During cell fate transitions, cells remodel their transcriptome, chromatin, and epigenome; however, it has been difficult to determine the temporal dynamics and cause-effect relationship between these changes at the single-cell level. Here, we employ the heterokaryon-mediated reprogramming system as a single-cell model to dissect key temporal events during early stages of pluripotency conversion using super-resolution imaging. We reveal that, following heterokaryon formation, the somatic nucleus undergoes global chromatin decompaction and removal of repressive histone modifications H3K9me3 and H3K27me3 without acquisition of active modifications H3K4me3 and H3K9ac. The pluripotency gene OCT4 (POU5F1) shows nascent and mature RNA transcription within the first 24 h after cell fusion without requiring an initial open chromatin configuration at its locus. NANOG, conversely, has significant nascent RNA transcription only at 48 h after cell fusion but, strikingly, exhibits genomic reopening early on. These findings suggest that the temporal relationship between chromatin compaction and gene activation during cellular reprogramming is gene context dependent. 

The dataset contains quantifications related to STORM, OligoSTORM, and smRNA-FISH data.

Cell lines and culture conditions

Primary human BJ fibroblasts (ATCC) were routinely plated at a density of 5x10³ cells/cm² and maintained in DMEM 1X (Gibco) containing 10% of Fetal Bovine Serum (Cytiva), Penicillin/Streptomycin 1X (Gibco) and Sodium Pyruvate 1X (Gibco). Culture media was changed every other day. Cells were passaged every three to four days (until reaching 80-90% of confluency) using 0.05% Trypsin-EDTA (Gibco). Tcf3^-/- mouse embryonic stem cells (mESCs)¹ at passages 25-28 were plated at a seeding density of 15-20x10⁴ cells/cm² and cultured in KnockOut^TM DMEM, 1X (Gibco), containing 15% of Fetal Bovine Serum (Cytiva), Penicillin/Streptomycin 1X (Gibco), Sodium Pyruvate 1X (Gibco), GlutaMAX 1X (Gibco), MEM aminoacids 1X (Gibco), 2-Mercaptoethanol 50 µM (Gibco) and 10⁶ units/ml of LIF (Leukemia Inhibitory Factor, EMD Millipore). Media was changed daily and cells were passaged every three days using Accutase (Innovative Cell Technologies, Inc.). Human iPSCs TOMM20-GFP (Allen Cell Collection) were used at passages 37-39 and cultured in mTeSR plus (Stem Cell Technologies) on a hESC-qualified Matrigel-coated surface (Corning) and media was changed daily. hiPSCs were passaged in small clumps using Versene 1X to detach cells and when plated, media was supplemented with 10 µM of Y-27632 ROCK inhibitor (Stem Cell Technologies) during 12h. All cells in this study were cultured in a humidified hypoxic incubator at 37°C containing 5% CO₂ and 5% O₂.

Heterokaryon generation

5 x 10⁶ to 10 x 10⁶ BJ fibroblasts at passage 5-6 and a similar number of mouse ESCs Tcf3^-/- at passage 25-27 were harvested, centrifuged and resuspended in PBS 1X; cells were counted and mixed at 1:1 ratio and centrifugated in a 50 ml tube at 1 x 10³ RMP. Pelleted cells were disrupted by gently tapping the bottom of the tube and incubated 2 min at 37°C. Next, 1-2 ml of warm (37°C) polyethylene glycol (PEG 1500, Roche) was added slowly and cells were incubated during 2 minutes at 37°C under continuous agitation; then, 20 ml of DMEM F12 at 37°C was added slowly and cells were centrifuged and resuspended in PBS 1X. Finally, cells were centrifuged again and resuspended in PBS 1X containing 10% of Fetal Bovine Serum and incubated during 1h with both Phycoerythrin (PE) and Alexa Fluor® 647-conjugated antibodies against the mutually exclusive surface antigens CD90 (BD Pharmingen) for fibroblasts and E-Cadherin (BioLegend) for mESCs as 0.5 µl of antibody per ml per every 10 x 10⁶ cells. Cells were resuspended in PBS 1X supplemented with 5% of Fetal Bovine Serum and 25 mM of HEPES 1X and sorted in an Influx B sorter. Double positive (PE-CD90⁺, Alexa Fluor® 647-E-Cadherin⁺) cells were considered as fused cell and plated into borosilicate glass bottom 8-well chamber slides (Nunc™ Lab-Tek™) coated with Laminin (Sigma-Aldrich) at a density of 3 x 10⁴ cells/cm². Additionally, single positive cells were sorted as controls. All cells were cultured in mESC conditions for up to 48h after fusion. Media was changed at 24h.

Immunostaining

Cells were washed three times with PBS 1X at room temperature (RT) during five minutes each (all subsequent PBS washes were performed similarly) and permeabilized with Triton X-100 (Fisher BioReagents) 0.1% v/v for 10 minutes, RT and washed with PBS 1X. To avoid potential artefacts at chromatin², cells were fixed using an osmotically-balanced paraformaldehyde (PFA) 4% w/v solution (Electron Microscopy Sciences) during 10 minutes at RT and washed with PBS 1X. For STORM imaging, cells were incubated in blocking buffer consisting of bovine serum albumin (BSA), 10% w/v (Fisher) and Triton-X100 0.0025% in PBS 1X, for one hour at RT. Primary antibodies were diluted in blocking buffer and used as follows: H2B (Proteintech) 1:25; H3K9me3 (Thermo Fisher Scientific) 1:50; H3K27me3 (Thermo Fisher Scientific) 1:100; H3K4me3 (Thermo Fisher Scientific) 1:100; and H3K9ac (Thermo Fisher Scientific) 1:50; and incubated 12h-15h at 4°C in a humidified chamber. Cell were washed three times in agitation at RT in washing buffer containing 2% w/v BSA and 0.01% v/v Triton X-100 in PBS 1X. Secondary antibody was added in blocking buffer containing DAPI (Thermo Scientific) at 5 µg/ml during 1 hour at RT and then washed with PBS 1X. For STORM imaging, in-house made secondary antibodies conjugated with Alexa Fluor® 405-647 organic dye pairs were used at the same dilution as primary antibodies. DAPI and Lamin/AC staining patterns were used to distinguish the human and mouse nucleus in heterokaryons during imaging.

TRA-1-60 colony assay

Cells were fixed in PFA 4% for 20 minutes, RT; washed with PBS 1X and permeabilized with Triton X-100 0.3% for 20 minutes at RT. Anti-TRA-1-60-biotin antibody (eBioscience) 1:200 was incubated 12h at 4°C. Cells were then incubated in SAv-HRP (Streptavidin Horseradish Peroxidase, Biolegend) 1:200 for one hour, followed by PBS 1X wash. Finally, DAB staining (Vector Laboratories) was performed accordingly to manufacturer’s instructions.

RNA-FISH staining

Stellaris® reagents and probes (BIOSEARCH technologies) were used for all RNA-FISH experiments and RNA-FISH staining was based on manufacturer’s protocol for adherent cells available at www.biosearchtech.com/stellarisprotocols and were performed in RNase-free conditions. Briefly, cells were fixed in PFA 4% as above described, and washed in PBS 1X; then, cells were incubated in ethanol 70% v/v for 24h at 4°C. Subsequently, cells were incubated in Wash Buffer A for 5 minutes at RT, followed by 12-15h incubation in Hybridization Buffer containing 125 nM of RNA-FISH probe at 37°C covered from light. Next, cells where incubated in Wash Buffer A for 30 minutes at 37°C, then in Wash Buffer A containing DAPI (5 µg/ml) for 30 minutes at 37°C; and finally, incubated in Wash Buffer B for 5 minutes. RNA-FISH probes used contained the CAL Fluor® Red 590 dye or Quasar® 670 dye and were targeted against human exonic POU5F1/POU5F2 (OCT4); human exonic NANOG; human exonic LMNA; intronic OCT4 and intronic NANOG. Intronic RNA-FISH probes were designed using Stellaris Probe Designer v4.2 (www.biosearchtech.com/stellarisdesigner) (Supplementary Table 1).

Oligopaint probe preparation and staining

A 12K oligopool oligopaint library designed by (12) that contains probe sequences targeting the hg19 human genome assembly coordinates chr6:31,105,000-31,163,000 for OCT4-TCF19 (OCT4), chr12:7,931,000-7,970,000 for NANOG and chr12:6,641,500-6,666,000 for GAPDH-IFFO1 (GAPDH) was synthetized (CustomArray - Genscript). Sequences for OCT4, GAPDH and NANOG were quality checked by end-point PCR (Kapa Taq PCR kit, KAPA Biosystems). Next, probes were prepared using a protocol adapted from Neguembor et al.³ and Beliveau et al.⁴ Briefly, target sequences were amplified by end-point PCR; T7 promoter were added by touch up PCR followed by T7 RNA polymerase in vitro transcription (HiScribe T7 High Yield RNA Synthesis Kit, New England Biolabs) and reverse transcription (Maxima H Minus RT, Thermo Fisher Scientific) to generate single-strand DNA sequences containing the Alexa Fluor® 647 fluorophore-labeled primers (IDT). For oligonucleotide sequences, see Supplementary Table 2. Reaction products were mixed with 1 M NaOH and 0.5 M EDTA solution (1:1 v/v) and incubated at 95°C 10 minutes. Single strand DNA probes were purified and concentrated (DNA Clean & Concentrator–100 kit, Zymo Research).

For Oligopaint staining, cells were fixed in PFA 4% during 10 minutes at RT; washed with PBS 1X and incubated in Triton X-100 0.7% v/v for 10 minutes at RT. Then, cells were washed with PBS 1X and incubated successively in 2X SSCT for 5 minutes, RT; 2X SSCT + 50% (v/v) formamide (Ambion) for 5 minutes, RT; 2X SSCT + 50% (v/v) formamide for 7 min at 92°C; and 2X SSCT + 50% (v/v) formamide for 20 min at 60°C. Next, cells were incubated in hybridization mix consisting of 10% (w/v) dextran sulfate 2X SSC, 0.1% (v/v) Tween 20, 50% (v/v) formamide, 2.5 mM dNTPs, 250 ng/µl RNase A and 100 pmol of Alexa Fluor® 647-labelled oligopaint probe for 7 minutes at 92°C; followed by incubation at 37°C in a humidified chamber for 12h-15h covered from light. The next day, cells were incubated 15 minutes in 2X SSCT at 60°C; 10 minutes in 2X SSCT at RT and 10 minutes in 0.2 SSC + DAPI (5 µg/ml) at RT; finally, cells were transferred into 2X SSC.

Imaging acquisition

All image acquisitions in this study were performed in a Nanoimager-S 70 (Oxford Nanoimaging microscope) using NimOS Nanoimager^TM Software (Version: 1.7.2.10222 - 3a2ef15d. Build number: 4. 2016-2019 Oxford Nanoimaging Ltd. All Rights Reserved). Microscope was equipped with four lasers: 405 nm, 488 nm, 561 nm and 640 nm with 498-551 and 576-620 nm band-pass filters for channel 0 and 666-705nm band-pass filter for channel 1. A 100 x 1.4 NA oil immersion objective (Olympus) was used to collect fluorescence signal and imaged onto a Hamamatsu Flash 4 V3 sCMOS camera with a pixel size of 117 nm and a field of view of 50 µm x 80 µm. HILO (Highly Inclined and Laminated Optical sheet illumination) was used for all imaging sessions. For imaging, all cells in this study were randomly selected.

Direct STORM was used to image chromatin (H2B) and histone modifications and was performed using an imaging buffer containing 0.1 M cysteamine (MEA), 5% w/v glucose and 1% GLOX solution. Imaging buffer was replenished every 90 minutes. Images were acquired with a constant high-intensity (80 mW) 647 laser exposure during 4x10⁴ frames at 15 millisecond (ms) exposure time. 405 laser was activated in one-second pulses every 1x10⁴ frames and a constant exposure after frame 3x10⁴. OligoSTORM imaging was performed similarly to direct STORM with a total of 3x10⁴ frames and using an imaging buffer containing β-Mercaptoethanol 1.43 M, 1% GLOX solution, 10% w/v glucose, 10 mM NaCl and 50 mM Tris-HCl, pH 8.0 in PBS 1X. Imaging buffer was replenished every 90 minutes. For RNA-FISH imaging, all images were acquired as a series of optical sections along the z-axis for 10 µm total volume with stack images every 500 nm. Samples containing Quasar® 670 dye were excited with a 640 laser at 1s exposure-time and samples stained with CalFluor® Red 590 dye were exited with a 561 laser at 1s exposure time followed by 405 laser at 100 ms exposure-time for DAPI image. All RNA-FISH imaging was performed in a buffer containing 500 mM NaCl, 5% glucose and 1% GLOX solution in PBS 1X. Buffer was replenished every 90 minutes.

Sequential RNA-FISH/OligoSTORM staining and imaging

Heterokaryons were stained and imaged for RNA-FISH as previously described. However, before image acquisition, an overview tile-scan for DAPI was performed to identify and record the approximate position of individual heterokaryon cells in the chamber-slide related to the microscope stage position; cells were then stored in PBS 1x for up to five days. Next, cells were stained for Oligopaint as previously described except that the steps of fixation and permeabilization were omitted; then, the previously RNA-FISH imaged cells were manually found using their recorded approximate position and imaged for Oligopaint/OligoSTORM as previously described.

Super-resolution image processing and analysis

All STORM images were subject to automatic drift correction and filtered for a localization precision of 30 nm in X and Y coordinates through NimOS software after acquisition. Data was then exported into csv files and converted into bin files. For STORM imaging of chromatin, ROIs (nuclei) were manually cropped using a custom written ImageJ plugin. Before quantification, STORM bin files of nuclei were subject to a pre-processing step to remove over-localization artefacts using a custom made MATLAB® code⁵. Briefly, localizations were Voronoi segmented (see below) and clustered, where cluster area versus the number of localizations was plotted, which is expected to positively correlate for real clusters belonging to biological structures. Clusters deviating (> 3 S.D.) from the positive correlation of cluster density vs cluster area were considered as imaging artefacts and discarded. Voronoi Tessellation analysis of the super-resolution images^6,7 was performed in MATLAB® (R2019b Update 1 (9.7.0.1216025). 64-bit (win64)) similar to^8,9. Briefly, localizations in the x and y coordinates were processed by “delunayTriangulation” function, followed by the “Voronoidiagram” function to generate Voronoi polygons; finally, the Voronoi polygon areas were calculated from the shoelace algorithm. Voronoi density was calculated by taking the inverse of the Voronoi polygon area (1/Voronoi area). To quantify chromatin compaction, Voronoi density values at the 95^th percentile corresponding to H2B localizations were plotted for each individual nucleus. For Voronoi density visualization, 5^th to 95^th percentile of Voronoi polygon areas were rendered and color-coded based on area, where the smallest and biggest Voronoi polygons were set to red and blue, respectively. Normalized localization density was defined as the total number of localizations divided by the nuclear area. For OligoSTORM analysis, STORM bin files were overlapped with the corresponding conventional images (oligopaint foci) to manually crop the target area. As a quality control, ROIs below a threshold of 800 localizations for OCT4 and GAPDH and 500 localizations for NANOG were discarded. Bin files were then used for analysis of the degree of compaction by calculating the radius of gyration, (root-mean square distance of localization positions from their centroid position) similar to^2,3. Finally, obtained values were divided by the corresponding probe genomic length to normalize against different probe sizes.

RNA-FISH analysis

Acquired images were saved as TIF files and maximum intensity projection images were generated by combining all stack images. Subsequent steps for automatic spot quantification were performed using FISH-QUANT¹⁰, a MATLAB-based software with a graphical user interface, accordingly to software’s manual of instructions. Briefly, cytoplasmic and nuclear outlines were first defined using DAPI and RNA-FISH images, followed by image filtering by a two-step convolution with a 3D Laplace of Gaussian function; then, a spot pre-detection was performed using local maxima where localized voxels brighter than the immediate surroundings were defined as spot candidates. Spots were then identified based on a pixel intensity threshold and fitted with their corresponding spot 3D-Gaussian function. Finally, all images were processed in batch mode using the same settings for spot quantification for each type of probe.

Immunofluorescence image analysis

Conventional immunofluorescence quantifications were performed semi automatically, where a custom made ImageJ macro was applied to the average intensity projections of all images to generate a mask covering the fluorescence values at either the region containing the entire nuclear area for the histone modification marks or the LAMIN A/C at the nuclear periphery. Segmented images were manually inspected and any segmentation errors were corrected, and the resulting mean fluorescence intensity values were extracted. Background fluorescence subtraction was performed by extracting the mean fluorescence intensity values of random cytoplasmic regions and subtracted from the values of the segmented images.

Gene expression analysis

Total RNA from heterokaryons at 24h and 48h after fusion, as well as control human fibroblast BJ, mouse ESCs Tcf3^-/- and human iPSCs TOMM20-GFP from three biological replicates was extracted using RNeasy Micro Kit (Qiagen). RNA was quantified using NanoDrop^TM One^C (Thermo Fisher Scientific) and 500 ng of RNA of each condition was converted into cDNA using iScript™ cDNA Synthesis Kit (Bio-Rad) and diluted 1:6. RT-qPCR was performed by triplicate for each biological replicate using KAPA SYBR FAST Universal kit (KAPA Biosystems) on a QuantStudio^TM 7 Flex System (Thermo Fisher Scientific) using human specific primers (see Supplementary table 3) using the following program: 95°C for 3 min, then 40 cycles of 95°C, 3 sec and 60°C, 20 sec; and finally 95°C, 15 sec, 60°C, 1 min and 95°C, 15 sec. Gene expression is reported as 2^–∆Ct relative to the housekeeping gene GAPDH.

QUANTIFICATION AND STATISTICAL ANALYSIS

All statistical analyses were performed using GraphPad Prism 8.2.1. Errors reported in this study correspond to the standard error of the mean (SEM) unless otherwise indicated and n numbers in plots correspond to the number of individual values unless otherwise indicated.
The type of statistical test, n numbers, errors and exact p values are reported in the figures and described in the figure legends. Data with p values ≤ 0.05 where considered statistically significant.

1.          Merrill, B.J., Pasolli, H.A., Polak, L., Rendl, M., García-García, M.J., Anderson, K. V, and Fuchs, E. (2004). Tcf3: a transcriptional regulator of axis induction in the early embryo. Development 131, 263–274. 10.1242/dev.00935.

2.          Boettiger, A.N., Bintu, B., Moffitt, J.R., Wang, S., Beliveau, B.J., Fudenberg, G., Imakaev, M., Mirny, L.A., Wu, C., and Zhuang, X. (2016). Super-resolution imaging reveals distinct chromatin folding for different epigenetic states. Nature 529, 418–422. 10.1038/nature16496.

3.          Neguembor, M.V., Martin, L., Castells-García, Á., Gómez-García, P.A., Vicario, C., Carnevali, D., AlHaj Abed, J., Granados, A., Sebastian-Perez, R., Sottile, F., et al. (2021). Transcription-mediated supercoiling regulates genome folding and loop formation. Mol Cell 81, 3065-3081.e12. 10.1016/j.molcel.2021.06.009.

4.          Beliveau, B.J., Boettiger, A.N., Nir, G., Bintu, B., Yin, P., Zhuang, X., and Wu, C.-T. (2017). In Situ Super-Resolution Imaging of Genomic DNA with OligoSTORM and OligoDNA-PAINT. Methods Mol Biol 1663, 231–252. 10.1007/978-1-4939-7265-4_19.

5.          Gyparaki, M.T., Arab, A., Sorokina, E.M., Santiago-Ruiz, A.N., Bohrer, C.H., Xiao, J., and Lakadamyali, M. (2021). Tau forms oligomeric complexes on microtubules that are distinct from tau aggregates. Proc Natl Acad Sci U S A 118. 10.1073/pnas.2021461118.

6.          Otterstrom, J., Castells-Garcia, A., Vicario, C., Gomez-Garcia, P.A., Cosma, M.P., and Lakadamyali, M. (2019). Super-resolution microscopy reveals how histone tail acetylation affects DNA compaction within nucleosomes in vivo. Nucleic Acids Res 47, 8470–8484. 10.1093/nar/gkz593.

7.          Castells-Garcia, A., Ed-Daoui, I., González-Almela, E., Vicario, C., Ottestrom, J., Lakadamyali, M., Neguembor, M.V., and Cosma, M.P. (2022). Super resolution microscopy reveals how elongating RNA polymerase II and nascent RNA interact with nucleosome clutches. Nucleic Acids Res 50, 175–190. 10.1093/nar/gkab1215.

8.          Andronov, L., Orlov, I., Lutz, Y., Vonesch, J.-L., and Klaholz, B.P. (2016). ClusterViSu, a method for clustering of protein complexes by Voronoi tessellation in super-resolution microscopy. Sci Rep 6, 24084. 10.1038/srep24084.

9.          Levet, F., Hosy, E., Kechkar, A., Butler, C., Beghin, A., Choquet, D., and Sibarita, J.-B. (2015). SR-Tesseler: a method to segment and quantify localization-based super-resolution microscopy data. Nat Methods 12, 1065–1071. 10.1038/nmeth.3579.

10.        Mueller, F., Senecal, A., Tantale, K., Marie-Nelly, H., Ly, N., Collin, O., Basyuk, E., Bertrand, E., Darzacq, X., and Zimmer, C. (2013). FISH-quant: automatic counting of transcripts in 3D FISH images. Nat Methods 10, 277–278. 10.1038/nmeth.2406.