Microwell-seq 1.1 protocol

Materials

We taught a local company (G-Bio Biotechnology Co. Ltd) to provide Microwell-seq Kit/Service http://www.igeneseq.com/research_detail/dxbcp/microwell.html.

Fabrication of microwell device

The diameter of the microwells used in microwell-seq 1.1 was 28 μm and the depth is 35 or 30 μm, depending on different cell sizes, with the diameter 32μm and the depth 45 μm used in Microwell ATAC-seq to accommodate more nuclei. The new silicon plate was ordered from Suzhou Cchip Scientific Instrument Co.,Ltd.. The silicon microwell plate was then used as a mould to make a polydimethylsiloxane (PDMS) plate with the same number of micropillars. Before the experiments, a disposable agarose microwell plate was made by pouring 5% agarose solution onto the surface of the PDMS plate. Both the silicon and the PDMS plates are reusable.

Synthesis of barcoded beads

We optimized the previous protocol to improve the efficiency of three-segment molecular chain synthesis. Magnetic beads coated with carboxyl groups were provided by Suzhou Knowledge & Benefit Sphere Tech (diameter 20 μm used in Microwell-seq 1.1, diameter 28 μm used in Microwell ATAC-seq). The barcoded oligonucleotides on the surface of the beads were synthesized by three rounds of split-pool. All the sequences used for the Microwell-seq 1.1 beads and Microwell ATAC-seq beads were listed in supplementary Supplementary Table 4.
   For first batch of bead synthesis, 300–350 μL carboxyl magnetic beads (50 mg/mL) were washed twice with 0.1 M 2-(N-morpholino) ethanesulfonic acid (MES). The beads were then suspended in a final volume of 635 μL of 0.1 M MES. 1-Ethyl-3-(3-dimethyl aminopropyl) carbodiimide hydrochloride (EDC; 3.08 mg) was added to the beads, and 6.2 μL beads was then placed in each well of a 96-well plate. Amino-modified oligonucleotides (2.5 μL, 50 μM in 0.1 M MES) were then added to each well. After vortexing the mixture and incubating it for 20 min at ambient temperature, we distributed a 0.5-μL mix (6 mg EDC in 100 μL of 0.1 M MES) into each well. After an additional round of vortexing and incubation for 20 min at ambient temperature, an additional 0.5 μL mix (6 mg EDC in 100 μL of 0.1 M MES) was distributed into each well. After vortexing and incubation for 80 min at ambient temperature, the beads were collected in 1 mL of 0.1 M phosphate-buffered saline (PBS) containing 0.02% Tween 20. After centrifugation, the supernatant was carefully removed. The beads were then washed twice in 1 mL TE (10 mM Tris-HCl, 1 mM EDTA, pH 8.0).
   In the second split-pool, the beads were washed twice with 10 mM Tris-HCl (pH 8.0) and divided to the wells of another 96-well plate with beads mix consisting of 1x NEBuffer™ 2, 1 mM dNTP (5 μL each well), and 50 μM oligonucleotides were added (1 μL each well). The oligonucleotides in each tube encoded a sequence with reverse complementarity to linker 1, a unique barcode and a linker 2 sequence. The plate was then incubated in 95°C PCR machine for 10 seconds with pre-shaken and it was quickly followed by mixing beads on a rotary mixer (10 rpm) at room temperature for 10-20 min to promote the binding of the oligos. After that, we added the polymerase mastermix [1× NEBuffer™ 2, 200 U/mL Klenow (NEB)] to the plate (2.5 μL each well) and mixed them. The polymerization was carried out by incubating the beads for 1 hour at 37°C on a rotary mixer (10 rpm). Next, the plate was placed on ice and 0.15M EDTA was added to the plate (1.5 μL each well). After a short centrifugation, all beads were collected into one tube on ice. The beads were resuspended and incubated in 500 µL 0.1 M NaOH at room temperature for 30 sec twice to thoroughly denature the hybrid. Following denaturing, the NaOH was removed and beads were washed twice with 1mL TE-TW [10 mM Tris (pH 8.0), 1 mM EDTA, 0.01% Tween 20] and once with 200 µL of 10 mM Tris-HCl (pH 8.0).
   The third split-pool procedure was the same as the second one. The oligonucleotides used in each tube encoded a linker 2 reverse-complementary sequence, a unique barcode, a UMI sequence and a poly-T tail. After the third synthesis, the beads were washed twice by TE-SDS (1 × TE + 0.5% SDS) to inactivate the Klenow polymerase and then washed once by TE-TW and once by 10 mM Tris-HCl. To remove the chains without the third barcoded sequence, the beads were collected on ice and suspended in 200 μL exonuclease I mix (containing 1 × exonuclease I buffer and 1 U/μL exonuclease I) and incubated at 37 °C for 15 min on a rotary mixer (10 rpm), and then the reaction was terminated by washing with TE-SDS and TE-TW. To remove complementary chains, the beads were incubated in 0.1 M NaOH as mentioned above. The beads were then washed twice with 1mL TE-TW, and could be stored in 1 mL TE-TW for 4 weeks at 4 °C.

Protocol

Single cell collection and lysis

The cell concentration should be carefully controlled during Microwell-seq. Both cell and bead concentrations were estimated using a haemocytometer. The beads were resuspended in PBS before loading. The proper cell concentration is ~200,000/mL (with 10% of the wells occupied by single cell). The proper bead concentration is ~1,000,000/mL (with every well occupied by single bead). An evenly distributed cell suspension (~500mL) was pipetted onto the microwell array. To eliminate cell doublets, the plate was inspected under a microscope. Cell doublets were reduced by pipetting over the region of high cell density. After the expected number of cells were evenly and tightly dropped in the microwells, the supernatant containing the excess cells would be aspirated away and the bead suspension (~500mL) was then loaded into the microwell plate, and the plate was placed on a magnet, with beads being quickly shaken into the microwells. Excess beads were washed away slowly. Cold lysis buffer [0.1 M Tris-HCl (pH 7.5), 0.5 M LiCl, 1% sodium dodecyl sulfate (SDS), 10 mM EDTA, and 5 mM dithiothreitol] was added to the surface of the plate and removed after 12 min incubation. The beads were then collected, transferred to an RNase-free tube, and washed once with 1 mL of 6× SSC, once with 500 µL of 6× SSC, and once with 200 µL of 50 mM Tris-HCl pH 8.0. Finally, ~90,000 beads were collected in a 1.5 mL tube.

Reverse transcription

In this procedure, the instructions from the Smart-seq2 protocol were followed. Briefly, 20 µL RT mix was added to the collected beads. The RT mix contained 200 U SuperScript II reverse transcriptase, 1 × Superscript II first-strand buffer (Takara), 40 U Murine RNase inhibitor (Vazyme), 1 M betaine (Sigma), 6 mM MgCl2 (Ambion), 2.5 mM dithiothreitol, 1 mM deoxynucleoside triphosphate, and 1 µM TSO LNA primer. The sequences information for the primers was included in Supplementary Table 5. The beads were incubated at 42°C for 90 min with mixing on a rotary mixer (10 rpm) and then washed with 200 µL TE-SDS (1× TE + 0.5% SDS) to inactivate reverse.

Exonuclease I treatment

The beads were washed once with 200 µL TE-TW and once with 200 µL of 10 mM Tris-HCl (pH 8.0), resuspended in 100 µL exonuclease I mix containing 1× exonuclease I buffer and 50 U exonuclease I (NEB), and incubated at 37°C for 60 min with mixing on a rotary mixer (10 rpm) to remove oligonucleotides that did not capture mRNA. The beads were then pooled and washed once with TE-SDS, once with 1 mL TE-TW, and once with 200 µL of 10 mM Tris-HCl (pH 8.0).

Second Strand Synthesis

To improve the amplification efficiency, an additional second-strand synthesis step was introduced. The beads were resuspended and incubated in 200 µL 0.1 M NaOH at room temperature for 30 sec twice to thoroughly denature the mRNA-cDNA hybrid and remove mRNAs. Following denaturing, the NaOH was removed and beads were washed twice with 1mL TE-TW and once with 200 µL of 10 mM Tris-HCl (pH 8.0). The beads were resuspended in 20 µL dn-TSO oligo mix containing 5mM dn-TSO oligos, 10 mM Tris-HCl (pH 8.0) and 3mM MgCl2. The beads were then incubated in 95°C water bath for 30 seconds and quickly followed by mixing on a rotary mixer (10 rpm) at room temperature for 10 min to promote the binding of dn-TSO oligos to multiple sites of the single chains on beads. The sequence information is included in supplementary Table 5. After that, the supernatant was aspirated away, the beads were washed once with 100 µL of 10 mM Tris-HCl (pH 8.0) without being resuspended, and then combined with a 50 μL mastermix consisting of 1x RT buffer（Thermo）, 12% PEG8000 solution, 1mM dNTPs, and 6.25 U Klenow exo- (Vazyme). Second-strand synthesis was carried out by incubating the beads for 1 hour at 37°C on a rotary mixer (10 rpm). The beads were then washed twice with 200 µL TE-TW and once with 10 mM Tris-HCl (pH 8.0).

cDNA amplification

The beads from the same microwell plate were transferred into one PCR tube. To each tube, 25 µL PCR mix [1× HiFi HotStart Readymix (Kapa Biosystems) and 0.4 µM TSO-PCR primer (The sequences information is included in Supplementary Table 5)] was added. The PCR program was as follows: 95°C for 3 min; three cycles of 98°C for 20 s, 65°C for 45 s, and 72°C for 6 min; 72°C for 5 min; and a 4°C hold. After pooling all PCR products, 50 µL PCR mix [1× HiFi HotStart Readymix and 0.4 µM TSO-PCR primer was added to each DNA sample. The PCR program was as follows: 95°C for 3 min; 12 cycles of 98°C for 20 s, 67°C for 20 s, and 72°C for 3 min; 72°C for 10 min; and a 4°C hold. Finally, 0.9X VAHTS DNA Clean beads (Vazyme) were used to purify the cDNA library.

Transposase fragmentation, selective PCR and sequencing

The purified cDNA library was fragmented using a customized transposase that carries two identical insertion sequences. The customized transposase was included in the TruePrep Homo-N7 DNA Library Prep Kit for Illumina (Vazyme) or TruePrep Homo-N7 DNA Library Prep Kit for MGI (Vazyme). The fragmentation reaction was performed according to the instructions provided by the manufacturer. We used P5 primer for illumina and VAHTS RNA Adapters set 3-set 6 for illumina (Vazyme), or P5 primer for MGI and Indexed P7 primers for MGI to specifically amplify fragments that contain the 3’ ends of transcripts. Other fragments will form self-loops, impeding their binding to PCR primers. Meanwhile, we recorded the correspondence between p7 index and samples. The PCR program was as follows: 72°C for 3 min; 98°C for 1 min; five cycles of 98°C for 15 s, 60°C for 30 s, and 72°C for 3 min; 72°C for 5 min; and a 4°C hold. The PCR product was purified using 0.9X VAHTS DNA Clean beads. Then, 25 µL PCR mix (1×HiFi HotStart Readymix and 0.2 µM 2100 primer P5&P7 for illumina or MGI) was added to each sample. The sequences are listed in supplementary Table 5. The PCR program was as follows: 95°C for 3 min; five cycles of 98°C for 20 s, 60°C for 15 s, and 72°C for 15 s; 72°C for 3 min; and a 4°C hold. To eliminate primer dimers and large fragments, 0.55-0.15X VAHTS DNA Clean beads were then used to purify the cDNA library. The size distribution of the products was analysed on an Agilent 2100 bioanalyser, and a peak in the 400 to 700 bp range was observed. Finally, the samples were subjected to sequencing on the Illumina HiSeq or MGI DNBSEQ-T7. For MGI sequencing, we applied the protocal provided by VAHTS Circularization Kit for MGI (Vazyme) to obtain single-stranded circular cDNA available for DNB (DNA Nanoball) generation. We also replaced the official read1 sequencing primers with our custom read1 X-linker 1&2 (The sequences information is included in Supplementary Table 5) to ensure the completion of the sequencing.

Table 1 Primer used in Microwell-seq 1.1

Primer	Sequence
TSO LNA primer	ACACTCTTTCCCTACACGACGCCCAArGrG+G
dn-TSO oligos	ACACTCTTTCCCTACACGACGCCCANNNGGNNNB
TSO-PCR primer	ACACTCTTTCCCTACACGAC*G*C
P5 primer for illumina	AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCT*T*C
P5 primer for MGI	5'P-GAACGACATGGCTACGATCCGACTTACACTCTTTCCCTACACGACGCTCT*T*C
Universal P5 primer for MGI	5'P-GAACGACATGGCTACGATCCGACTTTCGTCGGCAGCGTC
2100 primer P5 for Illumina	AATGATACGGCGACCACCGA
2100 primer P7 for Illumina	CAAGCAGAAGACGGCATACGA
2100 primer P5 for MGI	5'P-GAACGACATGGCTACGATCCGAC
2100 primer P7 for MGI	TGTGAGCCAAGGAGTTGTTGTCTTC
Custom Read1 X-linker 1	CGCCGACGCACAGGGTGCCTCGACCGCATGGCGCGGAACCATGGTTCCGCGCACACTCTTTCCCTACACGACGCTCTTCCGATCT
Custom Read1 X-linker 2	CATGCGGTCGAGGCACCCTGTGCGTCGGCGGGCTGCATGCCGGCATGCAGCCACACTCTTTCCCTACACGACGCTCTTCCGATCT

Download oligo and barcoded beads used in Microwell-seq 1.1.

Microwell ATAC-seq protocol

Protocol

Tn5 tagging of nuclei

We took frozen fixed nuclei stored at -80°C and placed them on dry ice. With a 37°C water bath for 0.5-1 min, we thawed the nuclei aliquots and spun them at 600 × g for 5 min at 4°C to pellet the nuclei. The nuclei were washed twice with PBS-B buffer, resuspended in a small volume of PBS-B and counted using a hemocytometer with 1 % DNA double-stranded dye Ultra GelRed (Vazyme) under a fluorescence microscope to control the density to 200,000/mL. The Tn5 tagging plate was taken out, 6 µL tagmentation buffer mix (2.5 µL DEPC treated water, 2.5µL 4 × TD buffer and 1µL 0.1% Digitonin) to each well of the plate and centrifuged at 3000 × g for 5 min at 4°C to mix well. Then we added 2µL single-nuclei suspension to each well and recorded the tissue sources and mouse number. Tn5 tagging was carried out by incubating the nuclei at 55°C for 30 min on a rotary mixer (10 rpm). Tagmentation reaction was stopped by adding 10 µL 40mM EDTA to each well of the plate and incubated at 37°C for 10 min on a rotary mixer (10 rpm). After that, 20 µL of wash buffer was added to each well of the plate to adjust the osmotic pressure. Then tagmented and barcoded nuclei from each tissue sample in one plate were pooled to one 15mL centrifuge tube, washed with 3 mL cold RSB buffer and operated separately following the next steps to avoid contamination and confusion of sample sources.

SDS Stripping and exonuclease I treatment

The nuclei were suspended in 1.5 mL RSB-SDS in 15mL centrifuge tubes, incubated at room temperature for 10 min to release the binding sites of terminal deoxyribonucleotidyl transferase (TdT). After that, 8.5 mL RSBT-BD buffer were added to each of the tubes to quench the reaction and the nuclei were centrifuged at 600 × g for 5 min at 4°C. The pellet was wash by 1 mL RSBT-BD buffer twice and resuspended in 100 µL exonuclease I mix containing 1 × exonuclease I buffer and 50 U exonuclease I (NEB), and the mixture in each tube was then evenly distributed to one nuclease-free 8-strip tube, each well 12.5 µL, and incubated on ice for 5 min. The exonuclease I incision program was as follows: 37°C for 10 min; 45°C for 5 min; 50°C for 5 min; 55°C for 5min and a 55°C hold. Then the nuclei were quickly transferred onto ice and incubated for 3 min. Next, 12.5 µL exonuclease I mix containing 1 × exonuclease I buffer and 50 U exonuclease I (NEB) was added to each well of the 8-strip tubes. The above exonuclease I incision program was repeated to completely remove the primer residues. Then the nuclei from the same Tn5 tagging plate were collected into one 15 mL centrifuge tube and washed with cold WB-T buffer and then RSBT-B buffer.

Adding A-tail to DNA fragments

We counted the nuclei using a hemocytometer with 1 % DNA double-stranded dye Ultra GelRed under a fluorescence microscope to control the concentration of nuclei, ~1000,000 nuclei/1.5 mL tubes. The nuclei were pelleted at 600 × g for 5 minutes and the supernatant was carefully aspirated as thoroughly as possible. Then the nuclei were resuspended in 50 µL TdT mix containing 1 × TdT buffer, 5 mM CoCl2, 100 µM dATP, 400 U TdT (Roche) and incubated for 20 min at 37°C on a rotary mixer (10 rpm). After that, 50 µL 40 mM EDTA was added to the tube to stop the reaction.

Beads preparation

Corresponding aliquots of beads were taken out (~500,000 per microwell), incubated twice in 200 µL 0.1 M NaOH at room temperature for 30 sec, then washed twice with 1 mL TE-TW, twice with PBS-E and suspended in 500 µL PBS-E.

Nucleus collection and lysis

The nucleus concentration was carefully controlled to ~400,000/mL using a hemocytometer with 1 % DNA double-stranded dye Ultra GelRed under a fluorescence microscope. An evenly distributed cell suspension （~500mL）was pipetted onto the microwell array. Then we spun the microwell plates at 300 × g for 15 s at 4°C to speed up the loading of nuclei with most of the wells tightly occupied by one or two nuclei. Then the supernatant containing the excess cells would be aspirated away and the bead suspension（~500mL）was loaded into the microwell plate. The plate was then placed on a magnet, with beads being quickly shaken into the microwells. Excess beads were washed away slowly. Lysis buffer [1% SDS + 1 × Blue buffer (Vazyme) + 5% PEG 8000 +1 × PCR Enhancer (Vazyme) + 10mM EDTA + 1 mg/mL protease K (Sangon Biotech)] was added to the surface of the plate and removed after 30 min incubation on the magnet inside the 37°C incubator. The beads were then collected in 2 × SSC in a nuclease-free dish, transferred into a nuclease-free tube, and washed once with 500 µL of 2 × SSC, and once with 200 µL of 10 mM Tris-HCl pH 8.0.

Exonuclease I treatment and DNA strands extension

After aspirating the supernatant from the tube on the magnetic stand, we added 50 µL exonuclease I mix containing 1 × exonuclease I buffer and 50 U exonuclease I (NEB) and incubated the mixture for 15 min at 37°C on a rotary mixer (10 rpm) to remove unbound A-tail single strands of beads and prevent contamination. The beads were washed with 200 µL TE-SDS to inactivate the incision, and then washed once with 200 µL TE-TW and once with 200 µL of 10 mM Tris-HCl (pH 8.0). After that, the beads were resuspended in 50µL extending mix containing 1 × NEBuffer™ 2, 1 × PCR Enhancer, 5% PEG8000 + 1 mM dNTP + 6.25 U Klenow (NEB), and incubated for 30min at 37°C on a rotary mixer (10 rpm) to extend the DNA strands. The beads were washed with 200 µL TE-SDS to stop the extension. The beads were then washed twice with 200 µL TE-TW and once with 10 mM Tris-HCl (pH 8.0).

cDNA amplification, library construction and sequencing

To each tube, 50 µL PCR mix [1× HiFi HotStart Readymix (Kapa Biosystems) 0.4 µM ATAC-F primer and 0.4 µM ATAC-R primer] was added and mixed, and then distributed to four wells of the 8-strip tube, 12.5µL for each well. The sequences information is included in supplementary Table 6. The PCR program was as follows: 95°C for 3 min; two cycles of 98°C for 30 s, 65°C for 60 s, and 72°C for 60s; four cycles of 98°C for 30 s, 65°C for 30 s, and 72°C for 60s; 72°C for 3 min; and a 4°C hold. We aspirated the supernatant from the tube on the magnetic stand and transferred the solution into new tubes. After that, the PCR product was purified using 0.9X VAHTS DNA Clean beads and the beads in each well were eluted with 21 µL DEPC treated water. Then, 29 µL master mix (1×HiFi HotStart Readymix, 0.4 µM P5-ATAC primers and 0.4 µM indexed P7-ATAC primers) was added to each well of the tube. The sequences information is included in Supplementary Table 6. Meanwhile, we recorded the correspondence between p7 index and samples. The PCR program was as follows: 95°C for 5 min; two cycles of 98°C for 30 s, 65°C for 60 s, and 72°C for 60s; eight cycles of 98°C for 30 s, 65°C for 30 s, and 72°C for 60s; 72°C for 3 min; and a 4°C hold. To eliminate primer dimers and large fragments, 0.4-0.5X VAHTS DNA Clean beads were then used to purify the cDNA library. Finally, the samples were subjected to sequencing on the MGI DNBSEQ-T7. For MGI sequencing, we applied the protocal provided by VAHTS Circularization Kit for MGI (Vazyme) to obtain single-stranded circular cDNA available for DNB (DNA Nanoball) generation. We also replaced the official R1 sequencing primers with Custom 2.0 Read1 X-linker 1&2 to ensure the completion of the sequencing. The sequences information is included in Supplementary Table 6.

Table 2 Primer used in Microwell ATAC-seq

Primer	Sequence
primer A	CTGTCTCTTATACACATCT
Custom 2.0 Read1 X-linker 1	CGCCGACGCACAGGGTGCCTCGACCGCATGGCGCGGAACCATGGTTCCGCGCCCGCGGAAGCAGTGGTATCAACGCAGAGTACGT
Custom 2.0 Read1 X-linker 2	CATGCGGTCGAGGCACCCTGTGCGTCGGCGGGCTGCATGCCGGCATGCAGCCCCGCGGAAGCAGTGGTATCAACGCAGAGTACGT

Download ologo and barcoded beads used in Microwell ATAC-seq.