Which Cell Line Is Best for Studying DNA Methylation?

Choose by the question, then verify the line's methylation machinery and baseline at your locus:

• • Mechanism of maintenance or de novo methylation → a line with a well-characterized DNMT background (HCT116 is the classic, with documented alterations at DNMT1 and DNMT3B; verify rather than assume).
• • Editing-based studies (CRISPRoff, TET) → an easy-to-edit, fast-growing line with a stable karyotype and low background silencing, such as K562 for haematopoietic questions.
• • Tissue-specific or developmental methylation → a primary-like or iPSC-derived system, accepting slower workflows.
• • Diagnostic or biomarker work → a matrix-matched line, since methylation patterns are tissue-specific.

Avoid choosing a line because it is convenient and then treating its methylome as normal. Most popular lines carry documented abnormalities.

The last row is the one most often ignored: a slow-growing line needs longer timepoints to accumulate the same methylation change.

• • HCT116. Widely used for methylation work because its DNMT background is well studied, but it carries documented alterations in DNMT1 and DNMT3B. Excellent for maintenance questions; not a normal reference methylome.
• • HEK293. Easy to transfect and edit, hypotriploid, with an adenoviral insert; convenient for tool development, weak as a physiological model.
• • K562. Near-diploid, fast-growing, well-characterized chromatin; strong choice for haematopoietic regulatory questions and for editing panels.
• • HeLa. Aneuploid and highly rearranged; acceptable for tool validation, poor for dosage-sensitive claims.
• • H9/human iPSCs. Most physiologically relevant for developmental methylation, at the cost of slower growth, differentiation requirements, and imprinting instability in culture.
• • Primary cells. Best tissue fidelity, limited lifespan, and variable editing efficiency.

None of these is a defect of the line — it is a property to declare.

"Does methylation at this promoter silence this gene?" You need a line in which the promoter is methylated and the gene silent, plus the ability to demethylate it. Verify the baseline first; many immortalized lines have partial methylation at CpG islands.

"How is methylation maintained?" A DNMT-characterized line plus a knock-in or knockout panel; the maintenance question requires division-linked measurements.

"Does this variant alter methylation?" An isogenic panel in a line with a stable karyotype and reliable single-cell cloning.

"What does this mark do in this tissue?" A differentiated model matched to the tissue — accepting that editing efficiency and growth will be worse.

Three disclosures make the choice defensible:

1. State the known abnormalities of the line relevant to methylation.
2. Report the baseline methylome at the loci you study, not just the change.
3. Confirm the key finding in a second background where feasible, especially for claims about direction of effect.

That third point is what distinguishes a tool-development paper from a biological claim.

EDITGENE provides cell lines with documented DNMT/TET status, baseline methylation maps at customer-specified loci, imprinting checks where relevant, karyotype notes, and editing efficiency data.

Tell us the locus and the question. Contact EDITGENE to request a project consultation or quote.

Cell line characteristics, including DNMT/TET alterations, vary by subclone and passage; the descriptions here are general and should be verified for the specific stock used.

Methylation patterns are tissue-specific. Findings in an immortalized line should be corroborated in a matched system before being generalized.

• • Rhee I, et al. CpG methylation is maintained in human cancer cells lacking DNMT1. Nature. 2000;404:1003–1007.
• • Weisenberger DJ, et al. Role of the DNA methyltransferase variant DNMT3b3 in DNA methylation. Molecular Cancer Research. 2004;2:62–72.
• • Loyfer N, et al. A DNA methylation atlas of normal human cell types. Nature. 2023;613:355–364.
• • Bar S, et al. Epigenetic aberrations in human pluripotent stem cells. EMBO Journal. 2019;38:e101033.