What it is
Glutathione is the endogenous tripeptide γ-L-glutamyl-L-cysteinyl-glycine (C₁₀H₁₇N₃O₆S, 307.32 Da, CAS 70-18-8, PubChem CID 124886). It is synthesized in the cytoplasm of nearly every mammalian cell in two ATP-dependent steps: γ-glutamylcysteine ligase (GCL, formerly γ-GCS) condenses glutamate and cysteine (the rate-limiting step, feedback-inhibited by GSH), and glutathione synthetase adds glycine. The unusual γ-peptide bond between glutamate's side-chain γ-carboxyl and cysteine's α-amino group resists most peptidases — only γ-glutamyltransferase (GGT) can cleave it — which is why exogenous GSH is efficiently hydrolyzed on the gut brush border and on the vascular endothelium. Intracellular concentrations reach 0.5–10 mM (>90% reduced form in healthy tissue), establishing GSH as the single largest non-protein thiol pool in the body. It is not a 'peptide therapeutic' — it is a ubiquitous metabolite, which makes the evidence framing fundamentally different from exogenous peptides like BPC-157 or semaglutide.
How it works
- 01
Glutathione peroxidase (GPx) — primary peroxide defense
Glutathione peroxidases are selenoprotein enzymes that reduce hydrogen peroxide and lipid hydroperoxides using GSH as the two-electron donor: 2 GSH + H₂O₂ → GSSG + 2 H₂O. GPx1 (cytosolic) and GPx4 (membrane-associated, lipid-peroxide-specific) are the dominant forms; GPx4 knockout is embryonic lethal and its activity is the gatekeeper of ferroptosis. Mitochondria lack catalase, so mitochondrial GSH + GPx is the sole enzymatic defense against matrix-generated H₂O₂ (Marí 2020, Antioxidants; Lu 2013, Mol Aspects Med). Resulting GSSG is reduced back to GSH by glutathione reductase using NADPH — the redox cycle that couples GSH metabolism to the pentose phosphate pathway.
- 02
Glutathione S-transferase (GST) — Phase II detoxification
The cytosolic GST superfamily (α, μ, π, θ, ζ classes in humans) catalyzes nucleophilic attack of the GSH thiolate on electrophilic xenobiotics and reactive endogenous metabolites (α,β-unsaturated aldehydes, quinones, epoxides), forming GSH conjugates that are exported via MRP1/MRP2 transporters and processed through the mercapturic-acid pathway (GGT → dipeptidase → N-acetyltransferase) for renal elimination (Hayes 2005, Annu Rev Pharmacol Toxicol). This is the biochemist's 'detoxification' — hepatocyte-localized conjugation, not the infusion-clinic marketing term.
- 03
Cysteine reservoir and the rate-limiting role of cysteine
De novo GSH synthesis is rate-limited by intracellular cysteine, not glycine or glutamate. Cysteine is the scarce, autoxidation-prone input; this is why NAC (N-acetylcysteine) rescues hepatic GSH in acetaminophen overdose and why GlyNAC (glycine + NAC) supplementation (Sekhar 2023, Nutrients; Kumar 2023, J Gerontol) reliably raises erythrocyte GSH in deficient elderly adults. Oral GSH is less efficient than oral cysteine precursors because most ingested GSH is hydrolyzed at the brush border before reaching the portal circulation (Witschi 1992, Eur J Clin Pharmacol).
- 04
Protein S-glutathionylation — redox signaling
Reversible formation of mixed disulfides between GSH and protein cysteine residues (Protein-SSG) is now recognized as a major redox post-translational modification, regulating metabolic enzymes (GAPDH, pyruvate kinase), signaling proteins (PTP1B, NF-κB), and apoptotic caspases (Mailloux 2024, Redox Biol). Glutaredoxins (Grx1/Grx2) and sulfiredoxin catalyze deglutathionylation. This positions GSH not just as an antioxidant buffer but as a signaling cofactor — and complicates the 'more GSH is better' intuition that drives consumer marketing.
- 05
Pharmacokinetics of exogenous GSH — why route matters
Oral GSH: Witschi (1992) administered 3 g oral GSH to 7 healthy volunteers and observed no significant rise in plasma GSH, concluding oral bioavailability is effectively zero; more recent analogue and liposomal work (Richie 2015; the 2025 N-methylated analogue work in MDPI Pharmaceutics) has tried to circumvent brush-border GGT hydrolysis with mixed results. IV GSH: Aebi (1991, Eur J Clin Pharmacol) administered 2 g IV to 5 volunteers and showed plasma GSH cleared with a half-life of ~10 minutes; total GSH disappears from the circulation within ~90 minutes. Intranasal: Mischley (2013) showed modest brain GSH increase on MRS, the basis of the Mischley 2015 PD trial. The short plasma half-life is the mechanistic argument against IV 'detox drips': a bolus that clears in 10 minutes does not meaningfully raise hepatocyte intracellular GSH.
- 06
Proposed skin-whitening mechanism (and why it does not justify IV use)
In-vitro work (Villarama 2005; del Rosario-Blasco 2008) shows that GSH at millimolar concentrations in melanocyte culture shifts tyrosinase product distribution away from eumelanin toward lighter pheomelanin, and that GSH inhibits tyrosinase activity. Translating this to systemic IV dosing requires that a plasma GSH spike reaches melanocyte intracellular pools at the required concentration for long enough to matter — a chain of assumptions not supported by the pharmacokinetic data above. The controlled oral trials (Arjinpathana 2012; Weschawalit 2017; Handog 2016) show small and inconsistent melanin-index changes that do not translate reliably to patient-reported outcomes.