Founder's role text (READ): "Removes small, non-helix-distorting base lesions arising from oxidation, alkylation, or deamination." Founder's key components (READ): OGG1, MUTYH, APEX1, POLB, XRCC1, PARP1. In the recommender: No.
This is the only node in the block with a direct human fasting measurement of a BER-relevant marker. The human result is null.
DRY
DRY: none located. Queries: the three dry-fast queries above, plus Ramadan fasting 8-OHdG oxidative DNA damage urinary and fasting 8-oxo-7,8-dihydro-2'-deoxyguanosine human lymphocyte OGG1 base excision repair. No absolute dry fast study measures 8-OHdG, 8-oxodG, OGG1, or any BER component. Papagiannopoulos-Vatopaidinos et al. 2020 measured plasma total antioxidant capacity, uric acid and vitamin C, and no DNA oxidation marker.
THE URINARY NORMALIZATION FLAG, STATED HARD
The founder asked specifically whether the human 8-OHdG measurement was urinary or leukocyte, and whether it was corrected for urine concentration, because an uncorrected urinary marker in a fluid-restricted fast is confounded by dehydration. The finding is this:
- The one human fasting study that measured oxidative DNA damage used a urinary marker, 8-OHdG by ELISA, not leukocyte DNA.
- Its result was null. No significant difference after fasting.
- Whether that specific marker was creatinine-corrected could not be verified. The companion marker in the same paper, malondialdehyde, is explicitly reported per gram of creatinine, so creatinine correction was in use in the laboratory. The accessible record does not state the normalization for 8-OHdG or for 8-isoPGF. This is an open retrieval debt and it is named as one.
- The protective logic runs the right way regardless. Dehydration concentrates urine and inflates an uncorrected urinary marker, which biases toward a false positive. This study found no increase. A dehydration confound cannot manufacture a null. There is no positive here for the confound to explain away.
- The protocol was a South Korean fasting program, not a fluid-restricted fast, and water status is not stated on the accessible record. So the dehydration confound is not even established as present.
- The operative warning for the atlas: any future row that cites an uncorrected urinary 8-OHdG rise during RAMADAN or DRY, where fluid is restricted by design, is confounded and must not be written as a BER or oxidative-damage effect. No such row exists in this file because no such positive result was located.
Disconfirming evidence, first
- 21.D1. The human fasting result for oxidative DNA damage is null. Lee KH, Bartsch H, Nair J, Yoo DH, Hong YC, Cho SH, Kang D. "Effect of short-term fasting on urinary excretion of primary lipid peroxidation products and on markers of oxidative DNA damage in healthy women." Carcinogenesis 2006;27(7):1398-1403. PMID 16401636. DOI 10.1093/carcin/bgi337. Published online 9 January 2006. Design MEASURED from the opened record: 52 healthy women, mean age 28, range 15 to 48, urine before and after a short-term fasting program in South Korea. Fasting averaged 7.2 days (range 3 to 11). MEASURED: "no significant difference in 8-OHdG or edA" after fasting. Urinary malondialdehyde significantly lower (n = 51, P < 0.01; 0.18 +/- 1.10 vs 0.37 +/- 1.11 mg/g creatinine). Urinary 8-isoprostaglandin F2-alpha significantly reduced (n = 47, P < 0.01). Defect MEASURED on the printed values: the standard deviations as printed (1.10 and 1.11) are larger than the means (0.18 and 0.37) by roughly sixfold and threefold. For a non-negative concentration that is arithmetically implausible as a simple standard deviation. Either the dispersion statistic is not an SD, or the values are transformed, or the printing is wrong. Report unresolved. Do not quote those two numbers on a customer surface without opening the full text. Tier H3 (human single-arm pre/post). Modality: PF (prolonged fasting, water permitted) is the closest label, but water status is not stated on the accessible record and the "program" may have included a juice or broth allowance. Modality labeled provisionally and flagged. Grade C for the node, because it reads out a lesion marker rather than BER activity.
- 21.D2. CR does not broadly induce BER, and in two tissues it goes the other way. Stuart JA, Karahalil B, Hogue BA, Souza-Pinto NC, Bohr VA. "Mitochondrial and nuclear DNA base excision repair are affected differently by caloric restriction." FASEB J 2004;18:595-597. DOI 10.1096/fj.03-0890fje. Design MEASURED: mouse, CR at 60% of caloric intake vs pair-fed controls at 100%, 14 months, starting at 8 weeks. Mitochondria from liver, brain, kidney, plus heart for UNG. Nuclear extracts from liver and kidney. Strain not stated on the accessible record. MEASURED, mitochondrial BER synthesis: liver about +18%, not significant; brain about -30%, significant; kidney about -30%, marginal, P < 0.065. AP endonuclease lower in all tissues, significant only in brain. Polymerase gamma 20 to 30% lower in brain and kidney. OGG1 about +20% in liver, no significance value given. NTH1 (5-hydroxycytosine incision) about +40% in kidney. UNG: small non-significant increases under 10% in heart, brain and liver, kidney about +40%, P < 0.05. MEASURED, nuclear: BER activity +26% in liver, not significant; +42% in kidney, P < 0.05. Glycosylase and AP endonuclease activities: no differences between CR and pair-fed. Authors' conclusion as recorded: mitochondrial BER was not broadly induced by CR; lower mitochondrial oxidative damage is attributed mainly to a roughly 45 to 50% reduction in mitochondrial ROS production rather than to enhanced repair. Tier A1. Modality CR. Grade C. This is the direct contradiction of the "CR raises BER" story and it must sit ahead of the supporting row.
- 21.D3. The standard review records BER effects as tissue-dependent and compartment-dependent, not general. Heydari 2007 (NAR 35(22):7485-7496), READ: in C57BL/6 mice "the effects varied by tissue and by mitochondrial versus nuclear DNA. Mitochondrial BER decreased in brain and kidney but increased in liver."
- 21.D4. Two located nulls on CR and DNA repair that could not be opened. Named as retrieval debts, not as results:
- "Evidence that DNA repair may not be modified by age or chronic caloric restriction." PMID 7680761. Title READ from the PubMed record URL and the search listing. Authors, journal, volume and pages not retrieved.
- Lipman JM, Turturro A, Hart RW. "The influence of dietary restriction on DNA repair in rodents: A preliminary study." Mech Ageing Dev 1989;48:135-143. DOI 10.1016/0047-6374(89)90045-6. Full citation READ from the reference list of van Galen et al. 2026. Abstract not opened.
- Heydari 2007 also names a null by "Shaddock et al." on unscheduled DNA synthesis under CR. Year, journal and full author list not retrieved.
Graded rows, supporting
| Row | Modality | Species / system | Tier | Grade | What was actually measured | Direction |
|---|---|---|---|---|---|---|
| 21.1 | CR | Rat, Fischer 344 | A1 | C | BER activity, DNA polymerase beta | Up, reverses age decline |
| 21.2 | CR | Mouse, multi-tissue | A1 | C | BER in brain, liver, kidney, spleen, testes | Up in nucleus, down in kidney and brain mitochondria |
| 21.3 | PF | Human, n=26 randomized, 6-day zero-calorie | H1 | B+ for oxidative stress, not for BER | MDA and total antioxidant capacity only. No DNA marker at all. | MDA up during fast, back to baseline on refeeding |
| 21.4 | PF (modified) | Human, n=109, 10-day Buchinger | H3 | C | Redox and oxidative damage panel. No DNA repair gene, no 8-OHdG reported. | Antioxidant capacity up |
| 21.5 | C1 lead, unopened | Cell culture, nutrient withdrawal | C1 | debt | PARP1 requirement for starvation-induced autophagy | Not verified |
21.1 source: Cabelof DC, Yanamadala S, Raffoul JJ, Guo Z, Soofi A, Heydari AR. "Caloric restriction promotes genomic stability by induction of base excision repair and reversal of its age-related decline." DNA Repair 2003;2:295-307. DOI 10.1016/S1568-7864(02)00219-7. Full citation READ from the reference list of van Galen et al. 2026 (ref 52). Pathway-level result READ from two independent reviews: Heydari 2007 ("CR is reported to reverse the age-related decline in BER in male Fischer 344 rats, mainly by upregulating DNA polymerase beta. CR also increased BER and resistance to two carcinogens in young animals") and van Galen 2026. Primary abstract not opened: the exact percentages, n, CR percent and duration are a retrieval debt.
21.2 source: van Galen I, Hoeijmakers JHJ, Vermeij WP. Nutrients 2026;18(2):246. DOI 10.3390/nu18020246. READ verbatim from the opened review: "DR protected against chemical mutagenesis and improved BER in brain, liver, kidney, spleen, and testes, possibly by preventing age-related decline in DNA polymerase beta. BER was downregulated in mitochondria of kidney and brain." Primary refs given there as Cabelof 2003 and Stuart 2004, both cited in full above.
21.3 source, opened: Brazaitis M, Zidoniene K, Eimantas N, Solianik R. Antioxidants 2025;14(3):269. DOI 10.3390/antiox14030269. Published 26 February 2025. ClinicalTrials.gov NCT05545943. Design MEASURED: single-center, parallel-group, single-blind randomized controlled trial, Lithuanian Sports University, February 2021 to September 2022. 26 healthy men, mean age 29.4 +/- 7.0, BMI 24.9 +/- 2.5. Fasting n = 14, zero-calorie diet with water ad libitum, 6 days. Control n = 12, previous eating habits. Measurements at day 0, 2, 4, 6, and day 7 of regular diet. MEASURED: only two oxidative markers, plasma MDA and total antioxidant capacity (TAC). MDA time effect P = 0.007 (partial eta squared 0.18), time x intervention P < 0.001 (0.29); increased after fasting days 4 and 6 (P <= 0.044); higher on day 6 in fasting than control (P = 0.007); fell after refeeding vs day 6 (P = 0.002) and returned to pre-fasting values (P > 0.05). TAC time effect P = 0.003 (0.18), time x intervention P = 0.011 (0.14); increased after fasting days 4 and 6 (P <= 0.002), no difference between trials; rose after refeeding vs day 6 (P = 0.025) and returned to pre-fasting values. TAC/MDA ratio: no significant time effect (Friedman P > 0.05), no trial difference (Mann-Whitney P > 0.05). Absolute means and SDs appear only in Figure 5 and are not readable from the text: retrieval debt. MEASURED negative: 8-OHdG, 8-oxodG, comet assay and every other DNA oxidation or DNA repair marker were not measured. No urinary marker was used, so there is no normalization question. The methods list every measurement and none of these appear. This is the cleanest human prolonged-fasting RCT in this space and it did not measure this node. That absence is itself the finding, and it is why node 21's human tier stops at H3.
21.4 source: Grundler F, Mesnage R, Goutzourelas N, Tekos F, Makri S, Brack M, Douretas D, Wilhelmi de Toledo F. "Interplay between oxidative damage, the redox status, and metabolic biomarkers during long-term fasting." Food and Chemical Toxicology 2020. DOI 10.1016/j.fct.2020.111701. Published 25 August 2020. Design READ from the German-language channel (Arztegesellschaft Heilfasten summary, opened): 109 participants, 10-day fast at Klinik Buchinger Wilhelmi, Uberlingen. 250 ml fruit juice daily, described as providing few exogenous antioxidants. Water allowance not stated. Study type not named on that page. MEASURED on that page: lipid peroxidation by TBARS; high baseline TBARS limited LDL reduction during long-term fasting. Total blood antioxidant capacity increased. Endogenous antioxidants rose significantly. Protein oxidation markers and antioxidant enzymes: no effect. Metabolic markers normalized. WHO-5 wellbeing improved. No DNA repair gene and no DNA oxidation marker is reported as measured. The source read is a summary page, not the primary. Volume and page numbers not retrieved. Full retrieval debt on the primary. MODALITY FLAG: 250 kcal-scale juice daily puts this between PF and FMD. It is not a pure water fast. Labeled PF (modified) and flagged. IDENTIFIER FLAG: the author name printed on that page as "Douretas" is unusual for this author group. The name is recorded here exactly as printed and is not corrected, because correcting it would be inventing an identifier. Verify against the primary before any publication use.
21.5 source, located but not opened: "ROS-induced DNA damage and PARP-1 are required for optimal induction of starvation-induced autophagy." PMID 22525338. PMCID PMC3391023. Title and identifiers READ from search listings. Content not opened. PARP1 is on the founder's BER component list, which is why this lead is recorded rather than dropped. It is a cell-culture lead (C1 at best) and no finding is stated from its title. Retrieval debt.
Open questions
1. The 8-OHdG normalization method in Lee 2006 is unresolved and is the single highest-value retrieval debt in this block. 2. Direction for BER under CR is compartment-dependent and unresolved: nuclear up, mitochondrial down in brain and kidney, liver equivocal. No reviewer has reconciled Cabelof 2003 with Stuart 2004. Report both. 3. No fasting modality has measured OGG1, MUTYH, APEX1, POLB or XRCC1 in humans. 4. Also located for context and deliberately not made into a row, because the intervention is a single-nutrient deficiency and not a fasting modality: "Folate Deficiency Alters Hepatic and Colon MGMT and OGG-1 DNA Repair Protein Expression in Rats but Has No Effect on Genome-Wide DNA Methylation," DOI 10.1158/1940-6207.CAPR-09-0231. Named here so nobody later files it as fasting evidence.
← 20. Homologous recombination (HR) All 41 nodes 22. Nucleotide excision repair (NER) →
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.