The challenge the researchers faced is a classic one in modern biology. Single-cell RNA sequencing can measure how much every gene is expressed in thousands of individual cells at once, revealing which genes are noisy and which are consistently switched to the same level across a population. But a list of variable genes does not by itself reveal the proteins responsible for the variability. Hundreds of chromatin factors, transcription factors and RNA-processing proteins could, in principle, influence how steadily a gene is transcribed. Testing them all one by one would be prohibitively slow, so the team built an integrative computational pipeline designed to narrow the field before any experiment began.

The logic of the pipeline combines two kinds of genome-wide maps. The first comes from single-cell RNA sequencing of mouse embryonic stem cells, which quantifies transcriptional variability for each gene across the population. The second comes from chromatin immunoprecipitation sequencing, or ChIP-seq, datasets that record where candidate regulatory proteins physically bind across the genome. By overlaying these maps, the researchers could ask a targeted question: are the genes that show unusually high or unusually low expression variability preferentially bound by particular chromatin-associated proteins? If a protein’s binding sites are enriched near genes whose expression is especially stable, that protein becomes a candidate stabilizer; if its sites cluster near conspicuously noisy genes, it becomes a candidate destabilizer. This cross-referencing strategy turns scattered public datasets into a ranked shortlist of potential regulators of transcriptional noise.

Among the candidates that emerged from this screen, one stood out: Nucleosome Assembly Protein 1-Like 1, or NAP1L1. NAP1L1 belongs to the NAP1 family of histone chaperones, proteins that shuttle histones around the nucleus and help assemble nucleosomes, the spool-like structures around which DNA is wrapped. Because nucleosome organization directly shapes how accessible DNA is to the transcription machinery, a histone chaperone is a plausible lever on transcriptional consistency. The pipeline flagged NAP1L1 as a potential transcriptional stabilizer, meaning that its presence appeared to be associated with steadier gene expression in the stem cells.

To test this prediction directly, the team turned to a precision-targeting technique built on a deactivated form of the CRISPR-associated protein Cas9. In this approach, the DNA-cutting activity of Cas9 is disabled, but the protein retains its ability to be guided to a specific DNA sequence. The researchers fused NAP1L1 to this dCas9 scaffold and used guide RNAs to deliver the fusion protein to the promoter of Nanog, the gene whose noisy expression has made it a textbook example of transcriptional heterogeneity in stem cells. The result was striking: tethering NAP1L1 to the Nanog promoter stabilized NANOG expression in the cells. In other words, simply parking the histone chaperone at the gene’s control region was enough to reduce the cell-to-cell fluctuations that normally characterize this pluripotency factor, providing direct experimental evidence that NAP1L1 can act as a stabilizer at a specific locus.

A single-gene experiment, however, could not reveal whether NAP1L1’s influence is local or global. To address that question, the researchers generated embryonic stem cells in which the Nap1l1 gene was knocked out and compared them with wild-type cells using genome-wide single-cell transcriptomics. The comparison showed that loss of NAP1L1 increased transcriptional variability across the genome. Genes that had been expressed at consistent levels in the presence of the protein became noisier in its absence, supporting the idea that NAP1L1 plays a broad, global role in keeping transcription steady rather than merely fine-tuning one or two loci. This genome-wide view transforms the finding from an interesting observation about Nanog into evidence for a general mechanism of transcriptional stabilization in pluripotent cells.

With a functional role established, the next question was mechanistic: what does NAP1L1 actually do at the molecular level to smooth out transcription? To find its working partners, the team performed co-immunoprecipitation followed by mass spectrometry, a technique in which NAP1L1 is pulled out of cell extracts along with any proteins bound to it, and the captured molecules are then identified by their mass fingerprints. This experiment identified Developmental Pluripotency Associated 3, or DPPA3, as a key interacting partner of NAP1L1. DPPA3 is a protein associated with pluripotency and the germline, and its physical association with a histone chaperone suggests a plausible mechanistic basis for NAP1L1’s function: the two proteins may cooperate in organizing chromatin at variable genes, thereby influencing how reliably those genes are transcribed from cell to cell.

The significance of the work extends beyond one protein in one cell type. Transcriptional heterogeneity in embryonic stem cells is widely believed to be functionally important, priming individual cells toward different developmental fates and contributing to the plasticity that makes these cells so valuable for research and regenerative medicine. Yet the field has lacked a systematic framework for identifying the molecules that govern this heterogeneity. The integrative pipeline described in the study, which fuses single-cell expression data with chromatin binding maps, offers exactly that: a reusable recipe for discovering stabilizers and destabilizers of gene expression noise, complete with a published list of identified candidate regulators in the supplementary materials. Other laboratories can now apply the same strategy to different cell types, different developmental stages or different single-cell datasets to build a fuller catalogue of the noise-controlling machinery of the genome.

There are also broader implications for how scientists think about chromatin. Histone chaperones are usually discussed in terms of DNA replication and nucleosome assembly, housekeeping roles that keep the genome packaged correctly as cells divide. The new findings suggest that at least one member of this family has a more dynamic regulatory function, actively shaping the consistency of transcription in living stem cells. If NAP1L1 stabilizes expression by maintaining a favorable chromatin configuration at prone-to-fluctuate genes, then the balance between chromatin assembly and disassembly may be one of the fundamental determinants of how noisy a gene is, connecting the physical packaging of DNA to the probabilistic behavior of transcription that single-cell technologies have made visible.

The study, which was supported by the Israel Ministry of Science and by the European Union’s Horizon Europe programme under the EIC Pathfinder-Open grant RT-SuperES, was published open access in Genome Biology, with Tamar Segal, Elad Dvir, Matan Sorek, Amal Gharbi, Xue Sun, Michal Shoshkes-Carmel, Oren Ram and Eran Meshorer as authors. For now, the immediate takeaway is conceptual: the randomness observed in stem cell gene expression is not an uncontrolled byproduct of cellular life but a property with identifiable molecular guardians. NAP1L1, a histone chaperone recruited to the Nanog promoter through a CRISPR-based targeting system, demonstrably quiets transcriptional fluctuations, and its removal loosens the reins on gene expression across the genome. As researchers continue to dissect the NAP1L1-DPPA3 partnership and the other candidates surfaced by the new pipeline, the noisy inner life of the embryonic stem cell is likely to become considerably less mysterious.

Subject of Research: Regulation of transcriptional variability by the histone chaperone NAP1L1 in mouse embryonic stem cells

Article Title: Systematic discovery of potential regulators of transcriptional variability reveals NAP1L1 as a transcriptional stabilizer in mouse ESCs

Article References: Segal, T., Dvir, E., Sorek, M., Gharbi, A., Sun, X., Shoshkes-Carmel, M., Ram, O., & Meshorer, E. (2026). Systematic discovery of potential regulators of transcriptional variability reveals NAP1L1 as a transcriptional stabilizer in mouse ESCs. Genome Biology. https://doi.org/10.1186/s13059-026-04267-9

Image Credits: AI Generated

DOI: 10.1186/s13059-026-04267-9

Keywords: embryonic stem cells, NAP1L1, DPPA3, transcriptional variability, pluripotency, Nanog, histone chaperone, single-cell RNA sequencing, ChIP-seq, dCas9, chromatin, Genome Biology