Promoter-proximal pausing of RNA polymerase II is a key process regulating gene expression. In latent HIV-1 cells, it prevents viral transcription and is essential for latency maintenance, while in acutely infected cells the viral factor Tat releases paused polymerase to induce viral expression. Pausing is fundamental for HIV-1, but how it contributes to bursting and stochastic viral reactivation is unclear. Here, we performed single molecule imaging of HIV-1 transcription. We developed a quantitative analysis method that manages multiple time scales from seconds to days and that rapidly fits many models of promoter dynamics. We found that RNA polymerases enter a long-lived pause at latent HIV-1 promoters (>20 minutes), thereby effectively limiting viral transcription. Surprisingly and in contrast to current models, pausing appears stochastic and not obligatory, with only a small fraction of the polymerases undergoing long-lived pausing in absence of Tat. One consequence of stochastic pausing is that HIV-1 transcription occurs in bursts in latent cells, thereby facilitating latency exit and providing a rationale for the stochasticity of viral rebounds.

This review summarizes current advances in the role of transcriptional stochasticity in HIV-1 latency, which were possible in a large part due to the development of single-cell approaches. HIV-1 transcription proceeds in bursts of RNA production, which stem from the stochastic switching of the viral promoter between ON and OFF states. This switching is caused by random binding dynamics of transcription factors and nucleosomes to the viral promoter and occurs at several time scales from minutes to hours. Transcriptional bursts are mainly controlled by the core transcription factors TBP, SP1 and NF-κb, the chromatin status of the viral promoter and RNA polymerase II pausing. In particular, spontaneous variability in the promoter chromatin creates heterogeneity in the response to activators such as TNF-α, which is then amplified by the Tat feedback loop to generate high and low viral transcriptional states. This phenomenon is likely at the basis of the partial and stochastic response of latent T cells from HIV-1 patients to latency-reversing agents, which is a barrier for the development of shock-and-kill strategies of viral eradication. A detailed understanding of the transcriptional stochasticity of HIV-1 and the possibility to precisely model this phenomenon will be important assets to develop more effective therapeutic strategies.

Gene regulation in eukaryotes emerges from a dynamic interplay between transcription factors (TFs), nucleosomes, and RNA Polymerase II (Pol II), whose competitive and cooperative binding shapes DNA accessibility and transcriptional output. Single-molecule footprinting (SMF) and long-read chromatin accessibility assays such as Fiber-seq now capture these interactions at nucleotide resolution on individual DNA molecules. However, existing computational tools remain insufficient to decode complex binding events from sparse methylation data. Here, we introduce HiddenFoot, a probabilistic modeling framework based on statistical mechanics that quantitatively infers TF, nucleosome, and Pol II occupancy profiles on single DNA molecules by systematically evaluating all thermodynamically plausible binding configurations. Applying HiddenFoot to SMF and Fiber-seq data from mouse, Drosophila , and human cells, we recovered known TF footprints, precisely resolved Pol II pausing, and identified extensive heterogeneity in nucleosome positioning driven by TF binding. HiddenFoot further distinguishes direct TF–TF cooperativity from nucleosome-mediated co-dependency by estimating pairwise interaction energies and comparing to null models under equilibrium. By integrating biophysical modeling with high-resolution single-molecule data, HiddenFoot offers a general, interpretable framework for dissecting regulatory logic in native chromatin with base-pair precision. Abstract Figure:

Pancreatic islet cells differentiate from a common progenitor pool through tightly regulated transcriptional and epigenetic programs. ISL1, a LIM homeodomain transcription factor, is essential for islet development, but its molecular functions remain poorly defined. Here, we demonstrate that ISL1 is critical for maintaining endocrine cell identity and enabling terminal differentiation, particularly of α- and β-cells. Using conditional Isl1 deletion in endocrine precursors, combined with single-cell RNA sequencing and chromatin profiling (H3K27ac and H3K27me3), we reveal disruption of the transcriptional and epigenetic landscape in Isl1-deficient islets. Loss of Isl1 results in the failure to establish α-cell identity, loss of δ- and γ-cell lineages, and the persistence of immature β-cells with impaired functional profiles in Isl1CKO mice. Longitudinal single-cell analysis shows that Isl1CKO endocrine cells exhibit sustained progenitor-like states and defective β-cell maturation. These defects are accompanied by activation of stress and diabetes-associated transcriptional programs, along with sex-specific responses that may influence disease onset and progression. Mechanistically, ISL1 represses intermediate progenitor programs and facilitates chromatin remodeling necessary for endocrine lineage commitment and terminal maturation. Our findings highlight a previously underappreciated role for ISL1 in preserving endocrine cell fate and function and offer insight into how its dysregulation may contribute to diabetes. Article Highlights: ISL1 is a known maturity-onset diabetes of the young candidate and type 2 diabetes susceptibility gene, yet its molecular role in pancreatic endocrine maturation has remained unresolved. Deletion of Isl1 in endocrine progenitors results in islets composed of dysfunctional α-cells lacking glucagon production and immature β-cells with impaired basal insulin secretion, ultimately accelerating diabetes progression. ISL1 functions as a transcriptional repressor guiding chromatin remodeling and transcriptional transitions toward hormone-producing endocrine cells. The metabolic phenotype resulting from Isl1 deletion is associated with sustained progenitor-like states and activation of diabetes- and stress-associated pathways, with distinct sex-specific responses observed between male and female mice.

Gene regulation in eukaryotes emerges from a dynamic interplay between transcription factors (TFs), nucleosomes, and RNA Polymerase II (Pol II), whose competitive and cooperative binding shapes DNA accessibility and transcriptional output. Single-molecule footprinting (SMF) and long-read chromatin accessibility assays such as Fiber-seq now capture these interactions at nucleotide resolution on individual DNA molecules. However, existing computational tools remain insufficient to decode complex binding events from sparse methylation data. Here, we introduce HiddenFoot, a probabilistic modeling framework based on statistical mechanics that quantitatively infers TF, nucleosome, and Pol II occupancy profiles on single DNA molecules by systematically evaluating all thermodynamically plausible binding configurations. Applying HiddenFoot to SMF and Fiber-seq data from mouse, Drosophila, and human cells, we recovered known TF footprints, precisely resolved Pol II pausing, and identified extensive heterogeneity in nucleosome positioning driven by TF binding. HiddenFoot further distinguishes direct TF--TF cooperativity from nucleosome-mediated co-dependency by estimating pairwise interaction energies and comparing to null models under equilibrium. By integrating biophysical modeling with high-resolution single-molecule data, HiddenFoot offers a general, interpretable framework for dissecting regulatory logic in native chromatin with base-pair precision.