The authors argue that the core of viral infection lies in the intricate interaction network between viral proteins and host proteins. As obligate intracellular parasites, plant viruses carry compact genomes encoding only a handful to a few dozen proteins, and they depend extensively on host factors to accomplish entry, replication, assembly, movement and release of progeny virions. Crucially, these interactions also serve as molecular interfaces linking infection to plant defense. Many viral proteins physically interact with key components of the salicylic acid, jasmonic acid and abscisic acid signaling pathways, often suppressing host immunity to promote infection, while plants can exploit specific interactions as molecular cues to activate immune signaling and restrict viral spread. This bidirectional character makes protein interactions simultaneously virulence factors and defense regulators.

A substantial portion of the review is devoted to the technical arsenal available for detecting these interactions. Biophysical techniques such as microscale thermophoresis, isothermal titration calorimetry and biolayer interferometry quantify binding affinities with high sensitivity, minimal sample volumes and low false-positive rates. Microscale thermophoresis monitors the thermophoretic movement of fluorescently labeled proteins across a temperature gradient, and is particularly useful for proteins that are difficult to express or unstable. Isothermal titration calorimetry directly yields full thermodynamic profiles, including enthalpy, entropy and Gibbs free energy, without the need for labeling. Biolayer interferometry measures binding-induced shifts in the interference pattern of light reflected from fiber-optic biosensors, enabling real-time observation of association and dissociation kinetics, and has been applied to quantify binding between viral effectors and host receptors.

Biochemical and cell-based methods complement these quantitative approaches. Co-immunoprecipitation examines interactions in their natural state inside cells, whereas GST pull-down assays test direct bait-prey interactions in vitro, albeit with the risk of tag-induced nonspecific binding. Bimolecular fluorescence complementation reconstitutes a fluorescent protein from two nonfluorescent fragments fused to the proteins of interest, allowing direct visualization of interactions in living cells, though the technique is notoriously temperature sensitive and can fail at higher physiological temperatures. On the computational side, databases such as STRING and Pfam integrate experimental evidence, domain annotations and literature data to predict and visualize interaction networks, providing structural and functional context for experimentally detected pairs.

The review highlights the growing dominance of mass spectrometry-based proteomics in interaction discovery. By the end of 2023, the BioGRID database had compiled more than 2.63 million protein interaction entries, with mass spectrometry techniques such as affinity capture-MS, proximity labeling-MS and co-fractionation collectively contributing over 45 percent of the data. Affinity purification coupled with mass spectrometry remains the predominant high-throughput method, but it struggles with membrane proteins and transient or weak interactions. Proximity labeling overcomes these limits by biotinylating proteins near a target of interest in living cells; the engineered ligases TurboID and miniTurbo, developed through directed evolution, achieve substantial biotinylation within ten minutes, dramatically shortening experimental timelines. Cofractionation mass spectrometry infers interactions from co-elution during chromatographic separation, while chemical cross-linking mass spectrometry captures direct physical contacts by covalently linking proximal amino acid residues, providing distance constraints for structural modeling. The lysis-free Virotrap method, which packages bait proteins into virus-like particles to capture associated host factors under near-native conditions, further reduces purification artifacts.

Despite these advances, the authors caution that no single method suffices. Mass spectrometry generates vast datasets that demand sophisticated processing, and variability in protein abundance complicates detection of low-abundance interactors. Protein interactions are inherently transient, low-affinity and dynamic, varying across tissues, developmental stages and infection time points. Artificial intelligence tools such as AlphaFold and RoseTTAFold have accelerated structure-based prediction of interactions, but their accuracy depends heavily on the quality of multiple sequence alignments and they perform poorly on orphan proteins and complexes with limited homologous sequences. The review therefore advocates an integrative, dynamic and network-centric strategy combining in vitro, in vivo and computational approaches.

The mechanistic core of the review documents how viruses manipulate host proteins at nearly every regulatory layer. RNA silencing is a central antiviral defense in plants, and viruses counter it with dedicated suppressors. The 38 kDa coat protein of turnip crinkle virus interacts with SGS3 to enhance its suppressor activity; the p19 protein of tomato bushy stunt virus sequesters small interfering RNAs and prevents their incorporation into the RNA-induced silencing complex; and the F-box protein P0 of potato leafroll virus promotes the autophagic degradation of ARGONAUTE1. Beyond post-transcriptional silencing, the P3 protein of rice grassy stunt virus triggers P3IP1- and SERK4-dependent degradation of NRPD1a, the largest subunit of RNA polymerase IV, thereby suppressing RNA-directed DNA methylation. A plant bunyaviral protein has also been shown to disrupt the phase separation of SERRATE, dismantling dicing bodies and reducing microRNA biogenesis.

Viruses likewise target signaling cascades and cellular quality-control systems. The coat protein of beet black scorch virus suppresses MAPKKKα-mediated innate immunity through interaction with 14-3-3a, while the βC1 protein of cotton leaf curl Multan virus inhibits MKK2 and MPK4. The NIb protein of turnip mosaic virus blocks SUMOylation-dependent phosphorylation of the salicylic acid receptor NPR1, and the P2 capsid protein of rice dwarf virus competes with the auxin receptor OsTIR1 for binding to OsIAA10, rewiring auxin signaling to promote symptoms. Tobacco mosaic virus coat protein suppresses the calmodulin-like protein NbCML30, and the RNA-dependent RNA polymerases of cucumber mosaic virus, potato virus Y and tobacco mosaic virus disrupt the phyA-FHY1 interaction to impair far-red light-mediated immunity. Autophagy is another contested hub: the γb protein of barley stripe mosaic virus interferes with the ATG7-ATG8 interaction, while the host receptor NBR1 mediates autophagic degradation of turnip mosaic virus HCpro, which the virus counteracts through VPg and 6K2.

Viruses also co-opt host machinery for replication and movement. Potyviruses use their genome-linked VPg protein to mimic a cap structure and recruit translation initiation factors of the eIF4E family, with different potyviruses requiring different eIF4E isoforms in different hosts. The conserved chaperone HSP70 supports the assembly of tomato bushy stunt virus replication complexes on peroxisomal membranes and the cell-to-cell movement of abutilon mosaic virus, and silencing HSP70 attenuates symptoms. Translationally controlled tumor protein supports potyvirus multiplication, and its silencing in tobacco nearly abolishes potato virus Y symptoms. Conversely, plants deploy sensors and resistance proteins: the E3 ubiquitin ligase RBRL recognizes the coat proteins of rice stripe virus and rice dwarf virus to activate jasmonate-mediated immunity; the tomato Tm-22 and Sw-5b immune receptors recognize viral movement proteins; the RTM genes restrict phloem-based systemic movement of potyviruses; and host kinases such as PKA and CK2 phosphorylate viral proteins to impair their function.

These insights are now feeding directly into crop protection. The review outlines how interaction maps distinguish susceptibility genes, which encode host factors hijacked by viruses, from resistance genes, which mediate immune recognition, enabling targeted genome editing of the former and overexpression of the latter. CRISPR/Cas systems have been used both to cleave viral genomes, as demonstrated for geminiviruses and, using FnCas9 and Cas13a, for RNA viruses such as cucumber mosaic virus and tobacco mosaic virus, and to edit host factors such as the cassava eIF4E-family proteins nCBP-1 and nCBP-2 that interact with cassava brown streak virus VPg. In a striking example of precision editing, base editing of a single residue in the rice strigolactone receptor D14 abolished binding by the rice grassy stunt virus P3 protein while preserving hormone perception, conferring robust resistance in both japonica and indica rice. The authors conclude that integrating interactome mapping, AI-assisted structural prediction and CRISPR-based functional validation offers a rational path toward broad-spectrum, durable virus resistance, while warning that single-point edits can be bypassed by viral mutation and that plant virus interaction databases remain far less mature than their animal-virus counterparts, underscoring the need for sustained, systematic data generation.

Subject of Research: Virus-host protein-protein interactions and their role in plant antiviral defense and crop resistance breeding

Article Title: The multidimensional significance of virus-host protein interactions and their implications in the antiviral defense of plants

Article References: Zhang, J., Qi, C., Wu, Z., Zhang, Y., Shi, J., Li, Z., Guo, J., Zhou, X., Tang, S., Yang, J., & Liu, P. (2026). The multidimensional significance of virus-host protein interactions and their implications in the antiviral defense of plants. Stress Biology, 6(1), Article 39. https://doi.org/10.1007/s44154-026-00309-1

Image Credits: AI Generated

DOI: 10.1007/s44154-026-00309-1

Keywords: plant viruses, protein-protein interactions, antiviral defense, RNA silencing, mass spectrometry, proximity labeling, AlphaFold, CRISPR, susceptibility genes, resistance genes, crop breeding, phytohormone signaling

The core challenge that PRADA addresses is a long-standing bottleneck in proximity labeling. The most widely used tools, such as APEX2 and TurboID, genetically fuse an engineered enzyme to a protein of interest. The enzyme then converts a supplied small molecule into a reactive species that tags nearby biomolecules, which can later be fished out and identified by mass spectrometry or sequencing. The catch is the trigger. APEX2 requires a bolus of exogenous hydrogen peroxide, a reactive oxidant that damages cells and diffuses poorly through intact tissues. TurboID requires biotin, which works in vivo but generates its own background labeling. Neither approach has proven ideal for the delicate, three-dimensional environment of a living organism.

The Shanghai- and Singapore-based team, led by Shuo Han of the Chinese Academy of Sciences together with Yue Wan of the Genome Institute of Singapore, took a different route: instead of supplying hydrogen peroxide directly, they engineered the system to make it on the spot. PRADA pairs a genetically fused peroxidase with an engineered D-amino acid oxidase, or DAAO, derived from the yeast Rhodotorula gracilis. When researchers deliver a nonproteinogenic D-amino acid such as D-phenylalanine or D-alanine — molecules that animal cells largely ignore — the oxidase converts them into hydrogen peroxide right next to the peroxidase. That locally generated oxidant then activates the peroxidase, which converts phenol probes such as biotin-phenol into short-lived radicals that covalently tag whatever proteins or nucleic acids sit within roughly a few tens of nanometers.

The engineering itself was nontrivial. The team used structural modeling and rational mutagenesis to delete a flexible C-terminal loop of the R. gracilis oxidase, producing a monomeric variant that fuses cleanly with APEX2 without disrupting either enzyme’s activity. Extended data show the researchers systematically tested fusion orientations, split-enzyme designs and D-amino acid concentrations, using fluorescence assays to confirm that the cascade produces hydrogen peroxide efficiently and that the labeling signal colocalizes precisely with the tagged protein’s known subcellular address. When the two enzymes were expressed separately rather than fused, labeling dropped sharply, confirming that the reaction is genuinely confined to the immediate vicinity of the fusion protein.

That spatial confinement is what makes the method safe enough for living animals. Because the hydrogen peroxide is generated in situ at nanometer scales and consumed almost immediately by the peroxidase, it never accumulates to toxic levels. The researchers measured malondialdehyde levels, a marker of lipid peroxidation, and protein carbonylation after PRADA labeling and found both below detection limits or unchanged compared with controls. Mitochondrial superoxide, a sensitive indicator of oxidative stress, was actually lower in PRADA-labeled cells than in cells labeled conventionally with APEX2 and exogenous peroxide. Cell viability assays showed no measurable harm from the D-amino acid treatment itself. In effect, PRADA converts the most dangerous step of proximity labeling into a self-limiting, locally contained reaction.

The versatility of the platform is striking. In cultured cells, the team targeted PRADA to the nucleus, nucleolus, endoplasmic reticulum, plasma membrane and mitochondria, and in each case recovered proteomic profiles consistent with the known composition of those compartments. They also showed that PRADA can drive functional polymer assembly — the genetically targeted chemical assembly of conducting or insulating materials inside living cells — extending the technique beyond mapping into materials synthesis. And, crucially, they demonstrated that the peroxidase step can label RNA as well as protein, opening a door that most proximity-labeling systems leave closed.

That RNA reactivity became the foundation of the study’s most inventive application. When peroxidase radicals oxidize RNA, they leave behind chemical lesions that cause reverse transcriptase to misread bases during cDNA synthesis. The team realized these misincorporations could be read as a molecular fingerprint of RNA structure: nucleotides that are chemically accessible — that is, single-stranded and exposed — accumulate more mutations than nucleotides buried inside base-paired helices. Building on the logic of mutational profiling sequencing, they developed PRADA-MaPseq, a strategy that converts peroxidase labeling directly into a spatiotemporally resolved map of RNA secondary structure inside living cells.

They benchmarked PRADA-MaPseq against the well-established mitochondrial DMS-MaPseq datasets, showing that uracil and guanine reactivity scores accurately reproduce known structures of mitochondrial messenger RNAs, with area-under-the-curve values reaching 0.60 to 0.82 across transcripts and inter-experimental correlation coefficients as high as 0.96 for individual transcripts. The approach worked with both 4-thiouridine and 6-selenoguanosine metabolic labeling, and the team optimized reverse transcriptase choice to maximize mutational signal. In short, they turned a proximity-labeling enzyme into a structure-probing reagent that reports on RNA folding in a defined cellular compartment at a defined moment in time.

The payoff came in a mouse xenograft model. The researchers implanted HEK293T cells expressing mitochondria-targeted PRADA into immunodeficient mice, then administered D-amino acids and phenol probes systemically. Labeling proceeded efficiently inside the tumors in living animals. Proteomic analysis recovered the expected mitochondrial proteome, transcriptomic enrichment captured mitochondrial RNAs with high specificity, and PRADA-MaPseq yielded the first in vivo structurome of mitochondrial RNA within a living mammalian tumor. Comparing these in vivo structure maps with in vitro measurements of the same transcripts revealed systematic differences — evidence that RNA folding inside a living organism is actively shaped by its environment rather than being a fixed property of the sequence.

Those differences carried biological meaning. The team focused on LRPPRC, a mitochondrial RNA-binding protein known to organize the folding of the mitochondrial transcriptome. When they knocked down LRPPRC in the xenografts, changes in RNA structure propagated to changes in mitochondrial gene expression, demonstrating that RNA architecture functions as a regulatory layer controlling how mitochondrial genes are read. This finding elevates the structurome from a descriptive atlas to a mechanistic insight: the three-dimensional shapes of RNAs inside mitochondria help determine how much protein the organelle makes, and those shapes are tunable in vivo.

Beyond mitochondria, the breadth of validated applications suggests PRADA could become a workhorse across biology. The team demonstrated labeling in the cytoplasm and nuclei of body wall muscle cells in Caenorhabditis elegans, recovering nuclear-enriched transcripts and detecting retained introns as expected; in rat neurons and Drosophila; and in zebrafish. Because the trigger molecule — a simple D-amino acid — is cell-permeable, inexpensive and largely inert, the method should scale to tissues and organisms where peroxide bolus delivery is impractical, such as dense neural tissue, developing embryos or intact tumors. The authors have filed patent applications, and the analysis code for RNA structuromics is freely available on GitHub under an MIT license, with sequencing and proteomic data deposited in public repositories.

Limitations remain, as with any new technology. The labeling window is measured in tens of minutes rather than seconds, so fast molecular events may still escape capture. The peroxidase chemistry, while gentler than exogenous peroxide, still relies on radical intermediates whose diffusion radius sets the spatial resolution. And the D-amino acid oxidase must be carefully engineered for each fusion context to avoid perturbing the protein under study. Nevertheless, PRADA represents a genuine conceptual advance: it decouples proximity labeling from externally supplied oxidants, unifies protein, RNA and polymer chemistries under one enzymatic umbrella, and — for the first time — brings RNA structure mapping into living animals. If the platform generalizes as its developers hope, the spatial organization of biomolecules in health and disease may soon be readable not just in culture dishes, but in the bodies of living organisms, one D-amino acid at a time.

Subject of Research: An engineered D-amino-acid-activated peroxidase cascade for in vivo multiomic proximity labeling of proteins and RNA structures

Article Title: Multiomic proximity labeling in vivo by D-amino-acid-activated peroxidase reaction

Article References: Liu, J., Han, J., Wang, Y., Song, M., Zhu, J., Liang, Y., Chen, F., Liu, Z., Yan, X., Wang, Z., Peng, W., Tu, R., Zhang, Z., Zhong, B., Sun, H., Yin, J., Chang, J., Long, Y., Men, Y., … Han, S. (2026). Multiomic proximity labeling in vivo by D-amino-acid-activated peroxidase reaction. Nature Chemical Biology. https://doi.org/10.1038/s41589-026-02336-5

Image Credits: AI Generated

DOI: 10.1038/s41589-026-02336-5

Keywords: proximity labeling, PRADA, D-amino acid oxidase, APEX2, RNA structure, structurome, mitochondria, spatial proteomics, chemical biology, in vivo imaging, xenograft, LRPPRC

Proximity labeling rests on a deceptively simple idea. Instead of trying to capture fragile or transient interactions directly, researchers fuse an engineered enzyme to a molecule of interest and let that enzyme chemically tag everything nearby. The tagged neighbors can then be purified and identified by mass spectrometry or sequencing, producing a snapshot of the local molecular environment. Enzymes such as APEX2, which uses hydrogen peroxide to drive radical generation, and TurboID, an evolved derivative of biotin ligase, have transformed the study of protein complexes and organelle proteomes. Applying the same logic to RNA, however, has proven far more difficult, because RNA molecules are chemically distinct, often abundant, and embedded in ribonucleoprotein assemblies that are easily disrupted by harsh labeling conditions.

Lantern was developed to address precisely this gap. The enzyme is designed to label molecules in the immediate vicinity of a chosen RNA, generating a record of the proteins and other RNAs that associate with it in living cells. Early versions of the tool, like many first-generation proximity labeling systems, faced limitations in catalytic rate, background activity, and the conditions required to drive the labeling reaction. Slow enzymes require long labeling periods, during which the cell continues to change, blurring the temporal resolution of the resulting map. High background activity, meanwhile, can swamp genuine neighbors in a haze of nonspecific tags, obscuring the very interactions researchers hope to detect.

Directed evolution offers a systematic way out of this impasse. The approach mimics natural selection in the laboratory: researchers generate large libraries of enzyme variants carrying random mutations, screen or select the variants that perform best on a defined task, and then iterate the process, accumulating beneficial mutations over successive rounds. Applied to Lantern, this strategy allowed the team to interrogate enormous sequence space and identify combinations of mutations that jointly improved catalytic turnover, reduced background, and preserved the enzyme’s ability to function inside the complex chemical environment of a mammalian cell. The result is an evolved Lantern variant that labels proximal RNA and protein molecules far more rapidly than its predecessors.

The significance of speed in proximity labeling is difficult to overstate. Cellular states are not static; signaling events, stress responses, and cell-cycle transitions can remodel the interactome of an RNA within minutes. A labeling reaction that requires hours effectively averages over all of these changes, producing a composite picture that may not correspond to any real biological moment. A fast enzyme, by contrast, can capture a molecular neighborhood on a timescale that approaches the dynamics of the underlying biology. This temporal precision matters enormously for studying processes such as RNA granule assembly, stress granule formation, and the rapid redistribution of RNAs during cellular responses to external stimuli.

Dual labeling of both RNA and protein by the same enzyme is another defining feature of the evolved Lantern system. Most existing tools are specialized: some tag proteins efficiently but leave RNA untouched, while RNA-targeting approaches often rely on separate chemistries that are difficult to reconcile in a single experiment. A unified catalyst that marks both classes of molecules in the vicinity of a target simplifies experimental design and enables genuinely integrated maps of ribonucleoprotein architecture. Because RNA-binding proteins and their RNA partners form tightly interwoven networks, the ability to profile both sides of the interface from a single labeling event provides a more complete and internally consistent picture than combining results from separate, independently optimized systems.

The technical challenges that directed evolution had to overcome are worth appreciating in detail. An ideal proximity labeling enzyme must remain inactive until deliberately deployed, tolerate fusion to diverse RNA-targeting modules such as Cas proteins or RNA-binding domains, operate at physiological temperature and pH, and generate reactive intermediates that diffuse only over a short range before reacting with nearby molecules. Balancing these competing demands is not intuitive; mutations that boost catalytic activity often increase background or alter substrate specificity in undesirable ways. Screening strategies that evaluate variants directly in cellular contexts, rather than in simplified biochemical assays, are therefore essential for identifying enzymes that perform well where it matters, inside living cells rather than in a test tube.

Beyond its immediate technical achievements, the evolved Lantern system points toward broader applications across biology and medicine. Mapping the protein companions of disease-associated noncoding RNAs could reveal how long noncoding RNAs execute their regulatory functions and how mutations disrupt these interactions in conditions ranging from cancer to neurodegeneration. Viral RNAs, which recruit host factors into specialized replication and packaging complexes, could be profiled with unprecedented temporal resolution, illuminating points of vulnerability for antiviral therapeutics. In developmental biology, tracking the changing molecular neighborhoods of specific transcripts as cells differentiate could clarify how post-transcriptional regulation shapes cell fate decisions. The combination of speed, dual specificity, and genetic encodability makes the tool adaptable to virtually any RNA that can be targeted with a suitable binding module.

The study also contributes to a growing appreciation of directed evolution as an engine of innovation in chemical biology. Time and again, natural enzymes have proven to be starting points rather than finished solutions, and laboratory evolution has repeatedly delivered variants with properties that no rational design effort could have predicted. The Lantern work exemplifies this pattern: by letting mutation and selection explore sequence space under experimentally defined pressures, researchers obtained a catalyst whose performance characteristics reflect the specific demands of proximity labeling in living cells. As screening technologies improve and libraries grow larger and more diverse, the pace at which such optimized tools emerge is likely to accelerate, equipping the community with an ever-richer toolkit for interrogating molecular proximity.

For the field of RNA biology in particular, the arrival of a rapid, dual-function proximity labeling enzyme marks a meaningful step forward. The interactomes of RNAs have long been studied through laborious biochemical purification methods that require large quantities of material and inevitably perturb the very assemblies under investigation. A genetically encodable, fast-acting labeling system brings the study of RNA neighborhoods into the same experimental regime that has already revolutionized protein interaction mapping, with all the advantages of sensitivity, scalability, and compatibility with living systems. As researchers begin to apply evolved Lantern to their own questions, the coming years are likely to see a substantial expansion in our understanding of the molecular ecosystems that surround RNA, and of the roles those ecosystems play in health and disease.

Subject of Research: Directed evolution of the Lantern enzyme for rapid proximity labeling of RNA and proteins in living cells

Article Title: Directed evolution of Lantern enables rapid RNA and protein proximity labeling

Article References: Fang, Y., Ren, Z., Zheng, F., Wang, R., Zhao, S., Zhang, Y., Wang, W., Li, C., Liu-Yang, L., Lin, C., Liu, J., & Zou, P. (2026). Directed evolution of Lantern enables rapid RNA and protein proximity labeling. Nature Chemical Biology. https://doi.org/10.1038/s41589-026-02313-y

Image Credits: AI Generated

DOI: 10.1038/s41589-026-02313-y

Keywords: directed evolution, Lantern, proximity labeling, RNA biology, protein labeling, RNA-protein interactions, chemical biology, ribonucleoprotein complexes, enzyme engineering, molecular interactome, living cells, Nature Chemical Biology

The technique, developed by Iwona Nowak, Ahsan H. Polash, Hang T. Huynh, Mahekdeep Kaur and colleagues in the laboratories of Aishe A. Sarshad, Markus Hafner, and Daniel Benhalevy, fuses two powerful approaches that had never been married at this scale. The first is APEX2-based proximity labeling, an enzymatic tagging system that marks proteins and nucleic acids within a narrow radius of a genetically targeted protein. The second is 4-thiouridine-enhanced ultraviolet crosslinking, which covalently locks RNA-binding proteins onto their RNA targets and allows precise identification of crosslinked nucleotides. By combining these with sequential affinity purifications, first for the RNA-binding protein of interest and then for the biotin tag, the researchers built a pipeline that isolates only those protein-RNA contacts that occurred in the marked compartment.

The logic of the method is elegant in its layering. A protein of interest, such as the RNA-binding factor being studied, is expressed as a fusion with APEX2, a peroxidase that in the presence of biotin-phenol and hydrogen peroxide converts nearby molecules into biotin-tagged derivatives. Cells are first treated with 4-thiouridine, which is incorporated into newly synthesized RNA and dramatically strengthens ultraviolet-induced crosslinking between RNA and any protein touching it, a strategy borrowed from fPAR-CLIP protocols. When ultraviolet light fires, RNA-binding proteins are frozen onto their targets with nucleotide precision. APEX2 then labels everything in its immediate vicinity with biotin, creating a spatial signature.

What follows is a double selection. The researchers purify the RNA-binding protein and its crosslinked RNA fragments through an initial affinity step, recovering crosslinked RNA with mutations that reveal the exact binding sites, the hallmark of enhanced crosslinking-and-immunoprecipitation approaches. They then recover only those complexes that also carry the APEX2-deposited biotin tag in a second purification. A transcript must pass both filters to be counted: it must have been physically touched by the protein and must have been within the proximity-labeling radius, effectively bracketing the interaction in space and time. The result is a map of protein-RNA contacts that is simultaneously compartment-specific and nucleotide-resolved.

Control experiments were central to validating the approach. The team confirmed that APEX2 fusion proteins localized correctly to their intended compartments and that proximity-dependent biotinylation was restricted to the expected neighborhoods, presenting this evidence across multiple supplementary figures alongside imaging and immunoblot analyses. Without accurate spatial confinement, any apparent compartment specificity could simply reflect leakage or overexpression artifacts, so establishing the fidelity of the labeling step was essential before biological conclusions could be drawn.

With the method proven, the researchers turned it loose on three of the most intensively studied RNA-binding proteins: AGO2, the central effector of the microRNA silencing machinery; YBX1, a multifunctional nucleic acid-binding protein involved in mRNA stability and translation; and ELAVL1, better known in some circles as HuR, a regulator of mRNA stability and stress responses. All three proteins are known to operate in multiple cellular compartments, making them ideal test cases for a technology designed to disentangle location-dependent behavior.

The findings were striking. Each of the three proteins displayed distinct, compartment-specific RNA-binding patterns, targeting different sets of transcripts and even different sequence contexts depending on where in the cell it was acting. Importantly, the researchers found that these spatial differences arose through canonical motif recognition: the same proteins continued to recognize their characteristic sequence motifs in each compartment, but the repertoire of available targets, and thus the functional consequences of binding, shifted dramatically with location. In other words, the proteins did not change their binding preferences so much as their access to substrates changed, partitioning the accessible RNA targets and reshaping each protein’s regulome across the cell.

This partitioning concept has broad implications. For AGO2, compartment-specific mapping opens a window onto how microRNA targeting might differ between the cytoplasm, where canonical silencing occurs, and other locales where AGO2 has been implicated in less conventional roles. For YBX1 and ELAVL1, both implicated in cancer biology and stress physiology, knowing which transcripts are bound in which compartments could clarify how these proteins execute distinct functions from seemingly identical biochemical activities. The method essentially converts a single flat binding profile into a set of spatially annotated maps, revealing that what looked like one interactome is actually several overlapping ones.

Technically, RBProximity-CLIP also demonstrates the value of the 4-thiouridine enhancement inherited from fPAR-CLIP. Enhanced crosslinking increases crosslinking efficiency and yields the characteristic signature of crosslink-induced mutations, allowing binding sites to be pinpointed to individual nucleotides. Combined with proximity tagging, this delivers a rare combination of spatial and sequence resolution in a single experiment, something neither classical CLIP nor proximity labeling alone can provide. The authors provide detailed reagent lists, including antibodies, detection reagents for imaging and immunoblotting, and oligonucleotide sequences for library preparation, alongside full-length uncropped gel and blot images, underscoring the reproducibility infrastructure that accompanies methodological advances of this kind.

The work was a genuinely international effort, jointly led from the University of Gothenburg’s Wallenberg Centre for Molecular and Translational Medicine, the RNA Molecular Biology Laboratory at the National Institute of Arthritis and Musculoskeletal and Skin Diseases in Bethesda, and Tel Aviv University, with support from the Swedish Research Council, the Knut and Alice Wallenberg Foundation, the Swedish Society for Medical Research, the National Institutes of Health Intramural Research Program, the Israel Science Foundation, and the Hellenic Foundation for Research and Innovation. Published as an open-access article in Genome Biology, the study arrives at a moment when the field is increasingly aware that spatial context is not a luxury but a fundamental dimension of gene regulation. As RBProximity-CLIP is adopted and extended, it promises to turn the cell from a featureless bag of molecules into a mapped territory, where every protein-RNA encounter has an address, and where the grammar of gene expression can finally be read with both sequence and location in view.

Subject of Research: A subcellular-resolution RNA-binding protein mapping method combining proximity labeling and enhanced crosslinking

Article Title: RBProximity-CLIP enables subcellular mapping of RNA-binding protein interactions at nucleotide resolution

Article References: Nowak, I., Polash, A. H., Huynh, H. T., Kaur, M., Lobo, V., Scutenaire, J., Fong, M., Alluhaibi, G., Anastasakis, D. G., Hafner, M., Benhalevy, D., & Sarshad, A. A. (2026). RBProximity-CLIP enables subcellular mapping of RNA-binding protein interactions at nucleotide resolution. Genome Biology. https://doi.org/10.1186/s13059-026-04278-6

Image Credits: AI Generated

DOI: 10.1186/s13059-026-04278-6

Keywords: RNA-binding proteins, RBProximity-CLIP, proximity labeling, APEX2, CLIP-seq, AGO2, YBX1, ELAVL1, 4-thiouridine, subcellular resolution, protein-RNA interactions, Genome Biology