ChIP-seq for H3K27ac Profiling of Active Enhancers in Formaldehyde-Crosslinked K562 Cells with Spike-in Normalization

Two conditions (vehicle vs. HDAC-inhibitor treated) with three biological replicates each. Each ChIP uses chromatin from ~5 x 10^6 K562 cells with a constant 5% Drosophila S2 spike-in chromatin (by cell number) added to every tube. Each IP has a matched input (5% of pre-IP chromatin saved). Sonication conditions are fixed to yield 200-500 bp fragments. An IgG isotype IP is run per condition. Samples are processed in randomized order; the analyst is blinded to condition during peak QC.

BSL-1 for K562 (use BSL-2 practices). Formaldehyde is a carcinogen and respiratory/skin sensitizer — handle methanol-free formaldehyde in a fume hood, wear nitrile gloves, lab coat, and goggles; use formaldehyde waste containers. Sonication generates aerosols — keep tubes capped and use a sealed system; wear hearing protection for bath sonicators. SDS and protease inhibitors require standard handling. Autoclave biological waste; dispose of organics per institutional guidelines.

Input control (5% reserved pre-IP chromatin) sequenced alongside each IP for background subtraction and peak calling. IgG isotype control IP quantifies non-specific bead/antibody binding. Drosophila S2 spike-in with anti-H2Av provides the exogenous normalization reference. A positive-locus qPCR check (GAPDH promoter, a known active enhancer) vs. a silent gene-desert region confirms enrichment (>10-fold) before sequencing. Antibody specificity validated by manufacturer's peptide array / dot blot.

qPCR enrichment >10-fold at active enhancers vs. background before sequencing. Post-sequencing: sharp peaks at promoters/enhancers with broad domains over super-enhancers; FRiP >1% (typically 2-10% for histone marks), >20,000 peaks, and clear separation of IP from input by cross-correlation (NSC >1.05, RSC >0.8). Spike-in reads constitute ~2-8% of total. Library insert ~250-350 bp, yield 5-30 nM.

• K562 cells, log-phase, ~5 x 10^6 per IP, in RPMI-1640 + 10% FBS
• Drosophila S2 spike-in chromatin (crosslinked, sonicated) + spike-in anti-H2Av antibody
• 16% methanol-free formaldehyde (final 1%); 2.5 M glycine (final 125 mM)
• Anti-H3K27ac antibody, ChIP-grade (e.g., 2-5 µg per IP)
• Protein A/G magnetic Dynabeads
• Lysis buffers: LB1 (50 mM HEPES-KOH pH 7.5, 140 mM NaCl, 1 mM EDTA, 10% glycerol, 0.5% NP-40, 0.25% Triton X-100), LB2, LB3 (sonication buffer with 0.1% Na-deoxycholate, 0.5% N-lauroylsarcosine) + protease inhibitors
• RIPA wash buffer; TE + 50 mM NaCl
• Elution buffer (50 mM Tris pH 8, 10 mM EDTA, 1% SDS); Proteinase K; RNase A
• AMPure XP beads; NEBNext Ultra II DNA library prep kit + UDI adapters
• Qubit dsDNA HS; TapeStation HS D1000

To perform chromatin immunoprecipitation against histone H3 lysine 27 acetylation (H3K27ac) in formaldehyde-crosslinked, sonicated K562 chromatin, with Drosophila S2 chromatin and an anti-H2Av antibody as an exogenous spike-in for quantitative cross-sample normalization. The goal is to enrich active regulatory regions (>20-fold over input at known enhancers) and build Illumina libraries that resolve enhancer-level acetylation differences between conditions.

1. Crosslink 5 x 10^6 K562 cells with 1% formaldehyde, 10 min, room temp with rotation; quench with 125 mM glycine 5 min. Repeat for S2 spike-in cells.
2. Wash cells 2x cold PBS; pellet 600 x g, 5 min, 4°C.
3. Lyse sequentially in LB1 (10 min, 4°C), LB2 (10 min, 4°C), resuspend nuclei in LB3.
4. Sonicate to 200-500 bp (e.g., Covaris focused-ultrasonicator or bioruptor; verify by agarose gel on a de-crosslinked aliquot).
5. Clarify 20,000 x g, 10 min, 4°C; combine K562 chromatin with 5% S2 spike-in chromatin.
6. Reserve 5% as input. To remainder add anti-H3K27ac (+ spike-in anti-H2Av) and pre-blocked Protein A/G beads; rotate overnight at 4°C.
7. Wash beads 5x RIPA, 1x TE+NaCl, 4°C.
8. Elute 65°C 30 min in elution buffer; reverse crosslink (input + IP) 65°C overnight.
9. RNase A 37°C 30 min, Proteinase K 55°C 1 h; purify DNA with AMPure XP (1.8x), elute 30 µL.
10. Build libraries with NEBNext Ultra II (end-prep, UDI adapter ligation, 0.9x/0.9x cleanup, 8-12 cycle PCR).
11. Size-select 0.6x/0.8x SPRI; QC by Qubit + TapeStation (insert ~250-350 bp).

Independent variable: treatment condition (vehicle vs. HDAC inhibitor). Dependent variables: H3K27ac peak intensity (spike-in normalized), number of enriched regions, fraction of reads in peaks, and enrichment fold over input at reference loci. Controlled variables: cell number per IP (5 x 10^6), antibody lot and amount, spike-in fraction (5%), sonication fragment size, crosslinking time, and library PCR cycle number.

If a fixed ratio of Drosophila spike-in chromatin is added to each K562 sample prior to IP and normalized using spike-in-derived scaling factors, then genuine global changes in H3K27ac between conditions (which standard read-depth normalization would mask) will be quantitatively recoverable, yielding accurate fold-change estimates at active enhancers.

Align K562 reads to GRCh38 and spike-in reads to the Drosophila dm6 genome with Bowtie2. Compute spike-in scaling factors from dm6 read counts (reference-adjusted). Remove duplicates and blacklist regions. Call peaks with MACS2 (broad mode --broad for histone marks, or narrow for sharp enhancer peaks) against matched input. Generate spike-in-scaled bigWigs (deepTools bamCoverage --scaleFactor). Annotate peaks (HOMER/ChIPseeker), define super-enhancers with ROSE.

1. No enrichment by qPCR: antibody may be inactive or sonication too harsh — verify fragment size (200-500 bp), increase antibody, and confirm crosslinking. 2. High IgG background: increase wash stringency (more RIPA washes) and pre-block beads more thoroughly. 3. Over-sonication (<150 bp): reduce sonication cycles — fragments too small lose nucleosome context. 4. Low library yield: increase PCR cycles modestly and verify input DNA mass (>1 ng); avoid over-crosslinking which reduces recoverable DNA. 5. Inconsistent spike-in fraction across samples: ensure spike-in chromatin is added by precise cell-equivalent before IP, from a single mixed master.

Differential binding with DiffBind/DESeq2 on a consensus peak set using spike-in-derived size factors instead of library-size normalization, n=3 biological replicates per condition, negative-binomial Wald test, Benjamini-Hochberg FDR at α = 0.05. Report log2FC and padj; require |log2FC| > 1. With n=3, ~80% power to detect 2-fold acetylation changes at well-covered enhancers (mean count >50). Cross-correlation QC (phantompeakqualtools) reported per sample.