Dry Fasting — The Record

Pathway atlas · node 39 of 41

Integrin / FAK adhesion signaling

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

Founder's role line (READ): "Transmits cell-extracellular-matrix attachment signals controlling adhesion, migration, and survival." Key components (READ): ITGA/ITGB integrins, PTK2 (FAK), SRC

DRY status

DRY: none located. Queries run:

MEASURED: neither opened human dry-fasting study reports ITGA or ITGB integrins, PTK2/FAK, or SRC. Mindikoglu 2022 reports FLNC down and FLNA down (filamins, which bind integrin tails) and MYL9 up. Those are cytoskeletal and contractile proteins, not the founder's listed components, and no integrin, FAK or SRC was measured. Stated so a filamin is not promoted into an integrin.

Disconfirming and null first

No fasting intervention in any species was located in which FAK (PTK2), an integrin or SRC was measured. That is the finding for this node.

Graded rows

#ClaimModalitySpecies / systemTierGradeSourceTag
39.1Every-other-day fasting decreased ECM-related protein sets in visceral WAT, including collagen IV and collagen VIIF-ADFmouse, visceral WAT, in vivoA1CHarney 2021, Cell Reports 34:108804MEASURED
39.2The same ECM decrease did not occur in subcutaneous WATIF-ADFmouse, subcutaneous WATA1CHarney 2021MEASURED
39.3Picrosirius red collagen staining showed no visible change in either white depotIF-ADFmouseA1CHarney 2021MEASURED, and it contradicts 39.1
39.4The ligand side of this node (matrix collagen) moves under fasting in one depot; the receptor and kinase side (ITGA/ITGB, PTK2, SRC) is unmeasured under any fasting modality in any species. Hop: less collagen IV and VI means less integrin engagement means less FAK signaling. That hop was not tested by anything locatedIF-ADFhops shownMCHarney 2021, hops statedMEASURED premise, inference shown and NOT accepted
39.5FAK Tyr397 phosphorylation rose after carbohydrate plus exercise and not after placebo plus exercise in 10 humans. This is a feeding result, not a fasting resultnot a fasting interventionhuman crossoverH2 by design, but not this node under fastingn/a for fastingWilburn 2020, Nutrients 12(10):3145MEASURED

Conflict left unresolved

Within Harney 2021: proteomic ECM loss in visceral fat against no change on collagen histology, in the same animals. Reported unresolved. There is no second fasting study to set against it.

Open questions


← 38. Epithelial-mesenchymal transition (EMT) All 41 nodes 40. Angiogenesis (VEGF axis) →


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.