SULT Sulfonation (Phase 2)

What sulfonation is, and why its capacity limit is the whole story human data

Sulfotransferases transfer a sulfonate group from the cofactor PAPS onto a hydroxyl or an amine, producing a sulfate conjugate that is usually far more water-soluble and more readily excreted than the parent. The chemistry is unremarkable; the kinetics are what make this pathway behave differently from every other one on this shelf. Sulfation is a high-affinity, low-capacity route and glucuronidation is a low-affinity, high-capacity route, which means that at low substrate concentrations sulfation dominates and at higher concentrations it saturates and glucuronidation takes over. Two things limit capacity: the amount of enzyme, and the supply of PAPS, which depends on available inorganic sulfate and on the cysteine and methionine pool that feeds it. A large substrate load can therefore exhaust the cofactor rather than merely occupy the enzyme. Sulfation is also not uniformly a detoxification: a sulfate ester can be a good leaving group, and for a handful of substrates the sulfonate conjugate is the reactive species rather than the inert one. Paracetamol is the familiar demonstration of the capacity limit — at therapeutic doses a substantial fraction is sulfated and most of the rest glucuronidated, and as the dose rises sulfation saturates first, which shifts proportionally more of the dose toward glucuronidation and toward the CYP2E1 oxidation that generates NAPQI.

Sources: Coughtrie MWH 2016 · Riches Z 2009 · Koster H 1981* · Slattery JT 1996

Competitive sulfation: why a large phenolic load changes the fate of something else contested in vitro

This is the phenomenon that makes SULT worth a page of its own in a library about plants. Because the pathway has low capacity and a depletable cofactor, two substrates that would each be comfortably sulfated alone can interfere when presented together: the more abundant or higher-affinity one occupies the enzyme and consumes PAPS, and the other is shunted to a different route. The classical demonstration is dose-dependent: as the load of a phenolic substrate rises, the proportion of it that is sulfated falls and the proportion glucuronidated rises, and the same shift can be produced in a co-administered substrate. The best-known deliberate exploitation is the old use of a competing sulfation substrate to divert paracetamol metabolism. For this library the relevance is direct and mostly unmeasured: a strong tea, a concentrated polyphenol extract, a high-dose curcumin preparation, a green-tea catechin supplement or a large quercetin or resveratrol dose all present a substantial phenolic load, and a botanical whose active constituents are themselves phenols is competing for the same limited pathway as anything else phenolic in the gut and liver at that moment. The mechanistic prediction is that heavy phenolic intake shifts a co-administered phenol away from sulfation and toward glucuronidation and oxidation. What is genuinely lacking is human quantification for almost any specific pair, and the sulfate-supply side of it is a real constraint that a table of enzyme inhibitors does not capture at all.

• Saturation is the mechanism, not just inhibition: the cofactor PAPS can be depleted as well as the enzyme occupied.
• The shift is usually from sulfation toward glucuronidation, and toward oxidation if glucuronidation is also loaded.
• A phenol-rich botanical is both a substrate and a competitor at this pathway.
• Dietary sulfate and sulfur-amino-acid supply are upstream determinants of capacity, which is a nutritional variable inside a pharmacokinetic pathway.
• Intestinal SULT1A3 is a first-pass barrier for phenolic amines, so this pathway also determines whether an orally taken amine reaches the circulation at all.

Contested — caveat. The saturation kinetics and the sulfation-to-glucuronidation shift are well established in animal and in-vitro systems and in the paracetamol dose-response in humans. Extrapolating to a specific botanical pairing in a specific person is an extrapolation, and almost none of the individual pairings implied here has been measured in a controlled human study. Treat this as a mechanism that predicts a direction, not a quantified interaction.

Sources: Koster H 1981* · Coughtrie MWH 2016 · Riches Z 2009 · Volak LP 2008

Curcumin as a SULT inhibitor contested in vitro

The comparative study that characterised curcuminoid effects on human drug-metabolising enzymes found inhibition of sulfotransferase activity alongside inhibition of multiple CYP isoforms and of UDP-glucuronosyltransferase, with piperine by contrast being a relatively selective CYP3A4 inhibitor. That result is why turmeric occupies an unusual position in this library: it is a single commonly consumed botanical that touches phase 1, glucuronidation and sulfonation at once, so its interaction profile cannot be reconstructed from a CYP table and cannot be reasoned about as though a single mechanism were in play. The countervailing fact, and it is a large one, is curcumin oral bioavailability, which is very low unless the preparation is designed to overcome it — the same reason piperine is co-formulated with it in the first place, a pairing measured at roughly a twentyfold increase in curcumin bioavailability in human volunteers. The uncomfortable implication is that the formulations most likely to deliver curcumin systemically, which is to say the enhanced-bioavailability ones marketed as superior, are also the ones most likely to produce a real phase-1 and phase-2 interaction. A standard culinary quantity of turmeric and a high-dose piperine-enhanced phytosome capsule are not the same exposure and should not be reasoned about as though they were.

Contested — caveat. The enzyme inhibition data are in vitro. The clinical consequence depends entirely on achieved systemic curcuminoid concentration, which varies by orders of magnitude between culinary turmeric, a plain curcumin capsule and a bioavailability-enhanced formulation. No controlled human study establishes a SULT-mediated clinical interaction for curcumin.

Sources: Volak LP 2008 · Bahramsoltani R 2017 · Shoba G 1998 · Bhardwaj RK 2002

The allylbenzene link: why phase 2 decides the outcome contested animal

The allylbenzenes — safrole, myristicin, elemicin, estragole, methyleugenol, apiole — recur throughout the plant-medicine corpus because they occur in nutmeg, sassafras, basil, tarragon, parsley, dill and calamus, and because the corpus material on oil-based preparations turns on how the body handles them. The metabolic story is a clean two-enzyme-family story and it is worth stating precisely, because it is the clearest example in this library of phase 2 determining whether a phase-1 metabolite is harmless or harmful. Phase 1 CYP enzymes act on these molecules in two competing directions: they can oxidise the allyl side chain or the methylenedioxy ring in ways that lead to ordinary excretable products, and they can hydroxylate at the 1-prime position of the side chain to give the proximate metabolite 1-prime-hydroxy derivative. The CYP isoforms responsible for that 1-prime-hydroxylation have been characterised — CYP2A6 and CYP2C9 with contributions from CYP1A2 and CYP2E1 for safrole. That proximate metabolite is then a substrate for sulfonation, and the sulfate ester is a good leaving group, so the 1-prime-sulfooxy conjugate ionises to a reactive electrophilic carbocation that binds DNA. The evidence that this is the ultimate reactive species is genetic and pharmacological: in brachymorphic mice, which have a defect in PAPS synthesis and therefore impaired sulfation capacity, and in animals treated with a sulfotransferase inhibitor, the hepatocarcinogenicity of the 1-prime-hydroxy metabolite is markedly reduced. In other words, blocking the phase-2 step blocks the toxicity, which establishes that this particular sulfate conjugate is the hazard rather than the detoxification. Two general lessons come out of it. First, sulfonation is not automatically safe conjugation. Second, the same logic that predicts a metabolic potentiation predicts a change in which metabolites form, and a shift in metabolite ratio is not a neutral event — it is the mechanism by which a metabolic manipulation becomes a toxicological one.

• Phase 1: CYP-mediated 1-prime-hydroxylation gives the proximate metabolite; CYP2A6 and CYP2C9 are principal for safrole, with CYP1A2 and CYP2E1 contributing.
• Phase 2: SULT-mediated sulfonation of that metabolite gives the reactive 1-prime-sulfooxy ester, which generates a DNA-binding electrophile.
• Impaired sulfation (brachymorphic mice, sulfotransferase inhibition) markedly reduces the carcinogenicity — the phase-2 step is required for the harm.
• Human urinary metabolite profiles for myristicin, elemicin and safrole are characterised in the forensic toxicology literature, which is how these exposures are detected at all.
• Consequence for anyone reasoning about CYP manipulation and allylbenzene-containing material: shifting which CYP route dominates shifts which metabolites form, and the bioactivation route is one of them.
• This section describes metabolic fate only. It contains no preparation, extraction, dosing or combination guidance of any kind, and none should be inferred from it.

Contested — caveat. The bioactivation mechanism is well established in rodent models and in vitro. Quantitative extrapolation of allylbenzene carcinogenic risk from rodent studies to human dietary or occasional exposure is contested, and regulatory bodies differ in how they treat it. The human metabolite data establish that the pathways operate in people; they do not establish a human cancer risk magnitude at any given intake.

Sources: Jeurissen SMF 2004 · Boberg EW 1983* · Beyer J 2006 · Coughtrie MWH 2016