Skip to content

Resveratrol as a "treatment": what the direct structural evidence actually shows

Investigation report — 2026-07-18

TL;DR

Resveratrol is the textbook case of a compound whose popular therapeutic narrative (a SIRT1-activating, caloric-restriction-mimetic geroprotector) is the part with the weakest direct evidence, while its best-validated molecular fact — a 1.5 Å co-crystal with quinone reductase 2 (NQO2/QR2), Kd ≈ 34 nM (PDB 1SG0) — is rarely what it is marketed for. Starting from the structural layer, as requested, resveratrol looks less like a targeted drug and more like a promiscuous, low-affinity polypharmacology probe with a bioavailability problem that has produced a largely null/mixed human clinical record.

Recommendation for dismech: do not attach resveratrol to the deregulated_nutrient_sensing / cellular_senescence geroprotector modules as a "SIRT1 activator" — that would import exactly the contested downstream theory those modules deliberately exclude. If resveratrol is curated at all, the only structurally-defensible home is as a transthyretin (TTR) tetramer stabilizer, an instance of the existing amyloidogenesis module's tafamidis drug-target pattern (see §5). The NQO2 and TyrRS axes are the two cleanest direct targets but have no current dismech disorder home.

1. The structural evidence layer (start here)

Resveratrol (C₁₄H₁₂O₃; a stilbenoid polyphenol, three hydroxyls) has an unusually large set of solved protein co-crystal structures. The Frontiers 2018 review "Polypharmacology or Promiscuity?" (Saqib et al., PMID:30405417 / PMC6207623) catalogs them. The key point is that these targets share no common fold, sequence motif, or binding-site chemistry — resveratrol simply exploits its flat, H-bond-capable stilbene scaffold opportunistically.

Target PDB Direction Notes
Quinone reductase 2 (NQO2/QR2) 1SG0 inhibitor Highest measured affinity, Kd 34 ± 15 nM; sits parallel to FAD isoalloxazine, all 3 –OH H-bonded. Buryanovskyy 2004, PMID:15350128
Tyrosyl-tRNA synthetase (TyrRS/YARS1) 4Q93 active-site mimic Tyr-mimetic; redirects TyrRS to nuclear PARP1 activation. Sajish & Schimmel 2015, PMID:25533949
Leukotriene A4 hydrolase 3FTS inhibitor
Phospholipase A2 4QER inhibitor
F1-ATPase 2JIZ inhibitor
Sulfotransferase 1B1 3CKL inhibitor (also a resveratrol metabolizer)
PPAR-γ 4JAZ inhibitor/modulator
Transthyretin (TTR) 5CR1 / 8W42 stabilizer Binds T4 pocket, stabilizes tetramer (tafamidis-like)
Troponin C 2L98 modulator
Myosin-2 motor domain 3MNQ inhibitor
Methionine adenosyltransferase 2B 2YDX inhibitor
Sirtuin-5 4HDA weak activator
Sirtuin-1 5BTR allosteric, substrate-dependent Two NTD-bound resveratrol molecules tighten SIRT1–peptide binding. Cao 2015, PMID:26109052
Estrogen receptor α 4PP6 / 4PPP modulator phytoestrogen activity

Two direct targets outrank SIRT1 on the structural/affinity evidence:

  • NQO2/QR2 (1SG0) — the flagship. High-affinity (nM), clean competitive active-site occupancy. Functional knockdown of QR2 phenocopies resveratrol (increased antioxidant/detoxification enzyme expression, reduced proliferation), supporting QR2 inhibition as a bona fide mechanism (Buryanovskyy 2004).
  • TyrRS (4Q93) — Sajish & Schimmel showed resveratrol occupies the TyrRS active site as a tyrosine mimic and drives an NAD⁺-dependent auto-poly-ADP- ribosylation of PARP1, a stress-signaling axis activatable at ~1000× lower concentrations than the SIRT1 assays — i.e. within a plausible physiological range (Nature 2015, PMID:25533949).

2. The SIRT1 story is the contested one

The marketed mechanism — resveratrol as a direct SIRT1 activator and caloric-restriction mimetic — is the weakest link:

  • The original activation was measured with the Fluor-de-Lys fluorophore-tagged peptide. Resveratrol (and the "STAC" compounds SRT1720/2183/1460) do not activate SIRT1 against native, unlabeled substrates; the effect tracks the fluorophore, not the enzyme (Beher 2009, PMID:19843076; Pacholec 2010 — NMR/SPR/ITC showed compound–fluorophore, not compound–SIRT1, binding).
  • The rehabilitated view (Hubbard 2013; Cao 2015, PMID:26109052, PDB 5BTR) is allosteric and substrate-sequence-selective: two N-terminal-domain-bound resveratrol molecules stabilize the SIRT1–peptide complex for substrates that happen to carry bulky hydrophobic (fluorophore-like) residues at the +1 position. This is real but is not the clean, general "turn up sirtuin activity" story — and it is exactly why kb/modules/deregulated_nutrient_sensing.yaml lists rapamycin and metformin as its geroprotector treatments and pointedly leaves resveratrol out, mentioning it only inside a review-snippet.

3. Bioavailability and clinical translation: mostly null

  • Extensive first-pass glucuronidation and sulfation → very low, transient plasma parent-drug levels; the in-vitro efficacious 3–30 µM range is essentially never reached in vivo by oral dosing.
  • 2024 systematic reviews (e.g. IJMS 25(2):747) and bioavailability meta-analyses report heterogeneous, mostly modest/mixed effects on intermediate biomarkers (vascular function, inflammation, insulin sensitivity) and no replicated longevity or hard-outcome benefit in humans.
  • dismech has already hit this in practice: PR #3924 (Multiple Epiphyseal Dysplasia) explicitly excluded the resveratrol pseudoachondroplasia trial (NCT03866200) because it was terminated with no efficacy result.

4. So is resveratrol a "treatment"?

At the mechanism-of-disease altitude dismech curates: not for any specific disorder on current evidence. It is a promiscuous polyphenol whose:

  • molecular structure evidence is strong (many co-crystals; NQO2 genuinely high-affinity),
  • marketed mechanism (SIRT1 CR-mimetic) is contested/assay-dependent, and
  • clinical-outcome evidence is weak (bioavailability, mixed/failed trials).

The correct dismech posture is to treat it as a cautionary structural probe, not to bolt it onto an aging module where it would smuggle in the very theory that module was written to avoid.

5. The one clean dismech fit: TTR stabilization

If resveratrol should appear anywhere in the KB with defensible direct structural evidence, it is as a transthyretin tetramer stabilizer (PDB 5CR1 / 8W42): resveratrol binds the T4 thyroxine pocket and kinetically stabilizes the TTR tetramer against the rate-limiting dissociation step of amyloidogenesis — the identical mechanism to tafamidis, which the amyloidogenesis module already encodes as a target_mechanisms INHIBITS edge on the amyloid-precursor node. This is the only place resveratrol slots into an existing dismech drug-target design pattern backed by a co-crystal rather than by a contested phenotypic assay. It would still warrant a caveat that TTR stabilization by resveratrol is weaker than tafamidis and not a clinically established ATTR therapy.

  1. Do not curate resveratrol into deregulated_nutrient_sensing or cellular_senescence as a SIRT1 activator / CR mimetic.
  2. If a concrete KB action is wanted, add resveratrol as a secondary, caveated TTR-stabilizer exemplar in the amyloidogenesis module (structural evidence: PDB 5CR1/8W42), explicitly ranked below tafamidis.
  3. Optionally record NQO2 (1SG0) and TyrRS/PARP1 (4Q93) as the two structurally-validated direct targets in a note, since both outrank SIRT1 on the evidence but have no current disorder home.

Key references

  • Buryanovskyy L, et al. Crystal Structure of Quinone Reductase 2 in Complex with Resveratrol. Biochemistry 2004;43(36):11417–26. PMID:15350128 (PDB 1SG0)
  • Sajish M, Schimmel P. A human tRNA synthetase is a potent PARP1-activating effector target for resveratrol. Nature 2015;519:370–3. PMID:25533949 (PDB 4Q93)
  • Cao D, et al. Structural basis for allosteric, substrate-dependent stimulation of SIRT1 activity by resveratrol. Genes Dev 2015;29:1316–25. PMID:26109052 (PDB 5BTR)
  • Beher D, et al. Resveratrol is not a direct activator of SIRT1 enzyme activity. Chem Biol Drug Des 2009;74:619–24. PMID:19843076
  • Saqib U, et al. Polypharmacology or Promiscuity? Structural Interactions of Resveratrol With Its Bandwagon of Targets. Front Pharmacol 2018;9:1201. PMID:30405417 / PMC6207623