Rosmaricine

Rosmaricine at a Glance

Rosmaricine is an amino-abietane diterpene — the catechol-bearing tetracyclic frame shared by rosemary’s carnosic acid, carnosol and rosmanol, carrying a lactone and, unusually for a plant terpenoid, a free amino group at the benzylic C-7 position ^{3Reference 3Hanson JR · 2016Rosemary, the beneficial chemistry of a garden herbView study →}. It is a discrete, well-defined molecule: C₂₀H₂₇NO₄, molecular weight 345.4, PubChem CID 148619119, CAS 3650-11-1.

Almost everything else about it comes with a caveat, and the caveat is the page.

It is very probably not a rosemary constituent. Rosmaricine is obtained from dried rosemary leaf treated with ammonia in the presence of air, and the current reading of that procedure is that the ammonia is where the nitrogen comes from: the molecule is formed by reaction between oxidation products of carnosic acid and the ammonia added to liberate the alkaloids the Soviet chemists believed were in the plant ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}. Wenkert, Fuchs and McChesney reached that conclusion in 1965, in a paper whose title says it plainly — Chemical Artifacts from the Family Labiatae ^{1Reference 1Wenkert E et al. · 1965Chemical artifacts from the family LabiataeView study →}. Sixty years later the standard review of rosemary chemistry still records rosmaricine as “an anomalous ‘alkaloid’” shown to be an artefact, with the C-7 amino group introduced from the isolation ammonia ^{3Reference 3Hanson JR · 2016Rosemary, the beneficial chemistry of a garden herbView study →}.

Its literature is six PubMed records. Two of them have been retracted. One is a computational docking exercise. One is a 1994 anticancer screen that came back negative. One is a rosemary-extract paper. Exactly one study has measured what the isolated molecule does to anything — and that study made it in the lab, on purpose, to use as a synthetic scaffold ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}.

So the job is to close a loop, not to open one. Rosemary’s diterpene pharmacology belongs to carnosic acid and carnosol; rosmaricine is what happens when you go looking for alkaloids in a plant that does not have any.

Is It Actually in the Plant?

The honest answer is: nobody has shown that it is, and the people who work with it say it is not.

The case against natural occurrence rests on three things.

The chemistry. Rosmaricine is recovered only from leaf treated with ammonia and air. Carnosic acid is notoriously unstable and oxidises readily — to carnosol, and onward to rosmanol, epirosmanol, isorosmanol and methyl carnosate — so an ammoniacal extract of rosemary is a flask full of reactive quinone-type oxidation intermediates sitting in a nitrogen nucleophile ^{12Reference 12Birtić S et al. · 2015Carnosic acidView study →}. Adding ammonia across such an intermediate at the benzylic position is an ordinary reaction, not a remarkable one. Wenkert and colleagues made exactly this argument in 1965 and grouped rosmaricine with other Labiatae “isolation products” that were never in the plant ^{1Reference 1Wenkert E et al. · 1965Chemical artifacts from the family LabiataeView study →}.

The attribution record. A PubMed search for rosmaricine AND "Rosmarinus officinalis" returns zero records. So does rosmaricine AND "Salvia rosmarinus". A search for rosmaricine AND rosemary returns one — a 2000 cosmetic-dermatology paper on an alcoholic rosemary extract whose substance indexing lists the compound ^{14Reference 14Calabrese V et al. · 2000Biochemical studies of a natural antioxidant isolated from rosemary and its application in cosmetic dermatologyView study →}. For a molecule that has been on rosemary constituent lists since the 1960s, that is a striking absence: sixty years of rosemary phytochemistry, including modern LC-MS profiling of the diterpene fraction, has not produced a paper that names it as something found in the plant.

What its own investigators say. The Milan group that has worked with rosmaricine longest describes it, in print and without hedging, as a compound “obtained from dry leaves of Rosmarinus officinalis L. treated with ammonia in the presence of air”, formed “through a complex reaction between some oxidation derivatives of carnosic acid and the ammonia used to liberate the alkaloids supposed to be present in the plant” ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}. They use it as a scaffold precisely because it is an unusual synthetic starting material, not because it is a phytochemical.

None of this is a proof of absence. A sensitive modern assay on ammonia-free extract has, as far as the indexed literature goes, never been run, so the possibility of a trace natural pool has not been formally excluded. But the burden of proof sits with the claim of natural occurrence, and there is nothing holding it up.

The Whole Literature, Listed

Six records. It is short enough to print.

Year	Study	What it was	Ref
1965	Legchaev	Soviet cardiovascular pharmacology, Russian-language, abstract only	^{7Reference 7Legchaev VYa · 1965Legchaev VYa (1965). [On the mechanism of action of rosmaricin on the cardiovascular system]. Farmakologiia i Toksikologiia 28:170–172. Russian. https://pubmed.ncbi.nlm.nih.gov/5849197/View study →}
1967	Grishaenkov et al.	Soviet radiation-sickness study in mice, Russian-language, abstract only	^{8Reference 8Grishaenkov VM et al. · 1967AnimalGrishaenkov VM, Ivanov MI, Legchaev VYa (1967). [The effect of rosmaricin on mice in acute radiation sickness]. Meditsinskaia Radiologiia 12:69–70. Russian. https://pubmed.ncbi.nlm.nih.gov/5621330/View study →}
1994	Boido et al.	Rosmaricine derivatives vs. P388 leukaemia — inactive	^{6Reference 6Boido A et al. · 1994Rosmaricine derivatives as potential antitumoral agentsView study →}
2000	Calabrese et al.	Rosemary extract in cosmetic dermatology; indexed to the compound	^{14Reference 14Calabrese V et al. · 2000Biochemical studies of a natural antioxidant isolated from rosemary and its application in cosmetic dermatologyView study →}
2011	Chang et al.	Docking against influenza haemagglutinin	^{11Reference 11Chang TT et al. · 2011Screening from the world’s largest TCM database against H1N1 virusView study →}
2011	Yang et al.	Docking against EGFR — RETRACTED 2026	^{9Reference 9Yang SC et al. · 2011ReviewRETRACTED: Identification of potent EGFR inhibitors from TCM Database@TaiwanView study →}
2011	Chang et al.	Docking against H1N1 — RETRACTED 2026	^{10Reference 10Chang SS et al. · 2011RETRACTED: Two birds with one stone? Possible dual-targeting H1N1 inhibitors from traditional Chinese medicineView study →}
2017	Gabriele et al.	Sulfurated derivatives; the only measured pharmacology	^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}

Two of the eight are retracted. Three of the eight are molecular docking. Two are Soviet abstracts from the 1960s that are not retrievable in full. That leaves one paper with a number in it that belongs to this molecule.

What It's Used For

One real dataset exists, and it is in vitro.Rosmaricine binds the STAT3-SH2 domain in a cell-free AlphaScreen assay with IC₅₀ 1.9 ± 0.1 µM, inhibits a STAT3 reporter in HeLa cells at 42.4 µM, inhibits NF-κB promoter activity in HCT-116 cells at 43.4 µM, and is antiproliferative against those cells at 65.2 µM ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}.

The selectivity is poor.It hits the closely related STAT1 at 10.0 µM — about five-fold — so it is not a selective STAT3 binder ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}.

The one animal test was negative.Rosmaricine and a set of its O-methylated and N-acyl derivatives were injected intraperitoneally in mice bearing P388 lymphocytic leukaemia. One compound was borderline; everything else, including rosmaricine, was inactive ^{6Reference 6Boido A et al. · 1994Rosmaricine derivatives as potential antitumoral agentsView study →}.

None of it transfers to rosemary.If the molecule is not in the plant, a µM STAT3 number is a fact about a laboratory compound, not about a herb. It should not be added to the rosemary evidence base, and the rosemary diterpene story belongs to carnosic acid and carnosol ^{12Reference 12Birtić S et al. · 2015Carnosic acidView study →}.

Three of the eight papers are docking studies and two of those are retracted^{9,10,11Reference 9Yang SC et al. · 2011ReviewRETRACTED: Identification of potent EGFR inhibitors from TCM Database@TaiwanView study →Reference 10Chang SS et al. · 2011RETRACTED: Two birds with one stone? Possible dual-targeting H1N1 inhibitors from traditional Chinese medicineView study →Reference 11Chang TT et al. · 2011Screening from the world’s largest TCM database against H1N1 virusView study →}. A docking score is a hypothesis; a retracted docking score is not even that.

Evidence by IndicationStrength of support · Beta

Support is an experimental score — weighted by study type (human > animal > in vitro) and volume. A beta, not a validated metric. Every score here is capped very low by design: the literature is six records, there is no animal efficacy result, no human data, and a live question over whether the molecule occurs in nature at all.

The only measured pharmacology rosmaricine has. In a 2017 study designed to build sulfurated drug hybrids, the parent molecule was profiled alongside its three new N-acyl derivatives. Rosmaricine bound the STAT3-SH2 domain strongly in the cell-free AlphaScreen assay — 99.7% inhibition at 30 µM, IC₅₀ 1.9 ± 0.1 µM — and carried that through into cells, inhibiting a STAT3 reporter in HeLa at IC₅₀ 42.4 ± 2.4 µM and NF-κB promoter activity in HCT-116 at IC₅₀ 43.4 ± 6.9 µM, with antiproliferative activity against HCT-116 at IC₅₀ 65.2 ± 2.3 µM ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}.

Two things temper it. Selectivity is weak — STAT1, which shares 78% sequence homology with STAT3 and has tumour-suppressive roles, was inhibited at 10.0 µM, a selectivity index of only about five ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}. And the 30-fold gap between the cell-free IC₅₀ (1.9 µM) and the cellular reporter IC₅₀ (42 µM) is the usual signature of a compound that binds a purified domain far better than it reaches the same protein inside a cell.

The paper’s own interest was in the hybrids, not the parent. Amidifying rosmaricine with a methanesulfonylthio-pentanoic acid gave a 10-fold gain on STAT3 (IC₅₀ 0.2 µM), and the allyldisulfanyl derivative was 19-fold more antiproliferative than rosmaricine (IC₅₀ 3.5 µM on HCT-116, versus 21.2 µM on normal human smooth-muscle cells) ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}.

Those numbers belong to synthetic derivatives and must not be read back onto the parent, let alone onto rosemary.

Gap: one laboratory, one paper, no replication, no animal work, no pharmacokinetics. And the compound was synthesised for the study — this is medicinal chemistry on a scaffold, not phytopharmacology.

Mechanisms

Target / pathway	Effect	System	Relevance
STAT3-SH2 domain	binding inhibition, IC₅₀ 1.9 µM	cell-free AlphaScreen ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}	in vitro only
STAT1-SH2 domain	binding inhibition, IC₅₀ 10.0 µM	cell-free AlphaScreen ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}	selectivity index ~5
STAT3 reporter	IC₅₀ 42.4 µM	HeLa cells ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}	in vitro only
NF-κB promoter	IC₅₀ 43.4 µM	HCT-116 cells ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}	in vitro only
Cell proliferation	IC₅₀ 65.2 µM	HCT-116 cells ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}	weak, non-selective
P388 leukaemia	inactive	mouse, i.p. ^{6Reference 6Boido A et al. · 1994Rosmaricine derivatives as potential antitumoral agentsView study →}	the only in-vivo efficacy test

Pharmacokinetics

None. There is no absorption, distribution, metabolism or excretion study of rosmaricine in any species, and no analytical method for measuring it in a biological matrix. The question is also somewhat moot: a molecule that is not present in the plant is not something an oral rosemary preparation can deliver, so there is no dietary exposure to characterise.

Clinical Trials

None, under any name, at any time.

Completed	Planned	Terminated	Preclinical
—(none)	—	—	1in-vitro paper, 1 negative mouse screen

Last checked: August 2026.

Where It Sits in the Rosemary Diterpene Cascade

Rosemary’s real diterpene story is one of instability. Carnosic acid is the plant’s principal phenolic diterpene and the active carried by food-grade rosemary extract, and it oxidises readily — first to carnosol, then onward to rosmanol, epirosmanol, isorosmanol and methyl carnosate. The composition of any given rosemary preparation is therefore a function of extraction solvent, heat, light and storage age as much as of the plant ^{12Reference 12Birtić S et al. · 2015Carnosic acidView study →}.

Rosmaricine is best understood as a branch off that cascade that only opens in the laboratory. Where carnosol and rosmanol are what carnosic acid becomes in air, rosmaricine is what one of those oxidation intermediates becomes in air plus ammonia ^{1,2Reference 1Wenkert E et al. · 1965Chemical artifacts from the family LabiataeView study →Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}. The skeleton is genuinely rosemary’s — it is a carnosol-family abietane, and the structure of carnosol was settled by an overlapping group of chemists in 1964 ^{13Reference 13Brieskorn CH et al. · 1964The structure of carnosolView study →} — but the nitrogen is the chemist’s, not the plant’s.

The Soviet group did not stop at one such compound: a fourth paper in the same series described isorosmaricine, a stereoisomeric partner recovered from the same kind of preparation ^{5Reference 5Yakhontova LD et al. · 1971A study of Rosmarinus officinalis IVView study →}, which is what one would expect if the amino group is being added non-stereospecifically to an oxidised intermediate rather than installed by an enzyme.

Discovery & Synthesis

The molecule was described in the early 1960s by L. D. Yakhontova and A. D. Kuzovkov at the All-Union Scientific Research Institute of Pharmaceutical Chemistry in Moscow, in a series titled Alkaloids of the Rosmarinus genus — the framing matters, because they were looking for alkaloids and used ammonia to liberate the free bases they expected to find. The third paper in the series, published in Chemistry of Natural Compounds in 1967, is titled The nature of rosmaricine from Rosmarinus officinalis ^{4Reference 4Yakhontova LD · 1967The nature of rosmaricine from Rosmarinus officinalisView study →}; a fourth, in 1971, added isorosmaricine ^{5Reference 5Yakhontova LD et al. · 1971A study of Rosmarinus officinalis IVView study →}.

In between, Wenkert, Fuchs and McChesney published Chemical Artifacts from the Family Labiatae in the Journal of Organic Chemistry in 1965 ^{1Reference 1Wenkert E et al. · 1965Chemical artifacts from the family LabiataeView study →} — the paper that is still cited, sixty years later, as the demonstration that the C-7 amino group came out of the ammonia bottle ^{2,3Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →Reference 3Hanson JR · 2016Rosemary, the beneficial chemistry of a garden herbView study →}. Several of the same chemists had settled the structure of carnosol the year before ^{13Reference 13Brieskorn CH et al. · 1964The structure of carnosolView study →}, which is how the connection was made.

There is no biosynthetic route, because there is no biosynthesis. Preparative access is semi-synthetic in the loosest sense: ammoniacal extraction of rosemary leaf in air, followed by chromatography ^{2Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →}. The only sustained chemistry on it has been derivatisation — the Genoa and Milan groups have made N-substituted, O,O-dimethylated, N-aminoacyl and sulfurated-hybrid derivatives across five decades ^{2,6Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →Reference 6Boido A et al. · 1994Rosmaricine derivatives as potential antitumoral agentsView study →}. Patents: not yet researched.

Toxicity & Safety

Not characterised, and there is no exposure to characterise. No acute-toxicity study, no genotoxicity panel, no reproductive or developmental data, no repeat-dose work. The unknown grading is literal: nothing has been looked for.

The one practical point is reassuring in an unusual way. Because rosmaricine is an artefact of an ammoniacal laboratory procedure, no rosemary product delivers it — not the culinary herb, not a tea or tincture, not the essential oil, and not the food-grade rosemary diterpene extract used as an antioxidant. Whatever the safety questions about rosemary are, this molecule is not among them.

Safety at a Glance

Toxicity unknown

No human exposure. Rosmaricine is not a constituent of rosemary food, tea, tincture, oil or extract ^{1,2,3Reference 1Wenkert E et al. · 1965Chemical artifacts from the family LabiataeView study →Reference 2Gabriele E et al. · 2017New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factorsView study →Reference 3Hanson JR · 2016Rosemary, the beneficial chemistry of a garden herbView study →}.
No adverse effect has been reported. It has also never been looked for — different statements.
No toxicology of any kind exists. No acute, repeat-dose, genotoxicity or reproductive study has been published on the isolated compound.
Two of the papers naming it have been retracted. Both 2011 PLOS Computational Biology docking papers were withdrawn in June 2026 ^{9,10Reference 9Yang SC et al. · 2011ReviewRETRACTED: Identification of potent EGFR inhibitors from TCM Database@TaiwanView study →Reference 10Chang SS et al. · 2011RETRACTED: Two birds with one stone? Possible dual-targeting H1N1 inhibitors from traditional Chinese medicineView study →}.
Do not read carnosol’s or carnosic acid’s safety record onto it. They are related skeletons, not the same molecule, and only they are actually in the plant ^{12Reference 12Birtić S et al. · 2015Carnosic acidView study →}.
Treat any commercial “rosmaricine” as a research chemical. It is a laboratory scaffold with a single in-vitro dataset behind it, not a supplement ingredient.

Dosage

There is none. Rosmaricine has never been given to a human being, and there is no animal dosing study to extrapolate from. Every published number is an in-vitro concentration at a purified protein or a cultured cell line.

Nor is there an incidental dietary exposure to describe, which is the unusual part: unlike almost every other molecule here, you cannot obtain rosmaricine by consuming the plant it is named after. Nothing here is a recommendation, and nothing on it should be transferred to rosemary.

References

Wenkert E, Fuchs A, McChesney JD (1965). Chemical artifacts from the family Labiatae. The Journal of Organic Chemistry 30:2931–2934. https://doi.org/10.1021/jo01020a012
Gabriele E, Brambilla D, Ricci C, Regazzoni L, Taguchi K, Ferri N, Asai A, Sparatore A (2017). New sulfurated derivatives of cinnamic acids and rosmaricine as inhibitors of STAT3 and NF-κB transcription factors. Journal of Enzyme Inhibition and Medicinal Chemistry. https://pubmed.ncbi.nlm.nih.gov/28738705/
Hanson JR (2016). Rosemary, the beneficial chemistry of a garden herb. Science Progress 99:83–91. https://pubmed.ncbi.nlm.nih.gov/27120816/
Yakhontova LD, Kuzovkov AD (1967). The nature of rosmaricine from Rosmarinus officinalis. III. Chemistry of Natural Compounds 3:118. https://doi.org/10.1007/BF00567185
Yakhontova LD, Sheichenko VI, Tolkachev ON (1971). A study of Rosmarinus officinalis IV. Isorosmaricine. Chemistry of Natural Compounds 7:396–398. https://doi.org/10.1007/BF00564722
Boido A, Savelli F, Sparatore F (1994). Rosmaricine derivatives as potential antitumoral agents. Farmaco 49:111–114. https://pubmed.ncbi.nlm.nih.gov/8003178/
Legchaev VYa (1965). [On the mechanism of action of rosmaricin on the cardiovascular system]. Farmakologiia i Toksikologiia 28:170–172. Russian. https://pubmed.ncbi.nlm.nih.gov/5849197/
Grishaenkov VM, Ivanov MI, Legchaev VYa (1967). [The effect of rosmaricin on mice in acute radiation sickness]. Meditsinskaia Radiologiia 12:69–70. Russian. https://pubmed.ncbi.nlm.nih.gov/5621330/
Yang SC, Chang SS, Chen HY, Chen CY (2011). RETRACTED: Identification of potent EGFR inhibitors from TCM Database@Taiwan. PLoS Computational Biology 7:e1002189. https://pubmed.ncbi.nlm.nih.gov/22022246/ — retracted 9 June 2026 over compromised peer review; do not cite as evidence.
Chang SS, Huang HJ, Chen CY (2011). RETRACTED: Two birds with one stone? Possible dual-targeting H1N1 inhibitors from traditional Chinese medicine. PLoS Computational Biology 7:e1002315. https://pubmed.ncbi.nlm.nih.gov/22215997/ — retracted 9 June 2026; do not cite as evidence.
Chang TT, Sun MF, Chen HY, Tsai FJ, Fisher M, Lin JG, Chen CY (2011). Screening from the world’s largest TCM database against H1N1 virus. Journal of Biomolecular Structure and Dynamics 28:773–786. https://pubmed.ncbi.nlm.nih.gov/21294588/
Birtić S, Dussort P, Pierre FX, Bily AC, Roller M (2015). Carnosic acid. Phytochemistry 115:9–19. https://pubmed.ncbi.nlm.nih.gov/25639596/
Brieskorn CH, Fuchs A, Bredenberg JB, McChesney JD, Wenkert E (1964). The structure of carnosol. The Journal of Organic Chemistry 29:2293–2298. https://doi.org/10.1021/jo01031a044
Calabrese V, Scapagnini G, Catalano C, Dinotta F, Geraci D, Morganti P (2000). Biochemical studies of a natural antioxidant isolated from rosemary and its application in cosmetic dermatology. International Journal of Tissue Reactions 22:5–13. https://pubmed.ncbi.nlm.nih.gov/10937349/