CRISPR-mediated multiplexed live cell imaging of nonrepetitive genomic loci with one guide RNA per locus

Design of chromatin interaction imaging experiments

Fastq files of ARPE-19 RAD21 ChIA-PET experiment (ENCSR110JOO) were downloaded from ENCODE database and processed by ChIA-PET2³³. Interactions were ranked by PET count. Top-ranked interactions with different PET distances were selected. To avoid interference of CTCF binding, design regions were selected by shifting external to PET regions. JACKIE³⁴ was then used to identify unique 1-copy CRISPR sites in the hg38 genome overlapping design regions of selected loops and further filtered for specificity by Cas-OFFinder³⁵. ChIA-PET loops are displayed on WashU EpiGenome Browser ³⁶.

Cloning for Casilio imaging

The lentiviral dCas9 expression plasmid (lenti-dCas9-Blast) was generated by PCR-based mutagenesis of lentiCas9-Blast plasmid (Addgene #52962, gift from Feng Zhang). Clover fused with PUF RNA-binding domain was pAC1447 (Clover_PUFc) (Addgene #73689). PUF9R-tethered iRFP670 and mRuby2 were created by a combination of PCR (from IDT gBlocks) and ligation cloning placing iRFP670³⁷ or mRuby2³⁸, respectively, downstream of PUF9R. PUFc-tethered iRFP670 and PUF₄₈₁₀₇ tethered Clover was created by a combination of PCR (from IDT gBlocks) and ligation cloning placing iRFP670³⁷ upstream of PUFc or Clover upstream of PUF₄₈₁₀₇, respectively. These PUF-fluorescent protein fusions contain nuclear localization signal (NLS) for their localization in the nucleus.

Guide RNA are under control of the human U6 promoter. gRNA spacer sequences were cloned into sgRNA-PBS expression vectors pCR8-sgRNA-15xPBSc or pCR8-sgRNA-15xPBS9R via an oligonucleotide-annealing protocol ³⁰.

All plasmids were subjected to restriction diagnostic tests and sequenced to ensure correct cloning. Plasmids are deposited on Addgene (Supplementary Table 2).

Target sequences for gRNAs

MUC4: GCCGGTGACAGGAAGAGTGC

RAD21 loops in ARPE-19 and HCT116/RAD21-mAID cells

MASP1-BCL6 loop Locus A: GGGTAAGAAGCCACTAGGGT

MASP1-BCL6 loop Locus B: GCATAGCCGCATTTGAAAGC

MASP1-BCL6 loop Locus R: ACGGATCGGACCCACCATGT

IER5L-P: GACTCCCGCGGGTCACTCGG

IER5L-SE: GCCATGCTCCGATAAGGATA

IER5L-M: GTTTACCCCAAGGGTCGCGG

RAD21-independent loop in HCT116/RAD21-mAID cells

Clover locus: GCTTGCCGACGATGGACCCAT

iRFP670 locus: GAGGATCGAAACGCTGATGCG

Competitor gRNA design

cGAL4: GAACGACTAGTTAGGCGTGTA

c(MASP1–BCL6)A (MASP1-BCL6 loop Locus A competitor, dSaCas9): GTGCTCCAGGGTAAGAAGCCAC

cIER5L-P: GCTTGACTCCCGCGGGTCACT

Cell culture

Human osteosarcoma U2OS cells (ATCC^® HTB-96^™) were cultivated in Dulbecco’s Modified Eagle’s Medium (DMEM) (Sigma #D5671) with 10% fetal bovine serum (Gibco, ThermoFisher Scientific #10437-028), 4% Glutamax (Gibco, ThermoFisher Scientific #35050-061), 1% Sodium Pyruvate (Gibco, ThermoFisher Scientific #11360-070) and 1% penicillin-streptomycin (Gibco, ThermoFisher Scientific #15140-163). Human retinal pigment epithelial ARPE-19 cells (ATCC^® CRL-2302^™) were cultivated in DMEM/F12 (Gibco, ThermoFisher Scientific #11330-032) with 10% fetal bovine serum, and 1% penicillin-streptomycin. Human chronic myeloid leukemia HAP1 cells (Horizon^™ C859) were cultivated in Iscove’s Modified Dulbecco’s Medium (IMDM, Gibco, ThermoFisher Scientific #12440-053) with 20% fetal bovine serum and 1% penicillin-streptomycin. The HCT116/RAD21-mAID cell line was a gift from Kanemaki Lab²⁸. The cells were cultured in McCoy’s 5A medium (ATCC #30-2007) with 10% fetal bovine serum, 2 mM L-glutamine (ATCC #30-2214), penicillin-streptomycin. Incubator conditions were humidified 37 °C and 5% CO₂. Cell lines expressing constitutive dCas9 were generated by transducing cells with lentiviruses prepared from a lenti-dCas9-Blast plasmid, followed by Blasticidin selection (Sigma #15205). Cell lines were obtained from trusted sources listed above and were not authenticated in our lab.

Transfection

U2OS/dCas9 cells were seeded at density of 100,000 cells/compartment in a 35 mm 4-compartment CELLview™ cell culture dish (Greiner Bio-one, VWR #89125-442) 24 h before transfection. Cells were transfected with 550 ng of sgRNA plasmid containing 15 Pumilio Binding Sites (15xPBS) and 50 ng of Clover-PUF fusion plasmid, using 1.5 μl Lipofectamine 3000 (Invitrogen, ThermoFisher Scientific #L3000075). Media was changed at 24 h post-transfection.

ARPE-19/dCas9 cells were seeded at density of 80,000 cells/compartment in 35 mm 4-compartment CELLview™ cell culture dish 17 h before transfection. Cells were transfected with 400 ng of each sgRNA-15xPBS plasmid, 40 ng of Clover-PUFc fusion plasmid, and 40 ng of PUF9R-iRFP670 fusion plasmid using 1.5 μl Lipofectamine LTX (Invitrogen, ThermoFisher Scientific #15338-100). For competitor experiments, 400 ng of the competitor plasmid was added. Media was changed at 24 h post-transfection.

HAP1/dCas9 cells were seeded at density of 150,000 cells/compartment in 35 mm 4-compartment CELLview™ cell culture dish 30 h before transfection. Cells were transfected with 400 ng of sgRNA-15xPBS plasmid and 40 ng of Clover-PUFc fusion plasmid DNA using 3.5 μl Turbofectin (OriGene #TF81001). Media was changed at 24 h post-transfection.

HCT116/RAD21-mAID/dCas9 cells were seeded at density of 60,000 cells/compartment in 35 mm 4-compartment CELLview™ cell culture dish the day before transfection. For each well, cells were transfected with 300 ng of each sgRNA plasmid and 40 ng of each fluorescent protein plasmid using 3.5 µL TurboFectin. For competitor experiments, 300 ng of the competitor plasmid was added. For three-color experiments, cells were transfected with 400 ng of each sgRNA plasmid, 40 ng of Clover-PUF₄₈₁₀₇ fusion plasmid, 40 ng of iRFP670-PUFc fusion plasmid, and 400 ng of PUF9R-mRuby2 fusion plasmid using 3.5 µL TurboFectin. Media was changed at 24 h post-transfection. Degradation of the AID-tagged RAD21 was induced by the addition of 500 μM (final concentration) indole-3-acetic acid (IAA/auxin; Sigma Aldrich #I5148) during media change 24 h prior to imaging if applicable.

Live-cell confocal microscopy

Imaging was performed at 41–51 h post-transfection. Prior to imaging, cells were stained with 0.5–1.0 μg/ml Hoechst 33342 prepared in cell culture media for 15–30 min, followed by two media washes. Images were acquired with the Dragonfly High Speed Confocal Platform 505 (Andor) using an iXon EMCCD camera and a Leica HCX PL APO ×40/1.10 W CORR objective or a Leica HC PL APO ×63/1.47NA OIL CORR TIRF objective mounted on a Leica DMi8 inverted microscope equipped with a live-cell environmental chamber (Okolab) at humidified 37 °C and 5% CO₂. Imaging mode was Confocal 25 μm. Hoechst images were acquired with a 200 mW solid state 405 nm laser and 450/50 nm BP emission filter. Clover images were acquired with a 150 mW solid state 488 nm laser and 525/50 nm BP emission filter. mRuby2 images were acquired with a 150 mW solid state 561 nm laser and 620/60 nm BP emission filter. iRFP670 images were acquired with a 140 mW solid state 637 nm laser and 700/75 nm BP emission filter. Z-series covering the full nucleus was acquired at 0.27 μm step size for ×40 objective or 0.13 μm step size for ×63 objective. For time-lapse imaging, the Z-series was acquired at 0.32 μm step size for ×40 objective or 0.16–0.2 μm step size for ×63 objective. Two-color images were acquired every 15 s. Three-color images were acquired every 15 or 24 s. Images are a maximum intensity projection of Z-series. Images were processed in Fusion software using ClearView-GPU deconvolution with the Robust (Iterative) algorithm (pre-sharpening 0, 5 iterations, and de-noising filter size 0.1). Linear adjustments in maximum and minimum levels were applied equally across the entire image to display only nuclear spot signals, and not background nuclear fluorescence. Example images of our image processing pipeline are shown in Supplementary Fig. 2.

Image analysis

To measure two targeted genomic loci distance in time-lapse images, Fiji image analysis software was used³⁹. Z-series acquired at 0.32 μm step size for ×40 objective, and 0.16–0.2 μm step size for ×63 objective was used. If the nucleus drifted over time, the Correct 3D Drift plugin was used⁴⁰. For segmenting and tracking spots, the TrackMate plugin was used⁴¹. Blob diameter was set at 2.0 μm. For each channel, threshold was set to include two spots with the maximum intensity in the 3D volume of the nucleus. Simple LAP Tracker used 2 μm linking max distance, 2 μm gap closing max distance, and 2 gap closing max frame gap. Analysis produced “Spots in track statistics” file which was used to run python script to calculate 3D distances between spots and generate 3D tracks. To determine and correct for potential chromatic aberrations, Invitrogen Tetraspeck 0.1 μm Microspheres (ThermoFisher #T7279) were imaged. On the Dragonfly High Speed Confocal Platform 505 (Andor), Z-series were acquired for both the Leica HCX PL APO ×40/1.10 W CORR objective and the Leica HC PL APO ×63/1.47NA OIL CORR TIRF objective. TrackMate was then used to localize the positions of the microspheres. iRFP670 and mRuby2 chromatic aberrations in each X,Y,Z dimension were calculated relative to Clover, and corrected by offsetting the shifts in each X,Y,Z dimension as follows:

DNA-FISH

Cells were cultured on 12 mm circle coverslip on 24-well plates and transfected with Casilio components. At 48 h, cells were fixed with 4% formaldehyde at room temperature for 10 min, briefly washed once with 1x PBS, permeabilized with 0.5% PBST for 20 min followed by x1 PBS wash, then dehydrated by a series of 75%, 85% and 100% ethanol for 2 min each. Let dry. FISH probe (Cy5-MUC4: /5Cy5/CTTCCTGTCACCGAC, IDT custom DNA oligo) and DNA hybridization buffer were prewarmed and mixed well, added onto slide, covered with cell-attached coverslip, sealed with rubber cement. The sealed cells and probe were then heated on a heat block at 80 °C for 2 min, incubated in a dark humidified chamber at 37 °C overnight. After hybridization, the cell-attached coverslip was unsealed and washed in sequence with ×2 SSC at room temperature for 2 min, 10% formamide/×2 SSC for 5 min at 45 °C, ×2 SSC at room temperature for 5 min, ×0.2 SSC at 45 °C for 10 min twice, then counterstained with DAPI for 5 min and mounted with SlowFade™ Diamond Antifade Mountant (Invitrogen, Thermofisher Scientific #S36967) for imaging.

Genome track and Hi-C maps

Genome tracks were visualized by WashU EpiGenome Browser³⁶ or UCSC Genome Browser⁴². Hi-C maps were generated using FAN-C⁴³.

Statistics and reproducibility

No statistical method was used to predetermine sample size. Sample sizes were chosen as customary in the field and were sufficient to obtain statistical significance. No data were excluded from the analyses. The experiments were not randomized. The Investigators were not blinded to allocation during experiments and outcome assessment.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.