Native ChIP-seq of H3K4me3 and H3K27me3 Bivalent Domains in Human iPSC-Derived Neural Progenitor Cells Using Micrococcal Nuclease Digestion

One cell type (iPSC-NPC) profiled for two marks plus input, biological triplicate (n=3) = 9 ChIPs. 1-2 million cells per mark. Each ChIP paired with a 10% sonication-free MNase input. Spike-in chromatin (Drosophila, EpiCypher) added for normalization. A terminally differentiated neuron sample included as a comparison for bivalency resolution. Antibody validation by SNAP-CUTANA nucleosome spike-in panel. Tubes coded and randomized for IP and library steps; operators blinded during peak-overlap QC.

BSL-2 for human iPSC/NPC culture (use a biosafety cabinet, treat as potentially infectious primary-derived material). MNase and protease handling require gloves. EGTA/EDTA buffers are mild irritants. Standard biohazard waste for cells, nuclei, and antibody reagents; sharps in puncture-resistant containers. Wear lab coat, gloves, and eye protection; decontaminate surfaces with 70% ethanol after nuclei work.

Negative control: normal IgG ChIP — minimal enrichment, FRiP <2%. Input control: 10% non-immunoprecipitated mononucleosomes for peak-calling background and normalization. Antibody specificity: SNAP-CUTANA spike-in nucleosome panel run with each antibody to confirm on-target binding and low cross-reactivity. Spike-in chromatin (Drosophila) for quantitative cross-sample scaling. MNase titration control to confirm mononucleosome predominance before IP.

MNase digestion yielding ~80% mononucleosomes (clear 150 bp band). H3K4me3: sharp peaks at active TSS (FRiP >25%). H3K27me3: broader domains over developmental genes. Bivalent TSS (both marks within ±1 kb) >8% of H3K4me3 promoters in NPCs. IgG FRiP <2%. Strong concordance of replicate peak sets (Jaccard >0.6). Library insert ~200-300 bp.

• Human iPSC-derived NPCs, 1-2 x10^6 nuclei per ChIP
• Nuclei isolation buffer: 10 mM Tris pH 7.5, 10 mM NaCl, 3 mM MgCl2, 0.1% NP-40, protease inhibitors
• Micrococcal nuclease (NEB M0247), titrated (e.g., 2-20 U per reaction)
• Anti-H3K4me3 (Active Motif 39159 or Diagenode C15410003), 1-2 µg
• Anti-H3K27me3 (Cell Signaling C36B11), 1-2 µg; normal IgG control
• Protein A/G Dynabeads (Thermo 88803)
• Native ChIP buffer: 20 mM Tris pH 8.0, 150 mM NaCl, 2 mM EDTA, 0.1% Triton X-100
• Drosophila spike-in chromatin + anti-Drosophila-H2Av antibody (EpiCypher 61686/13-0001)
• Proteinase K, MinElute columns, NEBNext Ultra II library kit, AMPure XP beads

To profile two histone modifications, H3K4me3 (active) and H3K27me3 (repressive), in native chromatin from human iPSC-derived neural progenitor cells (NPCs) by digesting chromatin to mononucleosomes with micrococcal nuclease and immunoprecipitating each mark separately. Native ChIP preserves true histone-DNA associations without formaldehyde-induced epitope masking, giving sharp, high-resolution mononucleosome-level maps ideal for detecting bivalent promoters.

1. Isolate nuclei from 1-2x10^6 NPCs in cold nuclei buffer, 5 min on ice; pellet 600xg 5 min 4°C.
2. Resuspend in MNase buffer (with CaCl2); digest 37°C 10 min with titrated MNase to yield ~80% mononucleosomes; stop with 10 mM EGTA on ice.
3. Lyse nuclei (brief hypotonic + gentle pipetting); clarify 10,000xg 10 min; recover soluble mononucleosomes (S1) and verify size on TapeStation (~150 bp).
4. Reserve 10% as input; split remainder for each antibody; add spike-in chromatin + spike antibody.
5. Add 1-2 µg primary antibody to each; rotate 4°C overnight.
6. Add 20 µL pre-blocked protein A/G Dynabeads; rotate 4°C 3 h.
7. Wash beads 4x in cold ChIP buffer (increasing NaCl to 200 mM on later washes), 5 min each.
8. Elute + digest with Proteinase K 55°C 2 h; purify on MinElute; elute 20 µL.
9. Build libraries (NEBNext Ultra II, no size selection, 10-12 cycles); sequence PE 50 to 20-30M reads per sample.

Independent variable: histone mark identity (H3K4me3 vs H3K27me3) and differentiation state (NPC vs neuron). Dependent variables: per-promoter normalized read density for each mark and bivalency status (co-occurrence). Controlled variables: cell number, MNase digestion level (~80% mononucleosomes), antibody amount, wash stringency, spike-in mass, and sequencing depth.

We hypothesize that a defined set of developmental and neuronal transcription-factor promoters (e.g., NEUROD1, GATA2, PAX6 targets) will carry both H3K4me3 and H3K27me3 in NPCs (bivalent), and that the fraction of bivalent TSS (co-marked within ±1 kb) will exceed 8% of all H3K4me3-positive promoters, more than in terminally differentiated controls.

Trim (Trim Galore), align to hg38 + Drosophila with Bowtie2; deduplicate (Picard). Compute spike-in scale factors from Drosophila reads; generate normalized bigWigs (deepTools). Call H3K4me3 (narrow, MACS2) and H3K27me3 (broad, MACS2 --broad or SEACR) using input as control. Define bivalent domains by intersecting H3K4me3 and H3K27me3 peaks within ±1 kb of RefSeq TSS (bedtools). Profile metaplots/heatmaps (deepTools computeMatrix/plotHeatmap).

Over-digestion (sub-mononucleosomal smear, naked DNA): reduce MNase units or time; titrate per cell prep. Under-digestion (oligonucleosome ladder dominates): increase MNase or digestion time to reach ~80% mononucleosomes. Weak H3K4me3 signal: epitope-validated antibody and avoid over-washing; H3K4me3 is sensitive to high-salt loss. Poor bivalency detection: ensure both IPs came from the same mononucleosome pool and adequate H3K27me3 depth (broad domains need more reads). High IgG background: pre-block beads and increase final wash NaCl to 250 mM.

Differential mark occupancy between NPC and neuron tested with DiffBind/csaw (edgeR negative binomial GLM, spike-in-scaled, n=3); BH FDR q<0.05, |log2FC|>1. Bivalency enrichment over random expectation tested by permutation (regioneR, 1000 shuffles) reporting empirical p. Replicate reproducibility quantified by IDR (irreproducible discovery rate <0.05 for retained peaks). Power: triplicate native ChIP at 25M reads detects 2-fold promoter changes at >0.8 power.