Bryo-delic! Diverse bibenzyl cannabinoids in the liverwort Radula marginata

Phytocannabinoids are the active ingredients in cannabis extracts, with tetrahydrocannabinol (THC) and cannabidiol (CBD) metabolites representing the two major subclasses with medicinal and economic value (Reekie et al., 2017). Early investigations of cannabinoid compounds prompted the discovery of human cannabinoid receptors and the endocannabinoid system, which has diverse roles in immunity and the nervous system (Ligresti et al., 2016). Despite their prominent association with the plant Cannabis sativa, cannabinoids have been isolated from a wider evolutionary spectrum of organisms, including other clades of flowering plants (Helichrysum umbraculigerum, Glycyrrhiza foetida, Amorpha fruticosa, Rhododendron dauricum, and Rhododendron anthopogonoides), nonflowering plants belonging to the Radula genus of liverworts (Radula marginata and R. perrottetii), and even fungi (Albatrellus and Cylindrocarpon olidum) (Gülck & Møller, 2020). Among these cannabinoid producers, Radula liverworts are notable for the inverted stereoconfigurations of their cannabinoids relative to other organisms.

The cannabinoid-like molecule cis-perrottetine (cis-PET) was first identified in R. perottetii liverworts in Japan and is a structural analog of (−)-Δ9-trans-tetrahydrocannabinol (Δ9-trans-THC) (Toyota et al., 1994). Since then, cannabinoid-like molecules have been identified in R. laxirameae growing in Costa Rica (Cullmann & Becker, 1999) and R. marginata growing in New Zealand (Toyota et al., 2002). Unlike the common model liverwort Marchantia polymorpha, which develops a thallus plant body, Radula liverworts display a ‘leafy’ morphology where small leaf-like appendages and root hair-like rhizoids emanate from a central stem-like structure (Fig. 1). They also grow epiphytically on the bark of forest trees and are notably slow-growing compared with model liverworts. Pharmacological studies of synthetic liverwort-type cis-PET have shown that this compound has a psychoactive role in mice, where it penetrates the blood brain barrier and stimulates cannabinoid receptor (CB1)-dependent processes with potentially fewer side effects compared with THC (Chicca et al., 2018). While this validates Radula cis-PET as a highly promising and pharmacologically relevant metabolite, the chemical and genetic diversity of bibenzyl cannabinoid-like metabolites produced naturally in Radula remained largely unexplored.

Cannabis cultivars generally present as one of three distinct categories of ‘chemotype’ capable of producing high levels of THC, CBD, or a mixture accumulating lower levels of each class (Jin et al., 2021). To begin to understand whether Radula liverworts exhibit a similar chemotype range, Andre et al. interrogated natural populations of R. marginata liverworts sampled across three disparate locations during several seasons in New Zealand. By collecting and analytically quantifying cannabinoid-like bibenzyl compounds in over 75 samples, the authors confirmed that wild liverworts produce the THC structural analog cis-PET. They also found accumulation of perrottetinediol (PTD, a structural analog of CBD) and bibenzyl-4-geranyl (BB4G, a structural analog of the THC/CBD precursor cannabigerol) among other cannabinoid-like compounds and their chemical derivatives (Fig. 1). Importantly, they observed that individual plants fell into definable chemotype categories displaying PET dominance, PTD dominance, or an intermediate mixture of the two. Remarkably, this mirrored the chemotype classes observed in Cannabis, where THC/CBD ratios are genetically defined by the presence, absence, and heterozygosity of functional CBDAS (CBD acid synthase) and THCAS (THC acid synthase) alleles of enzymes that both compete for the same precursor substrate CBGA (cannabigerolic acid) (Ren et al., 2021). While a full genetic understanding of PET vs PTD dominance remains to be resolved, the authors speculate that a comparable but independently evolved genetic framework underpins cannabinoid chemotypes in R. marginata.

Since varying PET vs PTD levels could be explained by seasonal variation in abiotic factors (light, climate), the authors tested whether R. marginata propagated in controlled and/or axenic conditions maintained their chemotype status. Plants cultivated indoors showed little difference in cannabinoid levels for the first 4 months when compared to the wild plants from which they were sourced, supporting the idea chemotypes are indeed stable and therefore likely to be genetically encoded. Indeed, the same plants cultivated for over 1 year in artificial lighting, supplemented with far-red light, eventually produced higher levels of PET or PTD, depending on their original chemotype. The authors also explored whether axenic cultures of R. marginata could be generated and used for chemical analyses. While sterile plant cultures propagated in tissue culture conditions were developmentally impacted and slower growing overall, they generally produced higher levels (c. twofold) of cannabinoids than their wild-grown counterparts. This again supports the idea that R. marginata chemotypes are genetically encoded, demonstrating promise for future efforts to cultivate these liverworts for cannabinoid production.

The chemical and ecological insights provided by Andre et al. significantly expand our understanding of cannabinoid structural diversity in a divergent land plant lineage. While it is clear that some Radula liverwort species have independently evolved the ability to make cannabinoid-like compounds, further research is required to clarify the genetic and biochemical pathways underpinning their synthesis. Importantly, future efforts to generate chromosome-level genome resources should include a diverse range of individual accessions that sample both geographically distinct ‘ecotypes’ and chemically diverse ‘chemotypes’. In addition, future explorations into the utility of R. marginata cannabinoid-like bibenzyls for liverwort fitness may also reveal the ecological context underlying their evolution. Initial hypotheses for this include a potential role in plant–herbivore interactions, given that these compounds accumulate within liverwort oil bodies that are known to repel insects in the model liverwort Marchantia (Kanazawa et al., 2020; Romani et al., 2020). In any case, the research and resources generated by Andre et al. represent an important first step toward such endeavors.

While it would be easy to focus solely on their fascinating scientific discoveries, the authors also take care to highlight the ethical and moral considerations of their work, which was performed with permission and assistance from indigenous authorities on tribal estates. Like all species endemic to New Zealand, R. marginata are considered natural treasures to Māori who are guardians of their conservation. The authors note and rightfully remind readers that future exploitation of R. marginata chemistry must be performed under international frameworks that recognize Indigenous peoples' assertions of authority over natural resources. This study presents a successful example of such collaborations.

The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.