Cannabis builds one precursor, CBGA, then three competing enzymes split it: THCA synthase, CBDA synthase and CBCA synthase.
CBD is what CBDA becomes when heat and time remove a carbon dioxide group.
CBG is what remains of CBGA once the synthases have finished. This is why CBG is scarce in most cultivars and why CBG raw material costs more. It is a botanical constraint, not a potency signal.
CBN is not on that diagram at all. It has no synthase. It forms when THC sits in contact with oxygen, heat and light and degrades. High CBN indicates old or heat-treated material, and concentrated CBN products are made by deliberately degrading THC.
That last point is why these three are not really peers. Two are plant products. One is a decay product.
CBD does little at cannabinoid receptors. Two of its genuinely potent actions are blocking the ENT1 adenosine transporter below 250 nanomolar and agonising TRPA1 at around 90 nanomolar. It also acts at TRPA1 at 0.09 micromolar, and at TRPV1, 5-HT1A and GPR55. A 2015 systematic review of more than 65 reported CBD targets concluded most are engaged only at concentrations people do not achieve.
CBG has the most potent single measurement of the three: alpha-2 adrenoceptor agonism at an EC50 of 0.2 nanomolar in mouse brain membranes, reported in 2010. The same study found the effect substantially weaker in isolated tissue, leaving the physiological relevance unresolved. CBG is also a 5-HT1A antagonist, the opposite direction to CBD at the same receptor.
CBN is a weak partial agonist at CB1, roughly 29 percent maximum effect against THC’s 72 percent. Its metabolite 11-hydroxy-CBN reaches equivalent brain concentrations and behaves like THC at CB1. If CBN produces central effects, that metabolite is the likely route.
Note that the compound with the most marketing behind it, CBN, has the weakest supporting result.
Neither CBG nor CBN has published human pharmacokinetics. All dosing guidance for either is extrapolation.
No health claim for CBD, CBG, CBN or any cannabinoid is authorised under Regulation (EC) No 1924/2006. Article 10(3) additionally excludes general wellbeing language, because such references are lawful only alongside an authorised claim and none exists. Regulation (EU) No 1169/2011 prohibits attributing disease-related properties to food.
This is why our EU product range is not organised by benefit. Category names such as “Sleep”, “Focus” or “Calm” are treated as implied claims by enforcement authorities across member states, and a CBD, CBG or CBN product sold lawfully in the EU cannot be positioned that way.
All three fall within the EU Novel Food Catalogue entry covering cannabinoid-containing extracts of Cannabis sativa L. None is authorised. Placing an unauthorised novel food on the market is unlawful in every member state.
EFSA’s February 2026 provisional safe level of 0.0275 mg per kilogram of body weight per day, roughly 2 mg for a 70 kg adult, applies to CBD isolate of at least 98 percent purity. Safety could not be established for anyone under 25, for pregnant or breastfeeding women, or for people on medication. No equivalent assessment exists for CBG or CBN.
CBN carries an additional question, because concentrated CBN is generally produced by converting THC rather than by extraction, which raises issues under national narcotics rules as well as novel food law.
National positions diverge sharply. Germany’s BVL considers CBD food supplements not marketable. Denmark requires authorisation and publishes THC action limits. Sweden applies a narcotics overlay to THC-containing extracts. Spain’s enforcement has channelled the category into external-use products.
All three appear on every Endoca batch certificate, alongside CBDA, CBC, THC and THCA, tied to the batch code on the bottle. We publish the CBN figure even when it is low, because low CBN indicates that material was extracted and stored well.
Under EU rules, that measured laboratory data is close to the entirety of what we are lawfully able to tell you. It is also, in our view, the most useful thing.
They are commonly combined and no interaction problem is documented. No evidence shows the combination outperforms any of them alone.
They are not on a common scale. CBG has the most potent single receptor measurement. CBD has the most human evidence. CBN is the only one weakly intoxicating at very high doses.
The plant converts nearly all its CBGA into other cannabinoids.
Usually a small amount, rising as the product ages.
Because no cannabinoid health claim is authorised in the EU, and general wellbeing language is excluded too.
Cascio MG et al. Br J Pharmacol 2010;159:129-141.
Cuttler C et al. Sci Rep 2024;14:16163.
Lavender I et al. J Sleep Res 2026;35:e70284.
Arnold JC et al. Neuropsychopharmacology 2025;50:586-595.
Ibeas Bih C et al. Neurotherapeutics 2015;12:699-730.
Moore RA et al. J Pain 2024;25:833-842.
Finlay DB et al. Front Pharmacol 2020;11:359.
EFSA Journal 2026;24:9862.
Last reviewed 1 September 2026.
Educational information about published research and European regulation. Not a health claim.