Founder's role line (READ): "Deposits the repressive H3K27me3 histone mark to silence developmental and lineage-specifying genes." Key components (READ): EZH2, EED, SUZ12, RBBP4
DRY status
DRY: none located. Queries run:
"dry fasting" DNA methylation epigenetic OR immune OR VEGF OR Myc OR EZH2 human study(English)сухое голодание метилирование ДНК иммунитет лимфоциты исследование Филонов(Russian)
MEASURED: neither opened human dry-fasting study reports EZH2, EED, SUZ12, RBBP4 or H3K27me3.
Disconfirming and null first
- EZH2 itself was UNCHANGED by fasting. In the one study that put a fasting intervention and the H3K27 mark in the same experiment, EZH2 expression did not move. MEASURED: "Fasting left EZH2 expression unchanged, and UTX was also unchanged." Source: Byun S, Seok S, Kim YC, Zhang Y, Yau P, Iwamori N, Xu HE, Ma J, Kemper B, Kemper JK. "Fasting-induced FGF21 signaling activates hepatic autophagy and lipid degradation via JMJD3 histone demethylase." Nat Commun. 2020;11:807. doi:10.1038/s41467-020-14384-z. PMID 32042044. PMCID PMC7010817. This is the central honest finding for node 35: fasting moved the mark, not the writer.
- EED, SUZ12 and RBBP4 were not measured in that study. MEASURED: they do not appear in the reported results. Three of the founder's four key components for this node are unmeasured under fasting.
- An ADF study that profiled H3K27me3 genome-wide in mouse liver reported no PRC2 subunit finding. Korenfeld N, Charni-Natan M, Bruse J, et al. "Repeated fasting events sensitize enhancers, transcription factor activity and gene expression to support augmented ketogenesis." Nucleic Acids Research. 2025;53(1):gkae1161. doi:10.1093/nar/gkae1161. PMID 39673515. MEASURED: H3K27me3 was one of four ChIP antibodies used; no PRC2 component is named in the reported results. Null for the writer complex, in a study that had the mark in hand.
- Human: no fasting intervention. Byun 2020 includes 15 human liver samples, but those are normal versus steatosis versus NASH-fibrosis disease-state comparisons, not a fasting intervention, and MEASURED: human H3K27me3 and PRC2 data are not reported. The authors describe the human results as correlative.
- Direction of the nearest human-adjacent evidence runs the other way for diet generally. EZH2-mediated H3K27 trimethylation in mouse liver is reported as an early event induced by high-fat diet exposure (located: Nutrients 2024;16(19):3260, mdpi.com; NOT opened, see debts). That is a feeding intervention, not fasting, and it is listed here only so the node is not read as "fasting acts on EZH2."
Graded rows
| # | Claim | Modality | Species / system | Tier | Grade | Source | Tag |
|---|---|---|---|---|---|---|---|
| 35.1 | Fasting markedly decreased hepatic H3K27me3, the PRC2 product mark | PF (16 h to 24 h food withdrawal; "overnight" in Methods) | mouse liver, in vivo | A1 | C | Byun 2020, Nat Commun 11:807 | MEASURED, direction; no numeric fold change given in the text |
| 35.2 | Fasting increased hepatic JMJD3 (KDM6B) protein, the H3K27me3 eraser | PF | mouse liver, in vivo | A1 | C | Byun 2020 | MEASURED |
| 35.3 | Fasting increased JMJD3 occupancy and decreased H3K27me3 at Tfeb, Ulk1, Atgl and Fgf21 | PF | mouse liver, in vivo | A1 | C | Byun 2020 | MEASURED |
| 35.4 | Liver-specific JMJD3 knockdown decreased 846 genes; 564 of them showed decreased expression with increased H3K27me3, enriched for autophagy and lysosomal function | n/a, genetic | mouse liver | A1 | C | Byun 2020 | MEASURED |
| 35.5 | EZH2 expression unchanged by fasting | PF | mouse liver, in vivo | A1 | C | Byun 2020 | MEASURED |
| 35.6 | Writing this node as "fasting suppresses PRC2" requires the hop EZH2 writes H3K27me3, therefore less H3K27me3 means less PRC2 activity. That hop is contradicted by 35.5 within the same study | n/a | hop shown | M | C | Byun 2020, hops stated | MEASURED premise, inferred conclusion, conclusion rejected |
Conflict left unresolved
The mark falls under fasting (35.1, MEASURED) while the writer does not change (35.5, MEASURED). Both sides are from the same opened study. Whether the mark falls because demethylation outruns a constant writer, or because PRC2 activity falls without a change in EZH2 abundance, is not resolved by any source located here. Both readings stay on the record.
Open questions
- EED, SUZ12 and RBBP4 under any fasting modality in any species: unmeasured.
- H3K27me3 under fasting in any human tissue: unmeasured.
- Whether the H3K27me3 fall is liver-specific: no second tissue located.
← 34. Myc transcriptional network All 41 nodes 36. SWI/SNF chromatin remodeling →
dryfasting.com has been published since 2014. This evidence record was compiled on 22 November 2026. Every row names the fasting modality the study actually ran, the system it was measured in and the grade. Where no source carries a DOI or PMID the row says so rather than supplying one. Nothing is sold on this site.