Dry Fasting — The Record

Pathway atlas · node 30 of 41

Serine–glycine / one-carbon (folate) metabolism

What fasting does to this node, by modality, by system and by grade. Disconfirming evidence is reported first.

Founder's key components: PHGDH, PSAT1, SHMT1/2, MTHFD1/2 Founder's description: "Generates one-carbon units used in nucleotide synthesis, methylation reactions, and redox balance."

DRY status

DRY: none located. Query run: the single Russian plus English Filonov-tradition query listed under node 28. No DRY-modality homocysteine, folate, serine, glycine or methylation measurement was located. Dedicated one-carbon DRY queries and the nine non-English channels could not be run. Recorded NOT RUN. Note for this node specifically: RAMADAN restricts FLUID as well as food, so it is the closest modality in the brief to a dry fast, and this node is the one place in the block where RAMADAN carries the primary evidence. That proximity is NOT equivalence and no row below is a DRY row.

Disconfirming first: HOMOCYSTEINE MOVES IN OPPOSITE DIRECTIONS ACROSS RAMADAN STUDIES. UNRESOLVED.

Four studies were retrieved. Two measured a decrease, two measured an increase. All four are RAMADAN modality. This file names all four and resolves nothing.

SIDE A, DECREASE. A1. Benli Aksungar F, Eren A, Ure S, Teskin O, Ates G. "Effects of Intermittent Fasting on Serum Lipid Levels, Coagulation Status and Plasma Homocysteine Levels." Ann Nutr Metab 2005;49(2):77 to 82. DOI 10.1159/000084739. PMID 15802901. 24 healthy volunteers (12 female, 12 male), aged 21 to 35. Venous blood 1 week before Ramadan, day 21 of Ramadan, and 20 days after. Homocysteine LOWER during Ramadan, p < 0.05, described as still within reference ranges, returning to pre-fasting levels afterward. HDL significantly elevated during Ramadan (p < 0.001) and at 20 days after (p < 0.05). D-dimer significantly lower at the end of Ramadan versus before and after (p < 0.001). Total cholesterol, triglycerides and LDL: no significant change. Prothrombin time, aPTT and fibrinogen stayed within physiologic limits at all three time points. Numeric homocysteine values are behind the paywall and were NOT retrieved. Retrieval debt R30.1. Modality RAMADAN. Tier H3 (single-arm pre/post). Grade B+.

A2. Hammouda O, Chtourou H, Aloui A, Chahed H, Kallel C, Miled A, Chamari K, Chaouachi A, Souissi N. "Concomitant Effects of Ramadan Fasting and Time-Of-Day on Apolipoprotein AI, B, Lp-a and Homocysteine Responses during Aerobic Exercise in Tunisian Soccer Players." PLOS ONE 2013;8(11):e79873. DOI 10.1371/journal.pone.0079873. 15 male professional soccer players, age 17.3 +/- 0.3 years. Homocysteine micromol/L, mean +/- SD:

ConditionTimeRest3 min post-exercise
Before Ramadan07:0015.82 +/- 0.8716.64 +/- 0.73
Before Ramadan17:0016.82 +/- 0.7718.25 +/- 0.70
Second week of Ramadan07:0015.39 +/- 0.6715.87 +/- 0.69
Second week of Ramadan17:0015.57 +/- 0.6515.92 +/- 0.57
Fourth week of Ramadan07:0014.83 +/- 0.6414.99 +/- 0.58
Fourth week of Ramadan17:0014.40 +/- 0.6314.61 +/- 0.64

ANOVA as reported: Ramadan effect F = 39.63, p < 0.05. Time-of-day effect F = 39.21, p < 0.05. Blood-sample effect (rest versus post-exercise) F = 47.67, p < 0.05. Ramadan x time-of-day F = 68.61, p < 0.05. Ramadan x blood sample F = 12.27, p < 0.05. Time-of-day x blood sample F = 3.38, p > 0.05. Resting and post-exercise homocysteine fell versus baseline at the same time of day. The normal diurnal pattern reversed by the fourth week, with higher morning values (p < 0.001). The authors attribute the homocysteine and inflammatory changes partly to chronobiological factors, not only to calorie restriction, and state the design cannot separate diet, time-of-day and physical activity. Modality RAMADAN. Tier H3. Grade B+.

SIDE B, INCREASE. B1. Oueslati I, Kardi A, Boukhayatia F, Hammami B, Cheikh M, Ben Romdhane N, Feki M, Yazidi M, Chihaoui M. "Impact of Ramadan intermittent fasting on metabolic and inflammatory profiles in type 2 diabetic patients." Journal of Diabetes and Metabolic Disorders 2022;21(1):751 to 758. DOI 10.1007/s40200-022-01046-8. 55 adults with type 2 diabetes on oral hypoglycemic drugs, age 54.5 +/- 10.1 years, fasting 29.3 +/- 2.3 days. Time points: before Ramadan, immediately after, and two months after. Homocysteine 12.2 +/- 6.2 to 13.5 +/- 6.4 micromol/L, INCREASED, p = 0.001. Insulin 9.7 +/- 5.5 to 7.98 +/- 5.05 mIU/L, decreased, p = 0.043. Fructosamine 303.6 +/- 46 to 333.49 +/- 59.49 micromol/L, p < 10^-3, returning to baseline at two months. Fibrinogen 3.7 +/- 0.8 to 3.4 +/- 0.6 g/L, p = 0.003. hs-CRP 4.8 +/- 5.7 to 3.7 +/- 4.5 mg/L, p = 0.058, not significant. Total daily calorie intake fell 19%, p < 10^-3. Weight 79.8 +/- 12.9 to 78.4 +/- 13.3 kg, p = 0.003, and weight loss correlated with number of fasting days (r = 0.348, p = 0.009). Fasting blood glucose, HOMA-IR and lipids: no significant change. Modality RAMADAN. Tier H3. Grade B+.

B2. Lamti F, Messous S, Toumia M, Trabelsi I, Dhaoui R, Bel Haj Ali K, Hajji E, Benamor B, Sekma A, Jaballah R, Yaakoubi H, Zorgati A, Beltaief K, Ben Soltane H, Mezgar Z, Khrouf M, Ben Abdallah I, Sghaier A, Jerbi N, Razgallah R, Hallara I, Bouida W, Habib Grissa M, Saad J, Boubaker H, Boukef R, Neffati F, Sakly N, Dridi Z, Amine Msolli M, Sriha A, Khochtali I, Nouira S. "Effects of fasting on endothelial function in type 2 diabetes mellitus patients: cohort study." Front Cardiovasc Med 2026;12:1646786. DOI 10.3389/fcvm.2025.1646786. 167 patients with type 2 diabetes, measured before and during Ramadan. Homocysteine INCREASED during Ramadan by 1.2 (95% CI 0.7 to 1.8), p < 0.001. UNIT AS PRINTED: the article's results text labels the unit mmol/L, which is inconsistent with its own figure; the physiologically sensible unit is micromol/L. Reported exactly as served and flagged, not corrected. Change in homocysteine associated with a decrease in reactive hyperemia index: OR 2.7 (95% CI 0.98 to 7.69), p = 0.054, NOT statistically significant. This paper cites NO prior study on homocysteine during Ramadan and does not flag the conflict. Modality RAMADAN. Tier H2 (prospective cohort, node measured). Grade B+.

THE CONFLICT, STATED AND LEFT OPEN. Two studies measured homocysteine DOWN. Two measured it UP. All four ran RAMADAN. No source opened reconciles them. One observation, offered as an observation and not as a resolution: the two decreases were measured in healthy young adults (n = 24) and young male athletes (n = 15); the two increases were measured in type 2 diabetes cohorts (n = 55 and n = 167). Whether population, medication, hydration, Ramadan diet composition, assay method, season or daylight hours accounts for the split was NOT established by anything I opened. No study opened tested the two populations head to head. No side is picked. On any customer-facing surface this node's homocysteine direction must be stated as contested, with both sides named, or not stated at all.

Graded rows

30-R1. Glycine and methionine RISE during a 10-day fast, with a surge at about day 5. Glutamine, glutamate and alanine fall. Grade B+, tier H3, modality PF. Steinhauser ML, Olenchock BA, O'Keefe J, Lun M, Pierce KA, Lee H, Pantano L, Klibanski A, Shulman GI, Clish CB, Fazeli PK. JCI Insight 2018;3(16):e121434. DOI 10.1172/jci.insight.121434. Design as in D29.1: 0-calorie 10-day inpatient fast, water ad libitum, multivitamin, 20 mEq KCl, 200 mg allopurinol; 12 enrolled, 8 completed; 7.0% to 11.7% body weight lost. Measured: glycine INCREASED; methionine INCREASED; glycine, asparagine, methionine and lysine surged at about day 5. Glutamine and glutamate decreased. Alanine decreased. Also measured in the same paper: leptin fell about 50% within the first day. Triacylglycerol C60:12 rose roughly 30-fold; fully saturated TAG C44:0 declined roughly 2-fold (day 10 versus baseline). SERINE was NOT reported in the content served, and no magnitudes or per-day values for glycine or methionine were given. The direction is solid; the size is unread. Retrieval debt R30.2. This is the only located human fasting measurement of this node's own metabolites, and it is single-arm, n = 8 completers.

30-R2. About 44% of 377 quantified plasma entities changed significantly when fasting was extended from 12 h to 36 h. Grade B, tier H3, modality PF. Rubio-Aliaga I, de Roos B, Duthie SJ, et al. (22 authors; Daniel H corresponding). "Metabolomics of prolonged fasting in humans reveals new catabolic markers." Metabolomics 2011;7(3):375 to 387. DOI 10.1007/s11306-010-0255-2. Design: four overnight fasting samples per volunteer over four weeks, plus one sample after a prolonged 36 h fast. Measured: of 377 quantified plasma entities, about 44% changed significantly between 12 h and 36 h. Known markers reproduced: free fatty acids, glycerol, ketone bodies. Newly identified: alpha-aminobutyrate and other amino and keto acids. Subject number is NOT stated in the abstract (described only as "a small cohort"), and no serine, glycine or one-carbon direction was retrievable. Retrieval debt R30.3. The 44% figure is included because it is a denominator: it establishes that a 36 h fast moves a large share of the measurable plasma metabolome, which is the context any single-metabolite claim about this node sits in. Note on alpha-aminobutyrate: it is a transsulfuration-adjacent metabolite, which is suggestive for this node, but the paper's own framing is catabolic markers generally. No one-carbon claim is made from it here.

30-R3. No human fasting measurement of this node's named enzymes was located. Tier M if written at all. PHGDH, PSAT1, SHMT1, SHMT2, MTHFD1 and MTHFD2 were not measured under any fasting modality in any source located. Any statement that fasting modulates the serine-glycine synthesis or mitochondrial folate enzymes in humans is M: inferred from the shared upstream node, not measured. It must not be written as a fasting effect.

Open questions, node 30


← 29. Glutaminolysis All 41 nodes 31. De novo lipogenesis →


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.