Cannabis Flavonoids: The Anti-Inflammatory Compounds Beyond Terpenes

Ask a cannabis consumer what’s in the plant and you’ll hear: cannabinoids and terpenes. Most of the entourage-effect literature is built on those two pillars. It’s also incomplete.

Cannabis produces a third category of bioactives that almost nobody talks about — even though one member was shown, in human rheumatoid synovial cells, to inhibit a key inflammatory mediator roughly thirty times more potently than aspirin. The compounds are flavonoids, and the cannabis-specific ones are called cannflavins. They were discovered in the 1980s, sat largely ignored for thirty years, and have re-emerged as one of the more interesting unsolved questions in cannabis pharmacology.

This is a serious look at what flavonoids are, what cannflavin A, B, and C actually do, why almost no consumer has heard of them, and what the practical pharmacology of flower really supports. The science is real, but the dose-from-a-joint reality is more modest than the headlines suggest. Both things can be true.

A note: Science article, not medical advice. If you are managing a chronic inflammatory condition, work with a clinician.

What Flavonoids Actually Are

Flavonoids are a vast family of plant secondary metabolites — small polyphenolic molecules built around a 15-carbon, three-ring scaffold (the flavone backbone). Roughly 6,000 distinct flavonoids have been characterized across the plant kingdom, responsible for much of the color of fruits and flowers.

They also do interesting things in animal cells: antioxidant activity, modulation of inflammatory pathways, kinase effects, binding to a wide variety of enzymes and receptors. Their planar polyphenol structure fits the active sites of many enzymes that handle planar substrates — including several that drive inflammation.

In cannabis, more than 20 flavonoids have been identified. Most are common across the plant world: apigenin, luteolin, kaempferol, quercetin, vitexin, orientin, myricetin — molecules you eat every day in parsley, kale, onions, and tea.

What makes cannabis distinctive is a small subset of prenylated and geranylated flavones essentially unique to the plant: cannflavin A, cannflavin B, and cannflavin C. Prenylation hangs a small isoprene-derived tail off the flavone — the same kind of decoration that turns olivetolic acid into cannabinoids. For practical purposes, cannflavins are a cannabis signature.

Cannflavin A: The Flagship

The story starts in 1985. Marilyn Barrett, a PhD student at the School of Pharmacy in London, was working in Frank Evans’s lab on a simple question: which cannabis compound is responsible for the plant’s anti-inflammatory effect independent of the cannabinoids?

Her assay used synovial cells — the lining cells of the joint — harvested from rheumatoid-arthritis surgery patients. In culture, those cells spontaneously released prostaglandin E2 (PGE2), the lipid mediator your body makes when a joint is hot, swollen, and painful. Whatever blocked PGE2 release in that dish was doing something inflammation cared about.

Barrett took a cannabinoid-free alcohol extract, fractionated it, ran each fraction through the assay, and kept going until a single compound fell out. She called it Cannflavin [Barrett et al., 1985]. A year later, the structure resolved into two prenylated flavones, Cannflavin A and B [Barrett et al., 1986].

The numbers from that 1985 paper have echoed through every flavonoid review since:

• Cannflavin mean IC₅₀ for inhibiting TPA-induced PGE2 release: 31 ng/mL
• Aspirin in the same assay: 840 ng/mL
• Indomethacin: 1.7 ng/mL
• Dexamethasone: 0.27 ng/mL

So cannflavin was about 27–30× more potent than aspirin on a mass basis — the famous “30× aspirin” line that echoes through popular write-ups. It was less potent than indomethacin and far less than dexamethasone. Cannflavin sits in the middle of the anti-inflammatory potency spectrum: comfortably above OTC aspirin and ibuprofen, comfortably below prescription COX inhibitors and steroids.

A second paper took the mechanism further. In purified enzyme assays, cannflavin A and B act on microsomal prostaglandin E synthase-1 (mPGES-1) — an inducible enzyme downstream of COX-2 selectively responsible for inflammatory PGE2 — and on 5-lipoxygenase (5-LOX), which makes leukotrienes [Werz et al., 2014]. IC₅₀ values fell in the low micromolar range. Critically, cannflavin A is only a weak inhibitor of COX-1 and COX-2 themselves — meaning it can quiet inducible inflammation without the gastric and renal side effects of long-term NSAID use. That is a more selective profile than an NSAID, closer to what a drug developer would design for a cleaner anti-inflammatory.

The 2019–2025 Reframing

Why did cannflavin sit on the shelf for thirty years? It is hard to extract, and prohibition kept cannabis pharmacology starved of funding through the 1990s and 2000s.

The modern revival started with a 2019 paper from the University of Guelph mapping the biosynthetic pathway to cannflavin A and B [Rea et al., 2019]. The team identified two enzymes — an O-methyltransferase (CsOMT21) and an aromatic prenyltransferase (CsPT3) — that take luteolin, methylate it to chrysoeriol, and attach either a geranyl group (cannflavin A) or a prenyl group (cannflavin B). That pathway opens a route to producing cannflavins in yeast, which solves the supply problem.

A 2020 Fitoterapia scoping review concluded the compounds had “promising therapeutic properties, most notably as an anti-inflammatory agent” [Erridge et al., 2020]. Since then, computational work has expanded the profile: cannflavin A binds the kinase TAK1 (a master regulator of inflammatory gene expression) with high predicted affinity, and the cannflavins dock favorably into the active site of the SARS-CoV-2 papain-like protease [Wanas et al., 2025]. These are not clinical claims — they are structural data that drug-discovery programs build on. A 2025 Journal of Cannabis Research paper also confirmed that cannflavin A and B dominate the prenylated-flavonoid profile in cultivated cannabis, with concentrations varying enormously by chemovar [De Leo et al., 2025] — an opening for breeders to select for cannflavin-rich genetics the way they once selected for CBG and CBC.

The pharmaceutical pipeline is real but early: synthetic cannflavin analogs for arthritis and GI inflammation, fermentation production licensed by several companies. We are roughly where THC and CBD were in 1995 — chemistry mapped, biology promising, clinic still distant.

The Other Cannabis Flavonoids

Cannflavins get the press, but cannabis carries a wider flavonoid profile worth knowing:

• Apigenin — also in chamomile and parsley. Mild anxiolytic via partial benzodiazepine-receptor binding. The base flavone scaffold cannflavin biosynthesis branches from.
• Luteolin — a 3’-hydroxylated cousin of apigenin and the direct biosynthetic precursor to chrysoeriol → cannflavin A/B. Inhibits NF-κB activation.
• Kaempferol — a flavonol common in kale and onions. Antioxidant, kinase modulator.
• Quercetin — the heavyweight of dietary flavonoids. Mast-cell stabilizer, COX-2 modulator, broad antioxidant.
• Vitexin and orientin — C-glycosylated flavones (sugar attached carbon-to-carbon, unusually stable). Anxiolytic and anti-inflammatory in small animal models.
• Cannflavin C — identified in 2008 by ElSohly at the University of Mississippi. Less studied; preliminary antiparasitic activity (notably against Leishmania).

The thing to notice: most of the activity overlaps with what cannabis terpenes already do, but through different molecular machinery. Apigenin is anxiolytic at the benzodiazepine site; linalool is anxiolytic via GABA. Caryophyllene is anti-inflammatory at CB2; cannflavin A is anti-inflammatory at mPGES-1. Same direction, different doors into the cell — exactly the kind of redundancy that makes full-spectrum behave differently than the sum of isolates.

Why Almost Nobody Has Heard of Them

Three honest reasons.

Concentration. Cannabinoids dominate flower at 10–25% of dry weight. Terpenes come in around 1–3%. The entire flavonoid pool sits at 0.1–1%, and cannflavins specifically are a fraction of that — an order of magnitude less per gram than terpene, two orders of magnitude less than cannabinoid.

Heat. Combustion at 600–900 °C — the temperature inside a lit joint — is brutally destructive to plant secondary metabolites. A meaningful fraction of flavonoid mass in smoked flower does not make it past the cherry. Vaporization at 175–215 °C is gentler but still non-trivial; cold-process methods preserve far more.

Pharmacokinetics. Bioavailability of prenylated flavones in humans is poorly characterized but generally low — lipophilic, heavily glucuronidated in gut and liver, hitting systemic circulation at a small fraction of ingested dose. Same story as quercetin and apigenin. The eventual pharmaceutical play is most likely a synthetic analog or formulated delivery system, not raw extract.

Add those together: cannflavins are real and interesting, but the dose you extract from a bowl of flower is small. That is why practical pharmacology has lagged the chemistry.

The Entourage Effect, Refined

The standard entourage-effect story is “cannabinoids and terpenes do more together than alone.” That has always been a simplification. The accurate version is that cannabis produces a layered cocktail of bioactives hitting overlapping inflammatory and analgesic pathways through different doors:

• THC and CBD modulate CB1/CB2 signaling and downstream cytokine production.
• Caryophyllene is a selective CB2 agonist with its own anti-inflammatory signal.
• Limonene and myrcene carry modest activity at NF-κB and other targets.
• Cannflavin A and B quietly inhibit mPGES-1 and 5-LOX in parallel — a pathway nobody else in the cocktail touches with the same selectivity.
• Apigenin and quercetin add antioxidant and mast-cell-stabilizing activity.

No single compound does the heavy lifting. The argument for full-spectrum is not that any molecule is a miracle, but that redundancy across pathways probably matters more than any single binding event. That helps explain the gap between purified-cannabinoid clinical data and the real-world responses people report from whole flower — a gap that shrinks when you factor flavonoids in. For chronic-inflammation use, that layered model is why I point readers toward our complete molecular science of cannabis and inflammation and pain management guide rather than any one strain or compound.

Routes That Preserve Flavonoids

If flavonoids are part of why you reach for cannabis, route matters more than usual. In rough order of preservation:

1. Fresh raw cannabis (juicing, salads) — highest theoretical preservation. No heat, full polyphenol load. Mostly limited to home growers.
2. Cold-process tinctures (alcohol or glycerin) — preserve a substantial flavonoid fraction. Ethanol is a good solvent for prenylated flavones.
3. Low-temperature edibles — long, gentle-heat extractions retain more than vaporization.
4. Vaporization (175–215 °C) — preserves cannabinoids and terpenes; degrades a meaningful fraction of flavonoids at the higher end.
5. Combustion (smoking) — worst for flavonoid preservation. The cherry destroys the bulk of the cannflavin pool.

This reframes the older “raw cannabis” advocacy — the Courtney protocol and similar — into something with clearer molecular logic. Track full-spectrum vs. isolate, and route vs. route, in the High IQ app once and the pattern in your own response gets visible.

Strain-Level Notes

Cannflavin content varies by chemovar more than most consumers appreciate, but the data remain incomplete. Two practical observations:

• High-CBD chemotypes like ACDC tend to have richer flavonoid profiles than THC-dominant lines, partly because metabolic flux into prenylation is not all locked up making cannabinoids.
• Indica-leaning, myrcene-rich chemotypes like Granddaddy Purple and Northern Lights often pair their terpene profile with a respectable flavonoid load, though quantitative data are sparse.

If you are choosing for the Relief or Balance High Family — the families most associated with chronic-inflammation use — the flavonoid argument is one more reason to favor whole-plant tinctures over isolate or smoked flower.

What This Means for Consumers

A few honest takeaways:

1. Cannabis flavonoids are real, bioactive, and underexplored. Cannflavin A and B have a selectivity profile (mPGES-1 and 5-LOX over COX-1/COX-2) that drug developers find attractive.
2. The “30× aspirin” line is technically correct but contextually limited. It is a 1985 cell-culture assay, mass basis, against a weak NSAID. It does not mean a joint hits like ibuprofen.
3. The dose from flower is modest. Flavonoids are 0.1–1% of dry weight; cannflavins a fraction of that; combustion destroys a share; oral bioavailability is low.
4. Full-spectrum still wins on logic. Cannflavins quietly close a loop in the entourage model — selective inducible-PGE2 inhibition — that cannabinoids and terpenes don’t cover.
5. Route matters. Fresh, cold-process, and gentle-heat methods preserve substantially more flavonoid mass than smoking.
6. The pharmaceutical pipeline is the watch-list. If a synthetic cannflavin analog reaches Phase 2 in arthritis or IBD, that is when this category gets its serious answer.

Cannabis has a third bioactive dimension. Small, fragile, and genuinely interesting. The chemistry is solid, the preclinical pharmacology promising, the practical contribution per dose real but modest — exactly the kind of underexplored territory that makes cannabis pharmacology worth following over the next decade.

For the broader picture — cannabinoids, terpenes that hit adenosine receptors, terpene synergy, and now flavonoids — start with our complete inflammation molecular science deep dive.

Sources

• Barrett, Gordon, & Evans (1985). Isolation from Cannabis sativa L. of cannflavin — a novel inhibitor of prostaglandin production. Biochem Pharmacol 34(11), 2019–2024. doi:10.1016/0006-2952(85)90325-9
• Barrett, Scutt, & Evans (1986). Cannflavin A and B, prenylated flavones from Cannabis sativa L. Experientia 42(4), 452–453. doi:10.1007/BF02118655
• Werz et al. (2014). Cannflavins from hemp sprouts target mPGES-1 and 5-LOX. PharmaNutrition 2(3), 53–60. doi:10.1016/j.phanu.2014.05.001
• Rea et al. (2019). Biosynthesis of cannflavins A and B from Cannabis sativa L. Phytochemistry 164, 162–171. doi:10.1016/j.phytochem.2019.05.009
• Bautista, Yu, & Tian (2021). Flavonoids in Cannabis sativa: Biosynthesis, bioactivities, and biotechnology. ACS Omega 6(8), 5119–5123. doi:10.1021/acsomega.1c00318
• Erridge et al. (2020). Cannflavins — From plant to patient: A scoping review. Fitoterapia 146, 104712. doi:10.1016/j.fitote.2020.104712
• Wanas et al. (2025). In silico investigation of cannflavins as anti-SARS-CoV-2 agents. Natural Product Research 39(5), 1081–1094. doi:10.1080/14786419.2023.2294111
• De Leo et al. (2025). Antioxidant, antimicrobial and anti-inflammatory activity of non-psychotropic Cannabis sativa L. J Cannabis Research. doi:10.1186/s42238-025-00336-1
• Computational assessment of cannflavin A as a TAK1 inhibitor (2023). Scientia Pharmaceutica 91(3), 36. doi:10.3390/scipharm91030036