Researchers have engineered a tobacco relative to produce five distinct psychedelic compounds simultaneously, pulling genetic instructions from plants, fungi, and animals into a single organism. The species used, Nicotiana benthamiana, generated DMT, psilocin, psilocybin, bufotenine, and 5-MeO-DMT through reconstructed biosynthetic pathways driven by a combined genetic construct. The result represents a new benchmark for multi-kingdom metabolic engineering and raises immediate questions about yield, scalability, and regulation.
What is verified so far
The central finding is well documented in a peer-reviewed study published in Science Advances. The team reconstructed full biosynthetic pathways in N. benthamiana that span three biological kingdoms. DMT and bufotenine are found naturally in certain plants and animals, psilocybin and psilocin originate in fungi, and 5-MeO-DMT occurs in both plants and the venom of cane toads. Producing all five in a single tobacco-family plant required stitching together genes that evolved independently across these kingdoms.
A key enabler was a highly active and promiscuous N-methyltransferase enzyme sourced from the cane toad Rhinella marina. As described in a commentary indexed at Trends in Biochemical Sciences, this enzyme efficiently converts primary indolethylamines into tertiary psychedelic amines. Because all five target compounds share a common indole backbone, the toad enzyme can process multiple substrates in the same cellular environment, which is why a single construct can yield several products at once rather than requiring separate engineered lines for each molecule.
The psilocybin branch of the pathway also has strong structural biology support. Research published in Nature Communications characterized the three-dimensional structures of the mushroom enzymes PsiD, PsiK, and PsiM, which catalyze sequential steps in psilocybin production. Those structural insights help explain why the fungal pathway functions when transplanted into a plant host: the enzymes fold and operate correctly even outside their native organism.
Tobacco species have a proven track record as biofactories for complex molecules. A landmark study in Nature Biotechnology demonstrated that five genes delivered on a single vector could drive production of the anti-malarial drug artemisinin in tobacco, with product quantification confirming measurable yields. That precedent established N. benthamiana as a workhorse for multi-gene metabolic engineering, which set the stage for the psychedelic work.
Separate research published in Frontiers in Plant Science showed that N. benthamiana can serve as a heterologous host for reconstructing and diversifying complex alkaloid pathways, including precursor-directed biosynthesis and the generation of analogs. That study demonstrated the plant’s capacity to produce not just known compounds but also new-to-nature derivatives, suggesting the psychedelic platform could eventually be expanded to create modified versions of these molecules.
What remains uncertain
The most significant gap in the current evidence is the absence of publicly detailed yield data. While the Science Advances paper confirms that all five compounds were produced, exact titers and purification efficiencies for each molecule have not been widely reported in accessible form. Without those numbers, it is difficult to judge whether this approach can compete with chemical synthesis or microbial fermentation on cost or throughput.
A second open question is whether the results come from transient expression or stable genetic transformation. A technical review in aBIOTECH surveying multiple case studies of metabolic engineering in N. benthamiana highlights that practical constraints differ sharply between these two approaches. Transient expression, which uses temporary gene delivery via bacterial infiltration, produces compounds quickly but does not create heritable plant lines. Stable transformation takes longer to develop but could support continuous, seed-based production. The review also flags pathway bottlenecks and expression system limitations as recurring challenges across N. benthamiana projects.
No official institutional statements or direct researcher interviews addressing ethical concerns or regulatory pathways for psychedelic plant production appear in the available primary literature. Psilocybin, DMT, bufotenine, and 5-MeO-DMT are all Schedule I controlled substances in the United States, and psilocin is similarly restricted. Any commercial or clinical application of plant-produced psychedelics would need to clear significant legal barriers that the published science does not address.
There is also no primary data available on whether the five compounds interact or compete for shared precursors inside the plant cell. If the toad N-methyltransferase processes multiple substrates simultaneously, the relative ratios of the five products may shift unpredictably depending on precursor availability, enzyme kinetics, and plant growth conditions. This metabolic competition could be a practical obstacle to producing any single compound at high purity.
How to read the evidence
The strongest evidence here comes from the peer-reviewed primary literature. The Science Advances study is the direct source for the headline claim and has undergone formal review. The Nature Communications structural biology work and the Trends in Biochemical Sciences commentary on the toad enzyme provide independent mechanistic support, confirming that the biological tools used in this system are well characterized.
The Nature Biotechnology artemisinin paper and the Frontiers in Plant Science alkaloid diversification study serve as contextual precedents rather than direct evidence for the psychedelic result. They establish that multi-gene engineering in tobacco works and that the platform can generate novel compounds. They do not themselves demonstrate psychedelic production. Readers should treat them as proof of concept for the chassis, rather than for the specific payload.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
