Cellular macromolecules-tethered DNA walking indexing to explore nanoenvironments of chromatin modifications

Exploring spatial organization and relationship of diverse biomolecules within cellular nanoenvironments is important to elucidate the fundamental processes of life. However, it remains methodologically challenging. Herein, we report a molecular recognition mechanism cellular macromolecules-tethered DNA walking indexing (Cell-TALKING) to probe the nanoenvironments containing diverse chromatin modifications. As an example, we characterize the nanoenvironments of three DNA modifications around one histone posttranslational modification (PTM). These DNA modifications in fixed cells are labeled with respective DNA barcoding probes, and then the PTM site is tethered with a DNA walking probe. Cell-TALKING can continuously produce cleavage records of any barcoding probes nearby the walking probe. New 3’-OH ends are generated on the cleaved barcoding probes to induce DNA amplification for downstream detections. Combining fluorescence imaging, we identify various combinatorial chromatin modifications and investigate their dynamic changes during cell cycles. We also explore the nanoenvironments in different cancer cell lines and clinical specimens. In principle, using high-throughput sequencing instead of fluorescence imaging may allow the detection of complex cellular nanoenvironments containing tens of biomolecules such as transcription factors.

Supplementary Table 1.Sequence information for oligonucleotides used in Cell-TALKING.

Oligonucleotides for lableing chromatin modifications in cells
Walking Five repeated experiments were performed for statistical analysis of spot count of randomly selected images.The "spot overlap ratio" means the proportion of the spot count of one kind (fluorescence overlaps or not) in total number of all kinds.The letters B, G, and R indicate spots of blue, green, and red, respectively.There are some spots that fail to overlap two or three fluorescence signals, indicating the false negative response of our method on this DNA substrate on glass.It is disccussed in the main text.The data of bar charts are presented as mean values +/-SD.

Supplementary Figure 2. Performance of in vitro
Cell-TALKING on DNA origami substrates.(a) Upper, the actual walking mechanism.Two copies of three barcoding probes (BPs) surround the blocked walking probe (WP) with about 7 nm radius distance on one origami molecule.During the nicking/walking process, the nicking enzyme firstly cuts the blocker sequence in the blocked WP duplex.Then the walking probe molecule in this duplex can be isothermally released to hybridize one BP molecule, forming a DNA duplex as the substrate for nicking enzyme.After cuting this BP molecule in the duplex, the walking probe molecule will walk to and hybridize another BP molecule.In this way, all these BP molecules can be cleaved and the newly generated 3'-OH ends can induce RCA reaction .Middle, statistical analysis of false negative response of our method.The images are corresponding to Figure 2c.Bottom, investigating the false positive response by several negative controls.There are some spots that fail to overlap two or three fluorescence signals, indicating the false negative response of our method on this DNA origami substrate.It is disccussed in the main text.(b) Investigating the probe lengthdetermined recognition range of our method using different walking probes and DNA origami substrates.Five repeated experiments were performed for statistical analysis of spot count of randomly selected images.The "normalized spot count" means the spot count of one sample relative to that of the given one (e.g., the positive control (IV) in Supplementary Figure 2a or the sample using walking probe containing 0 nt spacer in Supplementary Figure 2b).The data of bar charts are presented as mean values +/-SD.Almost complete crosslinking of 5hmU sites in cells with the 5hmU-primer probe.After the click reaction of the 5hmU-primer probe, the 5fU-primer probe was subsequently added.No RCA fluorescence signal of 5fU-primer probe was observed.It indicated all sites of 5hmU modification were crosslinked with the 5hmU-primer probe.(c) Successive labelling of 5hmU and 5fU in cells and simultaneous imaging analysis.MCF-10A cells were used.Scale bar, 10 μm.For the experiments of cell imaging in all figures of this work, five times were repeated independently with similar results.Supplementary Figure 5. Specific cell imaging using Cell-TALKING.(a) Scheme of sequential labeling of 5hmU, 5hmC and 5fU followed by the histone PTM in cells.The blocked walking probe is a DNA hybrid duplex that can be captured by the DNA-crosslinked secondary antibody.The detail is described in the protocol of Cell-TALKING as below.(b) Selected cell images for several negative controls and one positive sample.The nanoenvironment around histone H3K27ac in MCF-10A cells was detected here.Scale bar, 10 μm.(c) The spot counts of three combination sites of single cells.The cell numbers of Sample I-VII are 34, 10, 10, 10, 10, 10 and 10, respectively.The "Non-walking probe" negative control used the Non-walking probe instead of the walking probe.The Non-walking probe can not hybridize to all three barcoding probes, which fails to induce the DNA nicking reaction and walking process.The "Mismatched probes" negative control used a circularized padlock that can not hybridize to all three barcoding probes, which fails to induce RCA.And the "Pre-block modification sites" control firstly labeled all modification sites with blocked duplex barcoding probes, and then with single-stranded barcoding probes.The blocked duplex barcoding probes were prepared by the hybridization of the complementary oligo to the single-stranded barcoding probes.They can not hybridize to their circularized padlocks and fail to induce RCA.This negative control can indicate that almost all sites were labeled and crosslinked with DNA probes by the first labeling processes, and no residual sites were detected by second labeling processes.Similar results are shown in Supplementary Figure 4.The maxima, upper quartile, median, lower quartile, and minima are shown in the box plots.
the coverslips using plasma cleaner (Harrick plasma, PDC-002-HP), forming the chambers for cell culture and following chemical reactions.
Cell culture.MCF-10A, MCF-7 and MDA-MB-231 cells were cultured in Dulbecco's modified Eagle's medium with 10% fetal-bovine serum and 1% antibiotics penicillin-streptomycin (100 U/mL) in a humidified incubator containing CO2 (5%) at 37 o C. For a typical experiment, 6,000 cells were seeded on a collagen A-coated coverslip enclosed in a PDMS chamber at 37 o C overnight.
Cell fixation and permeabilization.The cells cultured on the PDMS chamber were fixed with 4% formaldehyde (Beyotime Biotechnology, cat.no.P0099) at room temperature for 10 min, and then were permeabilized with 0.5% Triton X-100 (Sangon Biological, cat.no.A110694) in 1x PBS for 5 min at room temperature.
The labeling of DNA modifications.We follow the labelling order (5hmU first, then 5hmC and finally 5fU) to achieve the discrimination of these three DNA modifications in cells. (

1-1). 5hmU phosphorylation reaction
The cells were incubated with following mixture at 37 o C for 2 h then washed with 1x PBS buffer for three times.

20x PBS 1
Water 17 Total 20 (3-1).The reduction of 5fU to 5hmU The cells were incubated with following mixture at room temperature in dark.Then the cells were washed with 1x PBS buffer for three times.

Reagents µL
Sodium borohydride (10 mg/mL, Aladdin, cat.no.S108355) 2 Water 18 Total 20 (3-2).The phosphorylation of newly generated 5hmU The cells were incubated with following mixture at 37 o C for 2 h then washed with 1x PBS buffer for three times.

ATP-γ-alkyne (10 mM) 1 10x Cutsmart buffer 2
Water 16 Total 20 (3-3).The labeling with barcoding probe-5fU The following solution was added to perform a copper(I)-catalyzed click reaction at room temperature for 1 h in dark.Then the cells were washed with 1x PBS buffer for three times.
The labeling with DNA-crosslinked secondary antibody.The cells were incubated with prepared DNAcrosslinked secondary antibody in QuickBlock™ Secondary antibody dilution buffer (Beyotime Biotechnology, cat.no.P0265) for 1h at room temperature.The cells were washed using PBST buffer three times for at least 15 minutes each time.Notably, this DNA sequence crosslinked on secondary antibody is used to capture the blocked walking probe via DNA hybridization.This design can regulate the length of walking probe without the change of DNA-crosslinked secondary antibody.And the blocked walking probe is a DNA hybrid duplex prepared as below.

The preparation of the blocked walking probe duplex
The following mixture was incubated in PCR tube at 37 º C for 2 h.

The hybridization of the blocked walking probe duplex
The cell samples were incubated with following mixture at 37 o C for 1 h then washed with 2x SSC buffer for three times.

Formamide buffer 4 20x SSC 2
Water 12 Total 20 Enzymatic and DNA reactions of Cell-TALKING.After the above labelling reactions, the Cell-TALKING was performed, which included the DNA proximity nicking/walking reaction, the hybridization reaction of circularized padlocks, 3' to 5' digestion and RCA by Phi 29 DNA polymerase, and the hybridization of fluorophore-labeled DNA probes. (

1). DNA proximity nicking/walking reaction
The cells were incubated with following mixture at 37 o C for 2 h then washed with 1x PBST buffer for three times.

Reagents µL
Nt.BbvCI nicking enzyme (New England Biolabs Ltd., cat.no.R0632S) 0.5 10x Cutsmart buffer 2 Water 17.5 Total 20 During this reaction, each walking probe molecule was isothermally released from the duplex of barcoding probe/walking probe, and was reused to recognize any nearby barcoding probes, and the cleaved barcoding probes formed new 3'-OH ends. (

2). Hybridization reaction of circularized padlocks
Then three circularized padlocks were added and captured by the barcoding probes.The following mixture was incubated at 37 o C for 3h.The cells were washed with 2x SSC buffer for three times.

Formamide buffer 4 20x SSC 2
Water 8 Total 20 The ssDNA 3'-OH overhangs in the duplexes of circularized padlocks/cleaved barcoding probes can be degraded by phi 29 DNA polymerase with its inherent 3' to 5' proofreading exonuclease activity.
(3).Barcoding RCA The cells were incubated with following mixture at 37 o C for 2 h.
The letter 'P' and the symbol '*' indicate phosphate group and phosphorothioate, respectively.The /LNA_G/ indicate the G base of locked nucleic acid (LNA).Italic letters in red indicate the nick sequence of Nt.BbvCI enzyme.Supplementary Figure 1.Other images of the experiments of in vitro Cell-TALKING performed on coverglass.(a) and (b) are corresponding to Figure 2a and 2b, respectively.
Figure 3.The structural characterization of ATP-γ-alkyne.See more details in the following Supplementary Materials.Supplementary Figure 4. Efficient labelling and imaging of DNA modifications in cells.(a) Bioorthogonal labelling of 5hmU in cells using 5-HMUDK and ATP-γ-alkyne.(b) Hybridization of fluorophore-labeled DNA probesAfter washing with 1x PBST, 200 nM of each fluorophore-labeled DNA probes in 2x SSC and 20% formamide buffer were incubated with the samples at 37 o C for 30 min.The nuclei were stained using DAPI.Cells were washed three times before fluorescence imaging.