Curcumin barely dissolves in water, so most of an oral dose passes through the small intestine undissolved, and an estimated less than 10% enters gut cells.Plain curcumin 1.5 g (common capsule dose) enters stomach → small intestine.
First gate: curcumin is extremely poorly water-soluble (similar to a fat-soluble pigment in olive oil), and most of it never dissolves in the small intestine, passing through with the chyme.
· 5 g of turmeric powder (one teaspoon of curry) contains ~ 150-250 mg of curcuminoids — less than 1/6 of a "therapeutic dose" · Oil-dissolved + taken with meals modestly raises absorption (~ 2-3×), still far short · Actual fraction entering enterocytes: estimated < 10%
So "more curry" = an extra 100 mg of curcumin at the mouth ≈ not enough to reach the active concentrations seen in in-vitro literature — this step alone loses 90%.
That is why the bottom of this step reads: "The story of oral curcumin has already lost half before it even begins".
2 · Gut + liver Phase II
The little curcumin that enters gut cells is pumped back by P-gp or inactivated by Phase II metabolism, so free curcumin in plasma stays nanomolar.The small amount of curcumin that luckily enters the enterocyte encounters two downstream barriers:
A · P-gp efflux pump (P-glycoprotein, -driven) · Actively pumps curcumin back into the gut lumen, as if it had never entered · Piperine (black pepper extract) inhibits P-gp + UGT — this is the scientific basis of the classic "curcumin + black pepper" combination (Shoba 1998)
B · Phase II metabolism (gut wall + liver) · UGT adds glucuronide → curcumin glucuronide (water-soluble, ready for excretion) · SULT adds sulfate → curcumin sulfate · Both steps happen in the gut wall — the liver doesn't even see "intact curcumin"
Plasma half-life < 2 hours — much faster than caffeine (5-7h) or vitamin D (~ 15 days).
Result: after gram-scale oral curcumin, plasma free curcumin is often only in the nanomolar range (~ 50 nmol/L), with 96-99% as already-inactivated metabolites (glucuronide / sulfate), which have almost no anti-inflammatory activity — a point often overlooked. 8 g turmeric powder is not 8 g curcumin.
So most in-vivo studies measuring "plasma curcumin" are actually measuring total metabolites, not active free curcumin.
3 · The 20-200× gap
After oral dosing, free curcumin in plasma reaches only about 50 nmol/L, while in-vitro studies need 1-10 µmol/L to see effects, a gap of 20-200 times.Core numerical gap:
· Oral 8 g curcumin → plasma free ~ 50 nmol/L (= 0.05 µM) · In-vitro literature "effective anti-inflammatory" concentrations: 1-10 µmol/L · Gap: 20-200× — and in-vitro cell-culture medium gets full drug contact, while in vivo you still have to subtract tissue penetration — so the real gap is only larger
This is the physical root of "strong in-vitro activity + unstable in-vivo effect":
· "Curcumin inhibits " → true, but only significant in vitro at 5 µM · "Curcumin activates Nrf2" → true, but again at the µM level · "Curcumin is anti-cancer" → cells die in vitro, but in-vivo 0.05 µM is nowhere near enough
Nelson 2017 J Med Chem warning: curcumin is a "PAINS + IMPS" (pan-assay interference + invalid metabolic panaceas) — it can show activity in almost any in-vitro assay, and this universal reactivity makes screening results overly optimistic. Nelson conclusion: no double-blinded placebo-controlled trial of curcumin has been successful; it is a highly improbable lead. Local tissue levels, metabolite activity, or enhanced forms reaching µM do not reverse that conclusion.
4 · Enhanced forms · form > dose
Adding piperine or using enhanced forms such as phospholipid complexes raises plasma exposure; trials use different forms, so form matters more than dose.Mainstream absorption-enhancement strategies (relative to plain curcumin):
· + Piperine 5-20 mg — inhibits P-gp + UGT → bioavailability ↑ about 20-fold (Shoba 1998: a small pharmacokinetic study in healthy volunteers — 2 g of curcumin alone was barely detectable in blood, and adding 20 mg of piperine raised it clearly) · Meriva® / phytosomal (Indena phospholipid complex) — complexed with lecithin, improving membrane penetration; plasma exposure rises vs plain powder, but the fold depends on whether the assay is free curcumin or hydrolyzed total, and whether doses were aligned · Theracurmin® (nanoparticulate, Japan) — particles < 200 nm; also raises plasma exposure — do not multiply 50 nM by a fold and call that µM · Longvida® SLCP (solid lipid microparticle) — designed for brain penetration, common in AD research · BCM-95 (mixed with turmeric essential oil) — the form used by Lopresti's depression
Key facts:
· Enhanced forms raise plasma exposure, but you cannot take the 50 nM from 8 g unenhanced curcumin, multiply by a fold, and treat that as having reached in-vitro µM · Different clinical trials use different forms → equivalent doses can differ a lot · When you see "curcumin trial positive / negative" → first ask which form was used
· First choice: curcumin + BioPerine® (piperine) — cheapest + the classic RCT form · Second choice: Meriva® — used in multiple OA / depression RCTs · Third choice: Theracurmin® — pricier, but commonly used in Japanese RCTs · Skip: products marketed as "nano super-soluble" / "patented novel" without clear bioavailability data
> "More curcumin" ≠ "more in the blood" > Wrong form, 10× the dose is useless; right form, 1/5 the dose works better
This is the sharpest lesson in learning "how to evaluate any supplement": mechanism → bioavailability → form → clinical evidence → dose — break any link in this chain and the product drops into the "expensive urine-coloring agent" category.