Structural Regulators

Recent studies have unveiled that certain ERs can extend their influence on transcription beyond their catalytic activities ¹¹ . Among the looposome components identified, the histone demethylase JMJD2 (KDM4) family stood out for its strong enrichment at E-P anchors genome-wide ⁸ . Traditionally, JMJD2 proteins have been studied for their enzymatic activity in removing H3K9 and H3K36 methylation marks ¹² . This study revises that view. Acute depletion of JMJD2 leads to a rapid and global weakening of E-P interactions without immediate changes to histone methylation or transcription, positioning it as a critical player in maintaining 3D genome architecture. This temporal separation reveals that chromatin architecture is disrupted before changes in epigenetic marks or gene expression, placing JMJD2 upstream of transcriptional consequences. Crucially, catalytic mutants of JMJD2 fully rescue E-P interactions and looposome integrity, demonstrating that JMJD2 regulates chromatin looping independently of its demethylase activity ⁸ . This represents direct demonstration that an epigenetic modifier can act as a structural organizer of chromatin rather than solely as a chemical modifier of histones.

This finding introduces a new conceptual framework: ERs can influence transcription by shaping chromatin topology itself, not merely by writing or erasing histone marks or recruiting transcriptional cofactors.

Phase Separation as a Mechanism for Loop Stabilization

Mechanistically, the study shows that JMJD2 contains extensive intrinsically disordered regions (IDRs) that drive biomolecular condensate formation.^4,8,13 JMJD2 forms dynamic, liquid-like nuclear puncta that colocalize with E-P interaction sites, display rapid fluorescence recovery, and dissolve upon treatment with aliphatic alcohols, hallmarks of phase separation ⁴ . In vivo experiments showed JMJD2 forming liquid-like condensates at E-P interaction sites, whereas in vitro experiments demonstrated IDR-driven droplet formation and nucleosome recruitment.

Remarkably, condensate-rescue mutants engineered by fusing heterologous IDRs restore loop integrity and looposome composition. These experiments provide compelling evidence that phase separation is the key physical mechanism by which JMJD2 stabilizes chromatin loops.

Engineering E-P Loops to Reprogram Cell Fate

Perhaps the most provocative aspect of the study is its functional leap from mechanism to manipulation. If E-P interactions can be stabilized by condensates, can they be engineered at will to control cell fate? To address this, the authors develop an inducible dCas9-based system to recruit JMJD2 condensates to selected enhancers and promoters. By assembling condensate-competent JMJD2 at predefined genomic loci, they induce de novo E-P interactions that persist even after withdrawal of the inducing signal. Applying this strategy to two-cell-stage genes such as Zscan4 and Nelfa, the authors successfully activate normally non-active E-P loops in ESCs, triggering transcriptional programs characteristic of two-cell-like cells (2CLCs). Conversely, engineering pluripotency-associated loops promotes reprogramming toward a pluripotent state. Notably, condensate-incapable or solid-like JMJD2 mutants fail to maintain these engineered interactions, underscoring the functional importance of phase separation. These results provide the first demonstration that cell-type-specific E-P loops can be engineered to enable direct control of cell fate transitions, potentially revolutionizing transplantation therapies.

Implications for Epigenetics, 3D Genomics, and Stem Cell Biology

This study delivers three major advances with broad implications (Fig. 1):

1. Technological innovation: LoopID establishes a new class of proteomics, proteomics based on chromatin structure, that enables direct interrogation of molecular assemblies at specific chromatin loops.

2. Mechanistic innovation: The discovery of a catalytic-independent, condensate-driven function of JMJD2 challenges the prevailing enzyme-centric view of ERs. It suggests that many chromatin modifiers may possess underappreciated structural roles in genome organization.

3. Functional innovation: The ability to engineer E-P interactions through phase separation provides a powerful strategy for cellular reprogramming. This approach may prove transformative for regenerative medicine, disease modeling, and synthetic genome engineering.

Figure 1.

Innovations in chromatin loop proteomics, mechanistic understanding, and functional application. (a) Technological innovation: Development of LoopID. Schematic of LoopID, combining CRISPR-Cas9 and split proximity labeling to identify proteins at chromatin loop anchors, revealing key regulators of enhancer–promoter (E-P) loops. (b) Mechanistic innovation: Catalytic-independent function of JMJD2. Discovery that JMJD2 mediates E-P loops via phase separation, independent of its catalytic activity, highlighting a new structural role for chromatin modifiers. (c) Functional innovation: Engineering E-P loops for cell fate manipulation. Illustration of establishing E-P loops through phase separation at specific loci, enabling gene expression control and cell fate reprogramming, with potential for regenerative medicine and synthetic biology.

More broadly, the work underscores a unifying principle: chromatin architecture, epigenetic regulation, and cell identity are inseparably linked through physical mechanisms of molecular self-organization.

Conclusion

By uniting chromatin-interaction-based proteomics, phase separation biology, and stem cell reprogramming, this study reshapes our understanding of how E-P interactions are built, maintained, and manipulated. Together, this work represents a dual breakthrough—technological and conceptual.

Technologically, the development of LoopID, a chromatin structure–based proteomics, provides the first systematic approach capable of identifying protein components specifically localized at looping anchors. Conceptually, the discovery that JMJD2 forms condensates to mediate E-P interactions uncovers the first catalysis-independent role of an epigenetic modifier in organizing chromatin architecture. Strikingly, the demonstration that E-P loops can be engineered to control cell fate heralds a new era in which 3D genome organization becomes a programmable dimension of biology.