Protocol for measuring endocrine-disruptive effects on transcriptional bursting using single-molecule imaging in human breast cancer cells Authors: Pelin Yaşar, Christopher R Day, Joseph Rodriguez Published: 2026 ABSTRACT SummaryTranscriptional bursts regulate gene expression by altering burst size or burst frequency. Here, we present a protocol to analyze estrogen-responsive transcriptional bursting of the TFF1 gene in human breast cancer cell lines by integrating fixed-cell single-molecule RNA fluorescence in situ hybridization (smFISH) and live-cell single-molecule imaging. We describe cell seeding, ligand dose-response treatment, and smFISH hybridization and imaging to quantify transcription sites. We then detail live-cell imaging of a TFF1-MS2 reporter and integration of both datasets to resolve how endocrine disruptors modulate bursting dynamics.For complete details on the use and execution of this protocol, please refer to Day et al.1Subject areas: Gene Expression, In Situ Hybridization, Single Cell, Single-molecule Assays Before you begin This protocol combines smFISH and live-cell MS2 imaging to analyze transcriptional bursting of the estrogen responsive TFF1 gene in human breast cancer cells (MCF7). To begin characterizing the effects of endocrine disruptor chemicals (EDCs) on transcriptional bursting, we recommend single-molecule RNA fluorescence in situ hybridization (smFISH) due to its speed and cost efficiency. This method uses multiple 18–22 bp probes labeled with fluorescent dyes.2 Hybridization of these probes to target RNAs produces localized fluorescence signals that appear as bright spots, and we quantify their intensity (Figure 1A). Colocalization of exon and intron smFISH signals identifies these bright spots as transcription sites (Figure 1B), with signal intensity serving as a proxy for burst size. Additionally, using smFISH to quantify shifts in the prevalence of active transcription sites across a population can provide meaningful insight into alterations in bursting kinetics. Quantification of transcriptional bursting using smFISH in response to ligand perturbations (A) Schematic of smFISH-based detection of nascent and mature TFF1 transcripts using exon- and intron-targeting probe sets. Transcription site (TS) intensity is used as a proxy of burst size. (B) Schematic of co-transcriptional splicing affecting intron probe detection. Early (5′) introns may be spliced before signal accumulation, leading to loss of intron signal at active transcription sites. (C) Schematic of ligand-dependent effects on transcriptional output. E2 and EDC treatment conditions are indicated. (D) Cell plating and coverslip preparation for smFISH imaging. (E) Representative smFISH image of TFF1 transcripts, shown as individual channels (DAPI, Exon, Intron) and merge. Exon and intron probe signals are pseudocolored cyan and magenta, respectively. Scale bar, 20 μm. (F) Magnified view of (E) highlighting colocalized transcription sites (TS; yellow arrowheads). Scale bar, 10 μm. (G) Illustrative box plot depicting the fraction of cells with transcription sites (% TS) for representative EDC and E2 treatments. This panel is a schematic representation intended to illustrate the type of comparison enabled by this protocol, not a plot of real experimental data. (H) Illustrative box plot depicting RNA abundance at transcription sites (used to estimate burst size) for representative EDC and E2 treatments. This panel is a schematic representation intended to illustrate the type of comparison enabled by this protocol, not a plot of real experimental data. In this workflow, we use smFISH to perform dose-response analysis, quantifying both the percentage of transcribing cells and transcription site intensities (Figure 1C). We plate cells on coverslips and prepare them for imaging as described (Figure 1D). These measurements allow quantification of transcriptional output and the fraction of responding cells across conditions (Figures 1E–1H). Based on these results, we then examine selected conditions using live-cell MS2 imaging to directly measure burst initiation and transcriptional dynamics. Live-cell imaging provides complementary information on burst initiation and burst duration that fixed-cell methods cannot capture3 (Figures 2A and 2B). By combining these approaches, this protocol enables clearer mechanistic interpretation of how different ligands affect transcriptional bursting kinetics. Measurement of transcriptional bursting dynamics using live-cell MS2 imaging (A) Schematic of the TFF1-MS2 reporter system with MS2 stem loops integrated into the endogenous locus, enabling visualization of nascent transcripts via MS2-GFP. (B) Representative live-cell images showing transcription sites (TS) appearing and disappearing over time, consistent with bursting behavior. Scale bar, 5 μm. Innovation Current descriptions of hormone-dependent gene regulation largely rely on bulk-cell measurements that average transcriptional outputs across populations, obscuring transcriptional bursting at individual loci. In contrast, single-cell studies demonstrate that only a subset of cells activates a given gene in response to hormone, revealing heterogeneous, burst-like transcriptional dynamics that bulk-based models fail to capture. Our protocol is innovative because it operationalizes a combined single-molecule RNA FISH (smFISH) and live-cell imaging workflow specifically designed to quantify transcriptional bursting in response to defined hormonal or ligand perturbations. smFISH provides rapid, high-throughput, and absolute measurements of RNA copy number and transcription-site intensity across large cell populations, enabling precise estimation of steady-state and burst-related parameters. In parallel, live-cell assays capture the timing, duration, and frequency of bursts at individual alleles, delivering longitudinal information that endpoint assays alone cannot provide. By integrating these complementary datasets into shared kinetic models, this protocol constrains otherwise underdetermined parameters and links single-cell bursting features to population-level gene activation. Thus, the protocol provides a practical and reproducible framework to move from descriptive imaging to quantitative, mechanistic analysis of hormone-responsive transcriptional regulation. Generate or obtain a TFF1-MS2 reporter cell line Timing: 3–6 months 1.Generate or obtain a TFF1-MS2 knock-in cell line in which 24x MS2 stem loops are inserted into the 3′UTR of the TFF1 gene. Generate or obtain a TFF1-MS2 knock-in cell line in which 24x MS2 stem loops are inserted into the 3′UTR of the TFF1 gene. Note: Live-cell imaging of transcriptional bursting requires a reporter cell line in which MS2 stem loops are integrated into endogenous TFF1 locus. Generation of edited clones typically requires 3–6 months, depending on the cell line and genome editing strategy. Additional details on reporter construction are described in Rodriguez et al.3 2.Express MS2 coat protein fused to GFP (MS2-GFP) in the reporter cell line to enable visualization of nascent transcripts. Express MS2 coat protein fused to GFP (MS2-GFP) in the reporter cell line to enable visualization of nascent transcripts. Prepare hormone-depleted culture conditions Timing: 2 days 3.Culture MCF7 cells in phenol red-free medium supplemented with charcoal-stripped fetal bovine serum. Culture MCF7 cells in phenol red-free medium supplemented with charcoal-stripped fetal bovine serum. Note: Hormone-depleted medium composition and storage are described in the Materials and Equipment section. 4.Perform hormone depletion for 2 days prior to ligand treatment to allow cells to reach steady-state transcriptional activity. Perform hormone depletion for 2 days prior to ligand treatment to allow cells to reach steady-state transcriptional activity. Note: To accurately measure ligand responses, cells should be pre-cultured in hormone-depleted medium before ligand treatment. Order smFISH probe sets Timing: 2 h 5.Order probe sets targeting exon TFF1 RNA labeled with one fluorophore (e.g., Quasar 570). Order probe sets targeting exon TFF1 RNA labeled with one fluorophore (e.g., Quasar 570). 6.Order probe sets targeting intronic TFF1 RNA labeled with a second fluorophore (e.g., Quasar 670) to enable two-color hybridization and transcription site identification. Order probe sets targeting intronic TFF1 RNA labeled with a second fluorophore (e.g., Quasar 670) to enable two-color hybridization and transcription site identification. Note: Detection of transcription sites requires simultaneous hybridization of intronic and exon RNA probes. Intron probe design must account for gene length and position. In long genes, cells may splice 5′ introns before signal accumulates enough to detect, which can preclude colocalization and cause missing transcription sites. Note: Probe set sequences for TFF1 are provided in Table S1. Prepare hybridization and wash buffers Timing: 3 h 7.Prepare the smFISH hybridization and wash buffers described in the materials and equipment section prior to beginning the experiment. Prepare the smFISH hybridization and wash buffers described in the materials and equipment section prior to beginning the experiment. Prepare ligand dilutions Timing: 15 min 8.Prepare 10-fold serial dilutions of ligand in hormone-depleted medium from a 10–3 M E2 stock solution or a 1 M EDC stock solution to obtain 7 working concentrations for E2 (10–13 M to 10–7 M), or 6 working concentrations for EDC of interest(10–9 M to 10–4 M). Prepare each dilution in a volume sufficient to provide 1 mL per well for the required number of replicates (see Step 2 of Cell seeding and ligand dose-response treatment), with a small excess (e.g., and additional 0.5–1 mL) to account for pipetting loss. Prepare 10-fold serial dilutions of ligand in hormone-depleted medium from a 10–3 M E2 stock solution or a 1 M EDC stock solution to obtain 7 working concentrations for E2 (10–13 M to 10–7 M), or 6 working concentrations for EDC of interest(10–9 M to 10–4 M). Prepare each dilution in a volume sufficient to provide 1 mL per well for the required number of replicates (see Step 2 of Cell seeding and ligand dose-response treatment), with a small excess (e.g., and additional 0.5–1 mL) to account for pipetting loss. Note: Concentration ranges are based on dose-response data for TFF1 induction; E2 saturates at approximately 1 nM (EC50∼0.018 nM), while EDCs require higher concentrations (saturating around 10 μM) to achieve a comparable response. CRITICAL: This protocol involves the use of formaldehyde (as a 4% formaldehyde solution) for cell fixation. Formaldehyde is toxic and a suspected carcinogen. Always prepare, handle, and dispose of formaldehyde-containing solutions in a certified chemical fume hood while wearing appropriate personal protective equipment (gloves, eye protection). Collect all formaldehyde waste in a labeled hazardous waste container for institutional disposal; do not pour down the drain. Key resources table REAGENT or RESOURCESOURCEIDENTIFIERChemicals, peptides, and recombinant proteins17-β-estradiol (E2)Sigma-AldrichCat# E2758Formamide (99.5%)ThermoFisher ScientificCat# BP228-10020x SSCQuality BiologicalCat# 351–003–131Dextran sulfate sodium salt from Leuconostoc spp.Sigma-AldrichCat# D8906ProLong Gold Antifade Mountant with DAPIThermoFisher ScientificCat# P3693518 mm #1.5 CoverslipElectron Microscopy SciencesCat# 72222–01MEMThermoFisher ScientificCat# 10370Phenol Red-Free MEMThermoFisher ScientificCat# 51200Penicillin/StreptomycinSigma-AldrichCat# P0781L-Glutamine (200 mM)ThermoFisher ScientificCat# 25030081FBSSigma-AldrichCat# F2442Charcoal/Dextran Treated FBSSigma-AldrichCat# F6765Paraformaldehyde, 32% aqueous solutionElectron Microscopy SciencesCat# 15714-SEthanol, 100%ThermoFisher ScientificCat# BP2818100Triton X-100Sigma-AldrichCat# T8787PBS, RNase-freeThermoFisher ScientificCat# 10010031Experimental models: Cell linesMCF-7ATCCCat# HTB-22; RRID:CVCL_0031MCF-7-TFF1-MS2Rodriguez et al.3N/AOligonucleotidesSingle-molecule RNA FISH probesLGC Biosearch TechnologiesSee Table S1Software and algorithmsMicro-ManagerEdelstein et al.4RRID:SCR_000415CellProfiler 3.1.9Carpenter et al.5RRID:SCR_007358https://cellprofiler.orgFijiSchindelin et al.6RRID:SCR_002285https://imagej.net/software/fiji/downloadsTrackMateTinevez et al.7https://imagej.net/plugins/trackmate/LocalizeLarson et al.8https://github.com/CBIIT/Larson-Lab-CCR-NCI/tree/masterPythonCustom codehttps://github.com/singlecelldynamics/AcuteBPAFusion 2.1.0.88Andor (Oxford Instruments)https://andor.oxinst.com/fusion-acquisition-softwareOtherASI RAMM (Rapid Automated Modular Microscope System) baseASI (Applied Scientific Instrumentation)www.asiimaging.comASI MS-2000 Small XY stageASIhttps://www.asiimaging.com/products/stages/xy-upright-stages/ms-2000-small-xy-stage/ASI High Speed Filter Wheel (FW-1000)ASICat# FW-1000Hamamatsu ORCA-Flash4 V3 CMOS cameraHamamatsuCat# C13440-20CUZeiss C-Apochromat 40x/1.2 NA water-immersion objectiveZeissCat# 421767–9971Lumencor SpectraX LED light sourceLumencorwww.lumencor.comSemrock FF409/493/596-Di02-25×36 dichroic filterSemrockCat# FF409/493/596-Di02-25x36Andor Dragonfly Spinning Disk Confocal/HILO microscopeAndor (Oxford Instruments)https://andor.oxinst.com/products/dragonfly-confocal-microscope-systemNikon CFI Apo TIRF 60x/1.49 NA oil-immersion objectiveNikonCat# MRD01691NUNC Lab-Tek II Chamber Slide, 2-well, No 1.5ThermoFisher ScientificCat# 154526Oko Lab Bold Line Cage IncubatorOko Labhttps://oko-lab.com/live-cell-imaging/microscope-incubators/cage-incubator-bl3Andor iXon Life 888 EMCCD cameraAndor (Oxford Instruments)https://andor.oxinst.com/products/ixon-emccd-camera-series/ixon-life-888 Materials and equipment Complete growth mediumReagentFinal concentrationAmountMEMN/Ato 500 mLFetal Bovine Serum (FBS)10%50 mLL-Glutamine (200 mM stock)2 mM5 mLPenicillin/Streptomycin (100× stock)1×5 mLTotalN/A500 mL Complete growth medium Note: Store at 4°C for up to 6 months Hormone-depleted mediumReagentFinal concentrationAmountPhenol Red-Free MEMN/Ato 500 mLCharcoal/Dextran-Treated FBS10%50 mLL-Glutamine (200 mM stock)2 mM5 mLPenicillin/Streptomycin (100×stock)1×5 mLTotalN/A500 mL Hormone-depleted medium Note: Store at 4°C for up to 6 months Formaldehyde solutionReagentFinal concentrationAmountParaformaldehyde 32% Aqueous Solution4%1.5 mL1X PBS, RNase-freeN/A10.5 mLTotalN/A12 mL Formaldehyde solution Note: Prepare fresh for each experiment CRITICAL: Formaldehyde is toxic and a suspected carcinogen. Wear appropriate personal protective equipment (gloves, eye protection) and prepare and handle this solution inside a chemical fume hood. Permeabilization solutionReagentFinal concentrationAmount100% Ethanol70%8.4 mLNuclease-free waterN/A3.6 mLTotalN/A12 mL Permeabilization solution Note: Prepare fresh for each experiment. CRITICAL: Ethanol is flammable. Prepare and handle away from open flames or heat sources. Alternatives: Permeabilize cells with 0.5% Triton X-100 in 1× PBS for 10 min at 20–25°C instead of the 70% ethanol method described above. smFISH hybridization bufferReagentFinal concentrationAmountDextran sulfate sodium salt (powder)10%5 gDeionized Formamide (99.5%)10%5 mL20× SSC, RNase-free2×5 mLNuclease-free waterN/Ato 50 mLTotalN/A50 mL smFISH hybridization buffer Note: Aliquot 700 μl into Eppendorf tubes (sufficient for one 12-well experiment, including 10% excess to account for pipetting loss) and store at −20°C for up to 1 year. Thaw at 20–25°C when needed. The buffer is stable through 2–3 freeze-thaw cycles. CRITICAL: Formamide is a reproductive toxicant and hazardous by inhalation and skin contact. Wear appropriate personal protective equipment (gloves, eye protection) and prepare and handle this solution inside a chemical fume hood. smFISH wash bufferReagentFinal concentrationAmount20× SSC, RNase-free2×5 mLDeionized Formamide (99.5%)10%5 mLNuclease-free waterN/Ato 50 mLTotalN/A50 mL smFISH wash buffer Note: Prepare fresh or freeze at −20°C and reuse. CRITICAL: Formamide is a reproductive toxicant and hazardous by inhalation and skin contact. Wear appropriate personal protective equipment (gloves, eye protection) and prepare and handle this solution inside a chemical fume hood. 2× SSC rinse solutionReagentFinal concentrationAmount20× SSC, RNase-free2×5 mLNuclease-free waterN/A45 mLTotalN/A50 mL 2× SSC rinse solution Note: Store at 20–25°C. Step-by-step method details Cell seeding and ligand dose-response treatment Timing: 8–9 days To ensure that observed effects are not due to suboptimal ligand concentrations, this step establishes a dose response of TFF1 bursting. We use the dose of maximal activity for comparison against the maximal E2 response. 1.Seed 100,000 MCF7 cells per 18 mm coverslip. This density yields 70–80% confluence on the day of fixation, given the approximately 2-days doubling time of MCF7 cells. Seed 100,000 MCF7 cells per 18 mm coverslip. This density yields 70–80% confluence on the day of fixation, given the approximately 2-days doubling time of MCF7 cells. Note: Plate MCF7 cells at 10–15% confluency as a starting point. MCF7 cells attach poorly to glass coverslips during initial recovery period. 2.Determine the number of samples to plate. For example, for 4 doses, plate three replicates per dose for a total of 12 samples. Determine the number of samples to plate. For example, for 4 doses, plate three replicates per dose for a total of 12 samples. 3.Place one 18 mm No. 1.5 coverslip into each well of a 12-well plate in the cell culture hood (Figure 1D). Place one 18 mm No. 1.5 coverslip into each well of a 12-well plate in the cell culture hood (Figure 1D). Note: Coverslip thickness affects objective correction and image quality; use the specified grade (see Key Resources Table). 4.Add 1 mL of complete growth medium to each well. Add 1 mL of complete growth medium to each well. 5.Gently press down on each coverslip with a sterile pipette tip to reduce cell growth underneath the coverslip. Gently press down on each coverslip with a sterile pipette tip to reduce cell growth underneath the coverslip. 6.Seed predetermined number of cells into each well. Seed predetermined number of cells into each well. 7.Rock the plate gently in up-down and left-right directions to disperse cells. Avoid circular motions, which can cause a ring of cell growth or growth concentrated in the center of the well. Rock the plate gently in up-down and left-right directions to disperse cells. Avoid circular motions, which can cause a ring of cell growth or growth concentrated in the center of the well. 8.Allow cells to recover for 2–3 days in a 37°C cell culture incubator, corresponding to approximately one cell cycle. Allow cells to recover for 2–3 days in a 37°C cell culture incubator, corresponding to approximately one cell cycle. 9.Aspirate the medium and add 1 mL of hormone-depleted medium to each well. Aspirate the medium and add 1 mL of hormone-depleted medium to each well. 10.Return the plate to the 37°C cell culture incubator for 1 h. Return the plate to the 37°C cell culture incubator for 1 h. 11.Repeat Steps 9–10 two more times for a total of three washes. Repeat Steps 9–10 two more times for a total of three washes. 12.After the final wash, return cells to the 37°C cell culture incubator for 3 days to reach steady-state hormone depletion. After the final wash, return cells to the 37°C cell culture incubator for 3 days to reach steady-state hormone depletion. 13.Prepare the ligand dilutions as described in before you begin (Step 8). Prepare the ligand dilutions as described in before you begin (Step 8). 14.Aspirate the hormone-depleted medium and replace with 1 mL of the appropriate dosed medium per well. Aspirate the hormone-depleted medium and replace with 1 mL of the appropriate dosed medium per well. 15.Return the plate to the 37°C cell culture incubator for 3 days to allow cells to reach steady-state transcriptional activity. Return the plate to the 37°C cell culture incubator for 3 days to allow cells to reach steady-state transcriptional activity. Cell fixation and permeabilization Timing: 1 h This step fixes cells with formaldehyde to preserve RNA and cellular structure, then permeabilizes the cell membrane to allow probe access during subsequent hybridization. 16.Wash cells three times with 1 mL PBS per well, swirling the plate by hand at 20–25°C (no shaking apparatus required). Aspirate PBS between washes. Wash cells three times with 1 mL PBS per well, swirling the plate by hand at 20–25°C (no shaking apparatus required). Aspirate PBS between washes. 17.Fix cells with 1 mL formaldehyde solution (4% formaldehyde in PBS; see materials and equipment) for 10 min at 20–25°C in a chemical fume hood. Fix cells with 1 mL formaldehyde solution (4% formaldehyde in PBS; see materials and equipment) for 10 min at 20–25°C in a chemical fume hood. 18.Collect all formaldehyde waste and washes in a labeled hazardous waste container. Do not aspirate directly to the sink. Collect all formaldehyde waste and washes in a labeled hazardous waste container. Do not aspirate directly to the sink. 19.Rinse cells two times for 10 min each with 1 mL PBS, collecting each rinse in the labeled hazardous waste container. Rinse cells two times for 10 min each with 1 mL PBS, collecting each rinse in the labeled hazardous waste container. 20.Permeabilize fixed cells by incubating coverslips with 2 mL of 70% ethanol for 12–16 h at 4°C. Permeabilize fixed cells by incubating coverslips with 2 mL of 70% ethanol for 12–16 h at 4°C. Alternatives: For same-day processing, permeabilize cells with 1 mL of 0.5% Triton X-100 in 1× PBS for 10 min at 20–25°C instead of Step 20. Rinse once for 10 min with 1× PBS, then proceed directly to smFISH hybridization. 21.Seal the 12-well plate with parafilm to reduce evaporation. Seal the 12-well plate with parafilm to reduce evaporation. Pause point: Fixed, permeabilized cells remain stable at 4°C for up to 2 weeks, allowing the protocol to be paused at this step if needed. Probe hybridization for smFISH Timing: 6–7 h This step hybridizes fluorescently labeled exon and intron probes to target RNA, enabling visualization and quantification of transcription sites. 22.Thaw the smFISH hybridization buffer, probe sets, and smFISH wash buffer (see materials and equipment). Thaw the smFISH hybridization buffer, probe sets, and smFISH wash buffer (see materials and equipment). CRITICAL: Protect probe sets from light exposure throughout handling to prevent photobleaching of the fluorescent dyes. Wrap probe tubes in aluminum foil and keep them in a closed drawer during thawing and handling. 23.Prepare a master mix of the probe-hybridization solution scaled to the total number of coverslips being processed. For each coverslip, add 0.4 μl of 12.5 μM exon probe stock and 0.4 μl of 12.5 μM intron probe stock per 49.2 μl of hybridization buffer, for a total of 50 μl per 18 mm coverslip. Prepare a master mix of the probe-hybridization solution scaled to the total number of coverslips being processed. For each coverslip, add 0.4 μl of 12.5 μM exon probe stock and 0.4 μl of 12.5 μM intron probe stock per 49.2 μl of hybridization buffer, for a total of 50 μl per 18 mm coverslip. Note: If the required probe volume from the 12.5 μM stock is less than 1 μl, it may be suitable to dilute the stock 1:10 in TE buffer to 1.25 μM for more accurate pipetting. Prepare 10% excess volume of the master mix to account for pipetting loss due to the viscosity of the hybridization buffer. 24.Pre-hybridization:a.Remove the 12-well plate from 4°C storage.b.Remove the 70% ethanol and collect it in a labeled flammable waste container.CRITICAL: Ethanol is flammable. Handle and dispose of ethanol waste away from open flames or heat sources.c.Add 1 mL of wash buffer per well and incubate for 2–5 min at 20–25°C.Note: Prepare the hybridization chamber (Step 25) while coverslips incubate in wash buffer.Alternatives: If cells were permeabilized with 0.5% Triton X-100 in 1X PBS (see Major Step 2 Alternatives) instead of 12–16 h ethanol, aspirate the Triton-PBS solution, rinse once with 1 mL PBS for 10 min at 20–25°C, then proceed with Step 25. Pre-hybridization:a.Remove the 12-well plate from 4°C storage.b.Remove the 70% ethanol and collect it in a labeled flammable waste container.CRITICAL: Ethanol is flammable. Handle and dispose of ethanol waste away from open flames or heat sources.c.Add 1 mL of wash buffer per well and incubate for 2–5 min at 20–25°C.Note: Prepare the hybridization chamber (Step 25) while coverslips incubate in wash buffer.Alternatives: If cells were permeabilized with 0.5% Triton X-100 in 1X PBS (see Major Step 2 Alternatives) instead of 12–16 h ethanol, aspirate the Triton-PBS solution, rinse once with 1 mL PBS for 10 min at 20–25°C, then proceed with Step 25. a.Remove the 12-well plate from 4°C storage. Remove the 12-well plate from 4°C storage. b.Remove the 70% ethanol and collect it in a labeled flammable waste container.CRITICAL: Ethanol is flammable. Handle and dispose of ethanol waste away from open flames or heat sources. Remove the 70% ethanol and collect it in a labeled flammable waste container.CRITICAL: Ethanol is flammable. Handle and dispose of ethanol waste away from open flames or heat sources. CRITICAL: Ethanol is flammable. Handle and dispose of ethanol waste away from open flames or heat sources. c.Add 1 mL of wash buffer per well and incubate for 2–5 min at 20–25°C.Note: Prepare the hybridization chamber (Step 25) while coverslips incubate in wash buffer.Alternatives: If cells were permeabilized with 0.5% Triton X-100 in 1X PBS (see Major Step 2 Alternatives) instead of 12–16 h ethanol, aspirate the Triton-PBS solution, rinse once with 1 mL PBS for 10 min at 20–25°C, then proceed with Step 25. Add 1 mL of wash buffer per well and incubate for 2–5 min at 20–25°C.Note: Prepare the hybridization chamber (Step 25) while coverslips incubate in wash buffer.Alternatives: If cells were permeabilized with 0.5% Triton X-100 in 1X PBS (see Major Step 2 Alternatives) instead of 12–16 h ethanol, aspirate the Triton-PBS solution, rinse once with 1 mL PBS for 10 min at 20–25°C, then proceed with Step 25. Note: Prepare the hybridization chamber (Step 25) while coverslips incubate in wash buffer. Alternatives: If cells were permeabilized with 0.5% Triton X-100 in 1X PBS (see Major Step 2 Alternatives) instead of 12–16 h ethanol, aspirate the Triton-PBS solution, rinse once with 1 mL PBS for 10 min at 20–25°C, then proceed with Step 25. 25.Prepare the petri dish hybridization chambers.a.Divide the bottom of each 10 cm petri dish into 6 sections by drawing lines from the center outward.b.Number each section 1–6, spacing the sections evenly to correspond to individual coverslip placement.c.Use two petri dishes to accommodate up to 12 coverslips total (sections 1–6 on the first dish, 7–12 on the second). Prepare the petri dish hybridization chambers.a.Divide the bottom of each 10 cm petri dish into 6 sections by drawing lines from the center outward.b.Number each section 1–6, spacing the sections evenly to correspond to individual coverslip placement.c.Use two petri dishes to accommodate up to 12 coverslips total (sections 1–6 on the first dish, 7–12 on the second). a.Divide the bottom of each 10 cm petri dish into 6 sections by drawing lines from the center outward. Divide the bottom of each 10 cm petri dish into 6 sections by drawing lines from the center outward. b.Number each section 1–6, spacing the sections evenly to correspond to individual coverslip placement. Number each section 1–6, spacing the sections evenly to correspond to individual coverslip placement. c.Use two petri dishes to accommodate up to 12 coverslips total (sections 1–6 on the first dish, 7–12 on the second). Use two petri dishes to accommodate up to 12 coverslips total (sections 1–6 on the first dish, 7–12 on the second). 26.Prepare the parafilm liner for each petri dish.a.Cut two sections of parafilm sealing film sized to cover the inside of each petri dish.b.Trim the corners with scissors.c.Remove the backing and press the parafilm flush against the petri dish bottom, new side up. Prepare the parafilm liner for each petri dish.a.Cut two sections of parafilm sealing film sized to cover the inside of each petri dish.b.Trim the corners with scissors.c.Remove the backing and press the parafilm flush against the petri dish bottom, new side up. a.Cut two sections of parafilm sealing film sized to cover the inside of each petri dish. Cut two sections of parafilm sealing film sized to cover the inside of each petri dish. b.Trim the corners with scissors. Trim the corners with scissors. c.Remove the backing and press the parafilm flush against the petri dish bottom, new side up. Remove the backing and press the parafilm flush against the petri dish bottom, new side up. Note: Hybridization on parafilm rather than directly on the petri dish reduces bubble formation in the probe-hybridization mixture during incubation. 27.Place 50 μl of the probe-hybridization working mix on each numbered position. Place 50 μl of the probe-hybridization working mix on each numbered position. 28.Using fine-tipped tweezers, gently lift each coverslip from the wash buffer and grip the edge. Using fine-tipped tweezers, gently lift each coverslip from the wash buffer and grip the edge. Note: Angled tweezers provide better maneuverability. 29.Gently dry the coverslip edge with tissue paper and place the coverslip cell-side down onto the probe-hybridization mixture. Gently dry the coverslip edge with tissue paper and place the coverslip cell-side down onto the probe-hybridization mixture. CRITICAL: Do not remove residual wash buffer from the coverslip before placement. 30.After placing all coverslips, add a folded, moistened piece of tissue paper (soaked with 1 mL nuclease-free water) to the center of each petri dish. Place the lid. After placing all coverslips, add a folded, moistened piece of tissue paper (soaked with 1 mL nuclease-free water) to the center of each petri dish. Place the lid. CRITICAL: The moistened tissue paper keeps the chamber humid. 31.Seal each petri dish with parafilm and place in a light-protected incubator. Incubate for 4 h at 37°C, or 12–16 h at 30°C. Do not shake. Seal each petri dish with parafilm and place in a light-protected incubator. Incubate for 4 h at 37°C, or 12–16 h at 30°C. Do not shake. CRITICAL: Place a small beaker or tray of water in the incubator to maintain humidity. Post-hybridization wash Timing: 1.5 h This step washes away unbound probe following hybridization to reduce background signal. 32.At 3.5 h into the hybridization incubation, place wash buffer into the 37°C incubator to pre-warm for 30 min. At 3.5 h into the hybridization incubation, place wash buffer into the 37°C incubator to pre-warm for 30 min. 33.Using fine-tipped tweezers, transfer each coverslip from the petri dish into a well containing 1 mL of pre-warmed wash buffer, cell-side up. Incubate for 30 min at 37°C. Using fine-tipped tweezers, transfer each coverslip from the petri dish into a well containing 1 mL of pre-warmed wash buffer, cell-side up. Incubate for 30 min at 37°C. 34.Aspirate the wash buffer and add 1 mL of fresh pre-warmed wash buffer per well. Incubate for 30 min at 37°C. Aspirate the wash buffer and add 1 mL of fresh pre-warmed wash buffer per well. Incubate for 30 min at 37°C. Note: Thaw Prolong Gold antifade mountant with DAPI at 20–25°C during this incubation. 35.Aspirate the wash buffer and rinse coverslips with 1 mL of 2× SSC rinse solution (see materials and equipment) swirling the plate by hand. Aspirate the wash buffer and rinse coverslips with 1 mL of 2× SSC rinse solution (see materials and equipment) swirling the plate by hand. 36.Aspirate the 2× SSC rinse solution and add 1 mL of 1× PBS per coverslip. Incubate for 5 min at 20–25°C, protect from light with aluminum foil. Aspirate the 2× SSC rinse solution and add 1 mL of 1× PBS per coverslip. Incubate for 5 min at 20–25°C, protect from light with aluminum foil. Coverslip mounting Timing: 10 min, plus 12–16 h drying This step mounts hybridized, stained coverslips onto glass slides for imaging. 37.Dry coverslips for 20–30 s by touching the edge against a tip box lined with tissue paper to remove excess liquid. Dry coverslips for 20–30 s by touching the edge against a tip box lined with tissue paper to remove excess liquid. 38.Add a small drop of mounting medium to a glass slide. Add a small drop of mounting medium to a glass slide. 39.Gently place the coverslip cell-side down onto the mounting medium. Cover the slides with foil, or store in a closed drawer, and let dry 12–16 h at 20–25°C. Gently place the coverslip cell-side down onto the mounting medium. Cover the slides with foil, or store in a closed drawer, and let dry 12–16 h at 20–25°C. Note: Lower the coverslip slowly onto the mounting medium to avoid trapping air bubbles, which can obscure imaging. smFISH image acquisition Timing: 20 min per coverslip (variable, depending on the number of samples) This step acquires fixed-cell smFISH images for quantification of transcription site number and intensity. Note: We acquired images on a custom widefield microscope built on an ASI Rapid Automated Modular Microscope System (RAMM) base, with a regular nosepiece focus drive (not a piezo stage). We collected signal using a Zeiss C-Apochromat 40×/1.2 NA water-immersion objective. We excited fluorophores using a Lumencor SpectraX LED light source equipped with violet, green, and red LEDs and matched excitation filters (DAPI: 395/25; Quasar 570: 550/15–25; Quasar 670: 640/18). We separated emission using a Semrock FF409/493/596-Di02-25x36 dichroic filter cube, followed by emission filters mounted in an ASI fast filter wheel (a multi-bandpass filter for DAPI and Quasar 670; a 593/40 filter for Quasar 570). We collected images on a Hamamatsu ORCA-Flash4 V3 CMOS camera (2048 × 2048 pixel field of view, 1×1 binning, 6.5 μm chip pixel size, default gain/readout mode, 16-bit depth). We controlled the microscope, LEDs, and acquisition using Micro-Manager4 (MMStudio version 2.0.0-gamma1). 40.Set the acquisition software to capture a 5×5 tile grid of non-overlapping regions per sample, using a 40x objective with MCF7 cells at 70–80% confluency, to image a minimum of 1000–2000 cells for reliable transcription site quantification. Set the acquisition software to capture a 5×5 tile grid of non-overlapping regions per sample, using a 40x objective with MCF7 cells at 70–80% confluency, to image a minimum of 1000–2000 cells for reliable transcription site quantification. 41.Set the Z-stack range to 10 μm with 0.5 μm intervals (21 Z-steps in total) to capture the entire nuclear volume. Set the Z-stack range to 10 μm with 0.5 μm intervals (21 Z-steps in total) to capture the entire nuclear volume. 42.Set the excitation power and exposure time for each channel: 100% LED power with a 500 ms exposure for the smFISH channels (Quasar 570, Quasar 670), and 25% LED power with a 25 ms exposure for DAPI. Set the excitation power and exposure time for each channel: 100% LED power with a 500 ms exposure for the smFISH channels (Quasar 570, Quasar 670), and 25% LED power with a 25 ms exposure for DAPI. 43.Acquire the image set for each sample. Acquire the image set for each sample. Transcription site image analysis Timing: Variable This step processes acquired smFISH images to identify transcription sites, quantify burst size, and compare transcriptional activity across treatment conditions. 44.Install CellProfiler5 (v3.1.9) to segment nuclei and cytoplasm from acquired images. Install CellProfiler5 (v3.1.9) to segment nuclei and cytoplasm from acquired images. 45.Install an integrated development environment (IDE) to run the analysis pipeline. We used Spyder, installed via Anaconda. Install an integrated development environment (IDE) to run the analysis pipeline. We used Spyder, installed via Anaconda. 46.Download the custom Python pipeline described in Day, Yasar et al.,1 available at https://github.com/singlecelldynamics/AcuteBPA, or use an analysis code with equivalent functionality (spot detection, colocalization, and RNA quantification per transcription site). Download the custom Python pipeline described in Day, Yasar et al.,1 available at https://github.com/singlecelldynamics/AcuteBPA, or use an analysis code with equivalent functionality (spot detection, colocalization, and RNA quantification per transcription site). 47.Run the pipeline on the acquired TIFF files. The pipeline generates maximum intensity projections of the input images, then detects smFISH spots by applying a bandpass filter and fitting a 2D Gaussian mask with local background subtraction. Run the pipeline on the acquired TIFF files. The pipeline generates maximum intensity projections of the input images, then detects smFISH spots by applying a bandpass filter and fitting a 2D Gaussian mask with local background subtraction. Note: Adjust the bandpass filter for each channel individually and determine threshold values specific to each channel to optimize spot detection. 48.Locate the pipeline output files for each sample: MaskSpot_ TIFF files, CellCounts and LocColoc CSV files. Locate the pipeline output files for each sample: MaskSpot_ TIFF files, CellCounts and LocColoc CSV files. Note: The MaskSpot_ TIFF files overlay the DAPI signal, smFISH signal, CellProfiler-generated cell and nucleus masks, and called smFISH spots for each probe channel; use these to spot-check samples and adjust CellProfiler segmentation or bandpass filter parameters as needed. The CellCounts CSV file (output separately for each smFISH channel) reports cell ID, total spots per cell, and the number of nuclear and cytoplasmic spots. The LocColoc CSV file lists every called spot, its cellular compartment, whether it colocalizes with a spot in the opposite channel, and a normalized intensity value for colocalizing spots. 49.After identifying transcription sites and quantifying burst size for each dose, compare the percentage of transcribing cells (%TS) and burst size at doses where activity plateaus. After identifying transcription sites and quantifying burst size for each dose, compare the percentage of transcribing cells (%TS) and burst size at doses where activity plateaus. Note: Differences in burst size between ligand- and E2-treated cells indicate different reinitiation rates during a burst. Differences in %TS indicate either a change in the number of active alleles or a change in burst initiation rate. Distinguishing between these two mechanisms requires live-cell imaging (see following major step). Seeding and treatment for live-cell imaging Timing: 8–9 days This step prepares TFF1-MS2 reporter cells for live-cell imaging by seeding, hormone depletion, and ligand treatment at previously determined saturating concentrations (see Transcription site image analysis, above). 50.Seed 100,000 TFF1-MS2 reporter cells onto NUNC Labtek two-chamber No. 1.5 glass dishes in complete growth medium. Seed 100,000 TFF1-MS2 reporter cells onto NUNC Labtek two-chamber No. 1.5 glass dishes in complete growth medium. Note: Confirm that cell confluency reaches 50–70% on the day of imaging. 51.Allow cells to recover for 2–3 days, corresponding to approximately one cell cycle. Allow cells to recover for 2–3 days, corresponding to approximately one cell cycle. 52.Aspirate the medium and add 2 mL of hormone-depleted medium per chamber. Return to the 37°C cell culture incubator for 1 h. Aspirate the medium and add 2 mL of hormone-depleted medium per chamber. Return to the 37°C cell culture incubator for 1 h. 53.Repeat Step 52 two more times for a total of three washes. Repeat Step 52 two more times for a total of three washes. 54.After the final wash, return cells to the 37°C cell culture incubator for 3 days to reach steady-state hormone depletion. After the final wash, return cells to the 37°C cell culture incubator for 3 days to reach steady-state hormone depletion. 55.Prepare the selected ligand at the saturating concentration identified from dose-response analysis (e.g., 1 nM E2 or 10 μM EDC) in hormone-depleted medium. Prepare the selected ligand at the saturating concentration identified from dose-response analysis (e.g., 1 nM E2 or 10 μM EDC) in hormone-depleted medium. 56.Aspirate the hormone-depleted medium and replace with 2 mL of the dosed medium per chamber. Aspirate the hormone-depleted medium and replace with 2 mL of the dosed medium per chamber. 57.Return cells to the incubator for 8 h following ligand treatment. Return cells to the incubator for 8 h following ligand treatment. Note: Transcription factor binding at TFF1 reaches steady state by 8 h post-induction; beginning imaging at this timepoint ensures that steady state, rather than transient, bursting dynamics are captured. 58.Transfer cells to the microscope for live-cell imaging. Transfer cells to the microscope for live-cell imaging. Live-cell MS2 image acquisition Timing: 14.2 h per imaging session This step acquires time-lapse images of TFF1-MS2 transcription sites in living cells to capture burst initiation, duration, and frequency. Note: We acquired images on an Andor Dragonfly Spinning Disk Confocal microscope fitted with an Oko Lab incubation chamber maintaining 37°C with 5% CO2. We collected signal using a Nikon CFI Apo TIRF 60×/1.49 NA oil-immersion objective with a fixed pinhole size of 40 μm. We excited MS2-GFP using 488 nm laser at 6% power, directed to the sample through a quad dichroic (488–13 band), and collected emission through a 521–38 emission filter. We collected images on an Andor iXon Life EMCCD camera (13 μm chip pixel size, no frame averaging, EM gain 200). We controlled the microscope and acquisition using Fusion (v2.1.0.88). We saved images in the microscope’s default acquisition format (.ims for the Dragonfly system). 59.Acquire a short test movie at 6% laser power to check for rapid photobleaching. If bleaching is excessive, reduce the laser power and repeat the test until signal is stable across the imaging session. Acquire a short test movie at 6% laser power to check for rapid photobleaching. If bleaching is excessive, reduce the laser power and repeat the test until signal is stable across the imaging session. Note: Cells expressing lower levels of MS2-GFP coat protein provide better signal-to-noise ratio; cells with the highest overall signal often have excessive diffuse nuclear background that obscures transcription site detection. 60.Enable autofocus to prevent focal drift over the course of long-timescale imaging. Enable autofocus to prevent focal drift over the course of long-timescale imaging. 61.Set the acquisition parameters to capture 7 Z planes with 0.8 μm interval, at 100 s intervals, for 512 frames, for a total imaging duration of approximately 14.2 h. Set the acquisition parameters to capture 7 Z planes with 0.8 μm interval, at 100 s intervals, for 512 frames, for a total imaging duration of approximately 14.2 h. Note: Total imaging time is flexible; capturing changes in inactive period duration is more important than the exact session length. For a gene with an average inactive duration of 1.5 h, 8 h of imaging is sufficient, allowing for two imaging sessions per day if needed. 62.Image 7–10 fields of view to obtain at least 30–50 transcribing cells per condition using a 40 ms exposure time per frame. Image 7–10 fields of view to obtain at least 30–50 transcribing cells per condition using a 40 ms exposure time per frame. Note: Include in the final analysis only cells observed continuously throughout the entire movie. Cell tracking and cropping for MS2 analysis Timing: Variable This step tracks individual cell nuclei across live-cell movies and crops each tracked cell into a separate time-series stack, in preparation for MS2 transcription site segmentation. Note: We used FIJI6 with the TrackMate7 plugin for nuclear tracking, and a custom Python script for cropping. This analysis pipeline requires TIFF-format image files; convert raw acquisition files (see Live-cell MS2 Image acquisition) to TIFF before proceeding, or modify the pipeline’s Python code to accept an alternative file format. 63.Convert the raw image files for each field of view (FOV) to TIFF format, if not already saved in this format. Convert the raw image files for each field of view (FOV) to TIFF format, if not already saved in this format. 64.Calculate and save the maximum intensity projection of each field of view (FOV) using FIJI, producing a time-series TIFF stack for each FOV. Calculate and save the maximum intensity projection of each field of view (FOV) using FIJI, producing a time-series TIFF stack for each FOV. 65.Open a single TIFF stack in FIJI. Open a single TIFF stack in FIJI. 66.Select TrackMate from the menu (Plugins → Tracking → TrackMate). Select TrackMate from the menu (Plugins → Tracking → TrackMate). 67.If not automatically populated from the image metadata, manually enter the calibration settings: pixel size (in microns) and the time interval between frames (100 s; see Live-cell MS2 image acquisition). If not automatically populated from the image metadata, manually enter the calibration settings: pixel size (in microns) and the time interval between frames (100 s; see Live-cell MS2 image acquisition). 68.Select the LoG detector and proceed to the next step. Select the LoG detector and proceed to the next step. 69.Enter the nuclei diameter (in pixels or microns) for the LoG detector. Click Preview to test detection accuracy; the software will draw circles around detected nuclei (Figure 3A). Enter the nuclei diameter (in pixels or microns) for the LoG detector. Click Preview to test detection accuracy; the software will draw circles around detected nuclei (Figure 3A). Note: If the diameter is set too small, the detector will identify too many nuclei; if too large, adjacent nuclei may merge into a single detection. 70.Proceed through the remaining detection steps until the software identifies regions in each time frame. Proceed through the remaining detection steps until the software identifies regions in each time frame. 71.Select the LAP tracker and proceed to the next step. Select the LAP tracker and proceed to the next step. 72.Enter the linking distance and gap-closing distance (e.g., 15 μm for both). Set the gap-closing maximum frame gap to 0. Enter the linking distance and gap-closing distance (e.g., 15 μm for both). Set the gap-closing maximum frame gap to 0. Note: The linking and gap-closing distances approximate one cell diameter, accommodating stage jitter or rapid cell movement between frames. Analyze only full-length movies in which the nucleus was identified in every frame. 73.Proceed to generate tracks. The software will overlay tracked paths on the image. Proceed to generate tracks. The software will overlay tracked paths on the image. 74.Play the movie and visually evaluate tracking accuracy. If tracking is inaccurate, adjust parameters and repeat. Play the movie and visually evaluate tracking accuracy. If tracking is inaccurate, adjust parameters and repeat. Note: Some cells may not be tracked for the entire movie using this method. Exclude these cells from downstream analysis, as partial tracks bias inactive (OFF) period durations toward artificially shorter values. 75.In the Display Options window, click Track, then export the tracked spot data to a CSV file for each FOV. In the Display Options window, click Track, then export the tracked spot data to a CSV file for each FOV. 76.Use the TrackMateTiffCrop.py Python script to crop cells using the exported TrackMate CSV data. Provide the file paths for the maximum intensity projections and corresponding TrackMate CSV files (multiple files can be processed at once). Use the TrackMateTiffCrop.py Python script to crop cells using the exported TrackMate CSV data. Provide the file paths for the maximum intensity projections and corresponding TrackMate CSV files (multiple files can be processed at once). 77.Enter a cropping window size. TrackMateTiffCrop.py uses this value to crop around the X,Y center coordinates of each nucleus (identified by TrackMate and output in the TrackMate CSV file), adding and subtracting the window size from the center coordinates to define the crop region. Enter a cropping window size. TrackMateTiffCrop.py uses this value to crop around the X,Y center coordinates of each nucleus (identified by TrackMate and output in the TrackMate CSV file), adding and subtracting the window size from the center coordinates to define the crop region. Note: For MCF7 cells (approximately 15–20 μm in diameter) imaged on the Dragonfly microscope, a window size of 75 pixels is suitable starting point, producing a final 150×150 pixel crop centered on each nucleus. Increase or decrease this value as needed after checking individual movies to ensure the nucleus is not cut off. The script also saves the first frame of original maximum intensity projection with cell IDs labeled, use this frame to check cropping accuracy. Analysis pipeline for transcriptional bursting from live-cell imaging (A) Detection of nuclei using TrackMate with Laplacian of Gaussian (LoG) segmentation. (B) Configuration of LAP tracking parameters for linking nuclei across time. (C) Identification of transcription sites using Localize for spot detection and intensity extraction. (D) Region-of-Interest (ROI) selection encompassing consecutive bursting events from individual alleles. (E) Hidden Markov Model (HMM) fitting of transcriptional states to extract ON (active) and OFF (inactive) durations. (F) Schematic of cumulative distribution analysis of ON and OFF times to quantify burst duration and initiation frequency. MS2 spot segmentation and tracking Timing: Variable This step segments and tracks individual MS2-labeled transcription sites within cropped, single-cell time-series movies to extract active and inactive burst durations. Note: We used Localize8 for spot calling and segmentation. Note: If a TFF1 allele is not bursting in the first frame of cropped movie, Localize will not capture the first inactive duration. To correct this, prepend the first instance of the transcription site to the beginning of the movie before tracking, then remove this duplicated frame from the data after tracking is complete. Note: Tracking alleles involves two major steps; segmenting transcription spots in each frame, then linking spots across adjacent frames into continuous tracks. 78.Using FIJI (Stacks → Tools → Substack, or the FirstFrameSubstack.ijm macro), add the first frame of the first burst to the beginning of the movie, and save the modified movie. Using FIJI (Stacks → Tools → Substack, or the FirstFrameSubstack.ijm macro), add the first frame of the first burst to the beginning of the movie, and save the modified movie. 79.Open Localize and load the modified TIFF stack. Open Localize and load the modified TIFF stack. 80.Enter the following parameters in Localize.a.Set Auto bpass to ON.b.Set the frame range from 1 to the total number of frames.c.Set FCS mode to ON. Enter the following parameters in Localize.a.Set Auto bpass to ON.b.Set the frame range from 1 to the total number of frames.c.Set FCS mode to ON. a.Set Auto bpass to ON. Set Auto bpass to ON. b.Set the frame range from 1 to the total number of frames. Set the frame range from 1 to the total number of frames. c.Set FCS mode to ON. Set FCS mode to ON. Note: With FCS mode ON, Localize retains the last known position of a transcription site if it cannot detect a site in the current frame, and extracts intensity at that position; this background intensity is required for downstream analysis. 81.Determine the last frame of the next burst and update the frame range entered in Step 80 to reflect this endpoint. Determine the last frame of the next burst and update the frame range entered in Step 80 to reflect this endpoint. 82.Select Localize → Segment with ROI. Draw a region of interest (ROI) encompassing two consecutive bursts from the same allele (Figure 3D). For the first ROI, include the first frame transcription site and the first true burst. Select Localize → Segment with ROI. Draw a region of interest (ROI) encompassing two consecutive bursts from the same allele (Figure 3D). For the first ROI, include the first frame transcription site and the first true burst. 83.Verify segmentation accuracy.a.After segmentation completes, load the newly created.loc file (a text file containing the coordinates and intensities of all identified spots).b.Set the frame to the first frame and confirm that a box appears around the correct spot, and that the software also identified spots during the first burst.c.If the software does not identify the correct spot, repeat segmentation with a larger ROI. Verify segmentation accuracy.a.After segmentation completes, load the newly created.loc file (a text file containing the coordinates and intensities of all identified spots).b.Set the frame to the first frame and confirm that a box appears around the correct spot, and that the software also identified spots during the first burst.c.If the software does not identify the correct spot, repeat segmentation with a larger ROI. a.After segmentation completes, load the newly created.loc file (a text file containing the coordinates and intensities of all identified spots). After segmentation completes, load the newly created.loc file (a text file containing the coordinates and intensities of all identified spots). b.Set the frame to the first frame and confirm that a box appears around the correct spot, and that the software also identified spots during the first burst. Set the frame to the first frame and confirm that a box appears around the correct spot, and that the software also identified spots during the first burst. c.If the software does not identify the correct spot, repeat segmentation with a larger ROI. If the software does not identify the correct spot, repeat segmentation with a larger ROI. 84.Save the.loc file with a name that indicates which frames it covers. Save the.loc file with a name that indicates which frames it covers. 85.Repeat Steps 81–84 for each subsequent pair of bursts in the same allele, adjusting the frame range and ROI each time, until all bursts for that allele are segmented. For the final burst, use an ROI that encompasses only the last burst. Repeat Steps 81–84 for each subsequent pair of bursts in the same allele, adjusting the frame range and ROI each time, until all bursts for that allele are segmented. For the final burst, use an ROI that encompasses only the last burst. 86.Combine all.loc files for the allele and remove any redundant, overlapping frame data. Combine all.loc files for the allele and remove any redundant, overlapping frame data. 87.Load the combined.loc file and select Track → Track. Once tracking completes, Localize generates a.trk file containing the coordinates of the tracked transcription site over time. Load the combined.loc file and select Track → Track. Once tracking completes, Localize generates a.trk file containing the coordinates of the tracked transcription site over time. 88.Verify tracking accuracy using the trkCheck.ijm macro.a.In FIJI, load the trkCheck.ijm macro and enter the file paths for the TIFF stack and the.trk file.b.Run the macro.c.Play the movie or step through frames to visually confirm the tracked transcription site overlays correctly onto the original time-series stack (Figure S1). Verify tracking accuracy using the trkCheck.ijm macro.a.In FIJI, load the trkCheck.ijm macro and enter the file paths for the TIFF stack and the.trk file.b.Run the macro.c.Play the movie or step through frames to visually confirm the tracked transcription site overlays correctly onto the original time-series stack (Figure S1). a.In FIJI, load the trkCheck.ijm macro and enter the file paths for the TIFF stack and the.trk file. In FIJI, load the trkCheck.ijm macro and enter the file paths for the TIFF stack and the.trk file. b.Run the macro. Run the macro. c.Play the movie or step through frames to visually confirm the tracked transcription site overlays correctly onto the original time-series stack (Figure S1). Play the movie or step through frames to visually confirm the tracked transcription site overlays correctly onto the original time-series stack (Figure S1). 89.Shift the data and remove the first frame of the.trk file. Step 78 added this duplicated frame to enable Localize to capture the first inactive duration; remove it after tracking to accurately quantify OFF (inactive) period durations. Shift the data and remove the first frame of the.trk file. Step 78 added this duplicated frame to enable Localize to capture the first inactive duration; remove it after tracking to accurately quantify OFF (inactive) period durations. 90.Load traceHMMFit.py in an IDE (e.g., Spyder) and enter the path to the directory containing all.trk files for a given condition. Load traceHMMFit.py in an IDE (e.g., Spyder) and enter the path to the directory containing all.trk files for a given condition. 91.Run traceHMMFit.py. The script generates intensity trace plots (.trk) and fitted HMM state plots (Figure 3E), and outputs:a.Median and mean ON/OFF (active/inactive) durations, in minutes.b.Image files of the plots (.hmm.png) and HMM state files (.hmm.txt) for plotting in other software.c.A summary file (TrkONOFFSummary.txt) containing the duration of each active and inactive period for all.trk files, with two columns: State and Duration. Run traceHMMFit.py. The script generates intensity trace plots (.trk) and fitted HMM state plots (Figure 3E), and outputs:a.Median and mean ON/OFF (active/inactive) durations, in minutes.b.Image files of the plots (.hmm.png) and HMM state files (.hmm.txt) for plotting in other software.c.A summary file (TrkONOFFSummary.txt) containing the duration of each active and inactive period for all.trk files, with two columns: State and Duration. a.Median and mean ON/OFF (active/inactive) durations, in minutes. Median and mean ON/OFF (active/inactive) durations, in minutes. b.Image files of the plots (.hmm.png) and HMM state files (.hmm.txt) for plotting in other software. Image files of the plots (.hmm.png) and HMM state files (.hmm.txt) for plotting in other software. c.A summary file (TrkONOFFSummary.txt) containing the duration of each active and inactive period for all.trk files, with two columns: State and Duration. A summary file (TrkONOFFSummary.txt) containing the duration of each active and inactive period for all.trk files, with two columns: State and Duration. CRITICAL: Inspect each HMM fit visually before accepting the results. 92.Generate cumulative distribution plots of the active and inactive duration distributions (Figure 3F). Generate cumulative distribution plots of the active and inactive duration distributions (Figure 3F). Integration of smFISH and live-cell MS2 datasets Timing: Variable This step combines smFISH and live-cell MS2 data to distinguish between changes in burst size, active allele frequency, and burst initiation across treatment conditions. 93.Normalize transcription site intensities from the smFISH dataset and compare this measurement across conditions to identify changes in burst size. Normalize transcription site intensities from the smFISH dataset and compare this measurement across conditions to identify changes in burst size. 94.Compare the percentage of transcribing cells (%TS) across conditions from the smFISH dataset to identify potential changes in active allele frequency or burst initiation. Compare the percentage of transcribing cells (%TS) across conditions from the smFISH dataset to identify potential changes in active allele frequency or burst initiation. 95.Compare burst initiation rates directly from the live-cell MS2 dataset to distinguish between these two possible mechanisms. Compare burst initiation rates directly from the live-cell MS2 dataset to distinguish between these two possible mechanisms. Note: If burst initiation rates do not differ between conditions, this indicates that the smFISH-observed change in %TS reflects fewer alleles being in an active state, consistent with previous observations in Day, Yasar et al.1 96.Compare burst duration (MS2 dwell time) between conditions using the live-cell MS2 dataset. Compare burst duration (MS2 dwell time) between conditions using the live-cell MS2 dataset. Note: Burst duration reflects the combined contribution of several processes, including burst size, elongation rate, termination, and transcript retention. Comparing changes in burst duration alongside changes in burst size can help clarify the underlying mechanistic interpretation. Expected outcomes We expect smFISH assays using exon and intron probes targeting the TFF1 to yield bright, punctate RNA spots spanning several pixels, with line-scan intensity profiles showing a single, centered peak above background. Intron-labeled spots should localize predominantly to the nucleus, whereas exon-labeled spots should appear in both the cytoplasm and nucleus. Nuclear colocalization of exon and intron spots will mark active transcription sites (TSs), and individual aneuploid MCF7 cells should exhibit 0–5 TSs under typical induction conditions. From these images, users should obtain heterogeneous RNA per cell distributions that are visually apparent in the raw images and quantifiable across at least ∼1000 cells per condition. Under saturating estradiol stimulation, we expect TFF1 expression to reach on the order of ∼70 RNA molecules per cell (cell line-dependent), and boxplot analysis of RNA at colocalized TSs should yield median burst sizes of approximately 1.5–3 RNA per TS for E2 doses. Comparing RNA per cell, fraction of cells with ≥ 1 TS, and RNA per TS across ligands or perturbations will reveal how total transcriptional output, fraction of responding cells, and burst size change between conditions. We expect live-cell MS2 imaging in the TFF1-MS2 reporter line to produce nuclei with a broad range of MS2-GFP expression levels; 20–30% of cells may lack detectable MS2-GFP signal and will not display visible transcription sites. Cells with excessively high MS2-GFP expression will show elevated nuclear background that precludes clear visualization of TSs. In optimally expressing cells, a minority of nuclei should show 1–3 bright punctate TSs clearly above background. These TSs will appear for multiple consecutive frames, disappear for variable intervals, and then reappear, consistent with transcriptional bursting. Some cells without initial TSs but with adequate MS2-GFP levels may show burst onset at later time points during acquisition. For quantitative analysis, users should obtain ≥ 30 high-quality intensity traces (tracked TSs) per condition, enabling construction of ON- and OFF-time histograms and their corresponding cumulative distributions. Comparing cumulative ON- and OFF-time distributions across ligands or conditions will indicate changes in burst duration (ON time) and burst frequency or initiation (OFF time). Integrating smFISH and MS2 outcomes will allow mechanistic interpretation of how specific ligands or perturbations modulate TFF1 transcriptional bursting. For example, E2 dose response may primarily alter burst frequency, increasing both the fraction of cells with TSs and the number of bursts per allele, whereas TRIM24 inhibition may reduce burst size with minimal effects on the fraction of active cells.3,9 Changes in ON-time distributions without changes in burst size would be consistent with altered burst duration (e.g., following enhancer perturbation), whereas changes in OFF-time distributions would support modulation of burst initiation frequency. Together, these combined readouts enable users to distinguish changes in burst frequency, burst size, burst duration, and number of active alleles, rather than attributing all effects to overall RNA output. Limitations The present protocol quantifies transcriptional bursting at a single endogenous locus, TFF1, and evaluates how ligands or perturbations modulate its bursting dynamics. Extending this combined smFISH and live-cell MS2 framework to additional genes would require generating new cell lines with MS2 stem loops integrated at each target locus, which is technically demanding and time consuming, and thus may limit throughput and generalizability across loci. Measurements of burst duration from MS2 imaging also represent the total dwell time of the nascent, MS2-labeled transcript on chromatin and therefore convolve polymerase elongation through the 3′ MS2 cassette, polyadenylation, and termination. As a result, without additional, locus-specific controls or complementary assays, we cannot unambiguously attribute changes in the observed ON time to a specific kinetic step. Troubleshooting Problem 1 Transcription sites are not visible or nuclear background is too high in MS2 imaging (see Live-cell MS2 image acquisition, Step 59). Potential solution MS2 coat protein (MCP-FP) expression is likely too high, causing diffuse nuclear fluorescence that obscures puncta. Reduce MCP-FP expression and use flow cytometry to select a subpopulation with intermediate nuclear signal, excluding very bright and very dim cells, so that nuclei show clear labeling but 1–3 bright transcription sites remain distinguishable. Problem 2 No transcription sites are visible and nuclei are poorly labeled in MS2 imaging (see Live-cell MS2 image acquisition, Step 59). Potential solution MS2 coat protein expression is likely too low to label MS2-tagged RNAs and nuclei efficiently. Adjust the expression cassette (for example, promoter strength or selection strategy) and repeat flow-based enrichment to obtain cells with moderate nuclear fluorescence, verifying suitable signal with short test movies before long time-lapse acquisition. Problem 3 Weak or no smFISH signal (see Probe hybridization for smFISH, Step 30). Potential solution Hybridization conditions are likely suboptimal. Optimize hybridization temperature in a narrow range around the recommended value and adjust formamide concentration to maximize punctate spot intensity while minimizing background, testing a small matrix of conditions for each new probe set. Problem 4 Poor probe penetration or inconsistent smFISH signal across samples (see Cell fixation and permeabilization, Step 19). Potential solution Fixation/permeabilization may not be optimal for the sample type. If formaldehyde plus ethanol permeabilization yields weak or heterogeneous signal, test formaldehyde fixation followed by brief permeabilization with Triton X-100 (see Alternatives, Step 20), titrating detergent concentration and incubation time to improve probe entry while preserving morphology. Problem 5 No signal from far-red smFISH probes (see smFISH image acquisition, Step 42). Potential solution The imaging system may not be properly configured for the far-red dye. Confirm that the microscope has an appropriate excitation source and filter set for your far-red fluorophore (for example, a Cy5-or Alexa Fluor 647-compatible filter set) and validate with a positive-control sample or fluorescent beads in the same spectral range. Resource availability Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Joseph Rodriguez (joseph.rodriguez@nih.gov). Technical contact Technical questions on executing this protocol should be directed to and will be answered by the technical contacts, Pelin Yasar (pelin.yasar@nih.gov) and Christopher R. Day (christopher.day@nih.gov). Materials availability This study did not generate new unique reagents. The TFF1-MS2 reporter cell line used in this protocol was previously generated and described in Rodriguez et al.3 Data and code availability The custom Python analysis pipeline used in this protocol is available at https://github.com/singlecelldynamics/AcuteBPA and has been archived at Zenodo (https://doi.org/10.5281/zenodo.21362043). Acknowledgments We thank the members of the Rodriguez Laboratory and the NIEHS Epigenetics and RNA Biology Laboratory for ongoing support, insightful discussions, and constructive feedback. We thank Jeff Tucker and Erica Scappini of the NIEHS Fluorescence Microscopy and Imaging Center for technical assistance with imaging. This research was supported by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (ES103331 to J.R.). The contributions of the NIH authors are considered Works of the United States Government. The findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services. Author contributions P.Y., C.R.D., and J.R. conceived the project. P.Y., C.R.D., and J.R. wrote the original draft and reviewed and edited the manuscript. Declaration of interests The authors declare no competing interests. Footnotes Supplemental information can be found online at https://doi.org/10.1016/j.xpro.2026.104821. Contributor Information Pelin Yaşar, Email: pelin.yasar@nih.gov. Christopher R. Day, Email: christopher.day@nih.gov. Joseph Rodriguez, Email: joseph.rodriguez@nih.gov. Supplemental information Document S1. Figure S1 and Table S1 References Associated Data Supplementary Materials Document S1. Figure S1 and Table S1 Data Availability Statement The custom Python analysis pipeline used in this protocol is available at https://github.com/singlecelldynamics/AcuteBPA and has been archived at Zenodo (https://doi.org/10.5281/zenodo.21362043).