Cannabis Has More Hidden Chemistry Than We Thought: Flavoalkaloids Just Got Their First Cannabis Sighting

Most of the conversation about Cannabis chemistry happens in two lanes: cannabinoids (THC, CBD, CBG, CBN) and terpenes (myrcene, limonene, pinene, the rest). That's roughly 750 known metabolites in the plant — and in 2025, a research team at Stellenbosch University in South Africa quietly added 25 more to the list. Sixteen of them belong to a class of compounds that had never been reported in Cannabis before.

That class is called flavoalkaloids. If you haven't heard of them, you're in good company — almost nobody in the cannabis space has, because until this paper, they weren't on the Cannabis map.

Let's go through what was actually found, how it was found, and what it does — and crucially, does not — mean.

The paper, briefly

• •Authors: Magriet Muller and Prof. André de Villiers, Department of Chemistry and Polymer Science, Stellenbosch University.
• •Published: Journal of Chromatography A, volume 1754, paper 466023, 2025.
• •DOI: 10.1016/j.chroma.2025.466023.
• •Sample: Three commercially grown Cannabis strains from South Africa.
• •Headline result: 79 phenolic compounds identified. 25 of them had never been reported in Cannabis. 16 of those 25 are tentatively classified as flavoalkaloids — a first for the genus.

That's the entire study at altitude. Now let's go down the layers.

Layer 1: What is a phenolic, and why should you care?

Phenolic compounds are a broad family of plant chemicals built around a phenol ring (a six-carbon ring with an attached hydroxyl group). They include:

• •Flavonoids — quercetin, apigenin, luteolin, kaempferol. Cannabis already has its own flavonoid lineage: the cannflavins.
• •Phenolic acids — caffeic acid, ferulic acid, p-coumaric acid.
• •Stilbenes — resveratrol's family.
• •Lignans, tannins, and a handful of rarer subclasses — including flavoalkaloids.

Phenolics are studied heavily in food chemistry. Tea, wine, cocoa, olives, rooibos, and grape skins are all classical phenolic-rich systems. The same lab at Stellenbosch had been characterizing rooibos, grapes, and wine before turning its instruments on Cannabis.

Important caveat right up front: phenolics are studied in laboratory and food-science contexts for things like antioxidant activity in vitro. That is not the same as a clinical claim about humans. We're not making one.

Layer 2: What's a flavoalkaloid, and why is finding one in Cannabis a story?

A flavoalkaloid is — structurally — a flavonoid skeleton fused with an alkaloid portion. The alkaloid portion is the "nitrogen-containing" piece. Most flavonoids have no nitrogen. Most alkaloids (caffeine, nicotine, morphine, the family of nitrogenous plant chemistry) have no flavonoid backbone. Flavoalkaloids are the rare hybrid.

The class is best known from a few specific places:

• •Tea (Camellia sinensis) — the so-called "puerins," 8-C-N-ethyl-2-pyrrolidinone-substituted flavan-3-ols.
• •Some Aconitum and Buxus species in the broader literature.

Cannabis was not on this list. Until now, even the best-characterized Cannabis chemistry maps did not include flavoalkaloids. So the headline isn't "Cannabis has a new wonder compound." The headline is "a class of compounds we already knew was rare in nature is also present, in small amounts, in Cannabis — we just couldn't see it before."

Layer 3: Why we couldn't see them before — the instrumentation story

This is the part of the paper that, if you care about analytical chemistry at all, is genuinely fun.

The reason flavoalkaloids hadn't been detected in Cannabis is not because the compounds weren't there. It's because in standard one-dimensional liquid chromatography (1D-LC), the rare flavoalkaloid peaks sit underneath the much larger peaks of common flavonoids. They co-elute — the timing on the column overlaps. With a single column, you're effectively asking the instrument to find a whisper underneath a brass band.

Muller and de Villiers used comprehensive two-dimensional liquid chromatography (LC×LC) coupled to high-resolution mass spectrometry. LC×LC runs the sample through two chemically different columns in series, separating compounds on two independent properties. The brass band gets pushed one way, the whispers get pushed another, and suddenly you can resolve them.

Prof. de Villiers said it directly in the press release: "The excellent performance of two-dimensional liquid chromatography allowed separation of the flavoalkaloids from the much more abundant flavonoids, which is why we were able to detect these rare compounds for the first time in Cannabis."

So the right summary is: the compounds were always there. The methodology to see them is what's new. That distinction matters when interpreting the result.

Layer 4: Strain variation is bigger than expected

Here's the result that should reset some assumptions in the cannabis industry.

Across just three strains, the phenolic profile varied dramatically. The 16 flavoalkaloids were concentrated mainly in the leaves of one strain. Not all three. Not evenly distributed. One strain, one tissue.

Three observations follow:

1. •Genotype matters more than people credit. If three strains produce three substantially different phenolic profiles, the popular shorthand of "indica vs sativa vs hybrid" is doing even less work than terpene researchers have been arguing for years.
2. •Tissue matters. Leaves are typically classified as waste in commercial cultivation. The data suggests leaves may carry a distinct phenolic profile that flower does not.
3. •Sample size is small. Three strains is hypothesis-generating, not definitive. Anyone repeating this work will need to cast a wider net to know whether flavoalkaloid presence is a quirk of one cultivar or a general feature of certain genetic lineages.

Layer 5: What this does not mean

We have to be precise here, because the press release headline ("rare compounds with medical potential") is doing a lot of work, and the underlying paper is more reserved.

This study did not:

• •Test these compounds in any human or animal model.
• •Isolate or purify any flavoalkaloid in quantities sufficient for biological assay.
• •Make any therapeutic, dosing, or "wellness benefit" claim.
• •Compare smoked, vaped, or extracted product profiles.
• •Make a claim that leaves are "more medicinal" than flower.

What it did was detect and tentatively identify a set of compounds. "Tentative" matters in mass spectrometry — full structural confirmation requires reference standards, and many of these compounds don't have commercially available standards yet.

So when you hear "flavoalkaloids in Cannabis," the right mental model is: new map, not new medicine. The map is the contribution. What you build on top of the map is the next ten years of work.

Layer 6: Why this still matters

Even with the appropriate caution, three things follow from this study:

1. •

   The non-cannabinoid chemistry of Cannabis is under-characterized. If three strains produced 25 first-time-reported phenolics, the genuine number across all cultivars is almost certainly in the hundreds. The plant has more chemistry than the THC/CBD/terpene conversation captures.

2. •

   Plant material currently treated as waste may have research value. Trim, fan leaves, and stems are often discarded or composted. If the leaves of even some strains carry distinct phenolic profiles, the by-product stream is worth a second look from a research standpoint. (For licensed operators: this is a research observation, not a green light to market leaves as a finished consumer product. Local rules govern what you can sell.)

3. •

   Methodology drives discovery. A lot of the next decade of Cannabis chemistry will come from instruments that resolve overlapping peaks better than 1D-LC ever could. Expect more "first-evidence-of" papers as LC×LC, ion mobility, and high-resolution MS become more common in cannabis labs.

What we'll be watching for

• •Replication studies on a wider strain panel.
• •Tissue-by-tissue mapping (flower vs leaves vs stems vs roots) for phenolics across cultivars.
• •Reference-standard confirmation of the 16 tentatively identified flavoalkaloids, so the "tentative" qualifier can be removed.
• •Any follow-up bioactivity work — but only in proper, peer-reviewed contexts.

We'll update this article as published work appears.

Bottom line

A research team in South Africa pointed a much sharper instrument at Cannabis and found compounds nobody had reported before. That is genuinely useful. It does not, on its own, change how anyone should think about consuming Cannabis. It changes what we know is in the plant — and that is the foundation everything else gets built on.

— Nug

---

Source citation (full): Muller M, de Villiers A. Comprehensive two-dimensional liquid chromatographic analysis of Cannabis phenolics and first evidence of flavoalkaloids in Cannabis. Journal of Chromatography A. 2025;1754:466023. DOI: 10.1016/j.chroma.2025.466023.

Disclosures: This article summarizes a peer-reviewed analytical chemistry study. It is not medical advice. It does not diagnose, treat, cure, or prevent any condition. Phytopedia is a knowledge platform, not a healthcare provider.

---

Editor checklist (before publish)

• • Wellness language: every "potential health benefit" reference framed as research-context, not consumer claim.
• • Cite the DOI in the article body and in metadata.
• • Confirm the topic taxonomy term — propose adding phenolics if not present.
• • Tag source: "research-literature".
• • Set wellness_reviewed: true only after a human pass.
• • Confirm character voice = Nug (no "feel free", no first-person consumer experience).
• • Decide whether to also surface on Mimea Learn Center (credit Phytopedia).