Cannabinoid Receptor Signalling

Roughly forty amino acid residues deep inside the plasma membrane, a G protein-coupled receptor waits. A lipid-soluble molecule drifts toward it, slots into the binding pocket, and something shifts , conformationally, chemically, functionally. That's the moment I find genuinely fascinating about cannabinoid biology: the signal doesn't travel through the bloodstream like a letter in the post. It's a local conversation, conducted in a language of structural change.

This piece is about the mechanics of that conversation. Not about what it might mean clinically , that's a different, heavily regulated discussion , but about how cannabinoid receptors actually signal at the molecular level.

The receptor landscape: CB1, CB2, and the wider cast

The two classical cannabinoid receptors are CB1 and CB2, both members of the Gi/o-coupled GPCR superfamily. If you want the structural detail on either, our CB1 receptor and CB2 receptor glossary entries cover the basics. The short version here: CB1 is expressed densely in the central nervous system, particularly in presynaptic terminals of the basal ganglia, hippocampus and cerebellum. CB2 sits predominantly in immune-associated tissues; spleen, tonsils, microglia, though that distribution is more fluid than early textbooks suggested.

Both receptors are seven-transmembrane proteins. Ligand binding at the orthosteric site (or, increasingly, at allosteric sites) triggers a conformational rearrangement in the intracellular loops that allows coupling to Gi/o heterotrimeric proteins. The Gα subunit dissociates, inhibiting adenylyl cyclase and reducing intracellular cyclic AMP. Gβγ subunits, meanwhile, go to work on inwardly rectifying potassium channels (opening them, hyperpolarising the cell) and voltage-gated calcium channels (closing them, reducing neurotransmitter release). That combination, less cAMP, more K⁺ efflux, less Ca²⁺ influx; adds up to a broadly inhibitory tone on neuronal firing.

But "Gi/o coupling" is not the full story. Long-term receptor stimulation also recruits β-arrestins, which both desensitise the receptor (uncoupling it from G proteins by physically blocking the intracellular loops) and initiate their own downstream signalling through ERK1/2 and other MAPK pathways. This β-arrestin pathway is increasingly understood as distinct in its functional consequences from G-protein signalling, a concept the pharmacology literature calls "biased agonism." Different ligands can push the same receptor toward different downstream outcomes, which is one reason cannabinoid pharmacology is far more complex than a simple on/off switch.

Endogenous ligands and the retrograde signal

The endocannabinoid system doesn't work like most neurotransmitter systems. Classical neurotransmitters are stored in presynaptic vesicles and released into the synapse to act on the postsynaptic cell. Endocannabinoids do the reverse: they're synthesised on-demand in the postsynaptic neuron, released, and travel backwards across the synapse to act on presynaptic CB1 receptors. This retrograde signalling is how endocannabinoids function as a feedback brake on neurotransmitter release.

Anandamide (N-arachidonoylethanolamine, AEA) and 2-arachidonoylglycerol (2-AG) are the two best-characterised endogenous ligands. Anandamide is synthesised primarily via NAPE-PLD-mediated cleavage of N-arachidonoyl phosphatidylethanolamine; 2-AG via diacylglycerol lipase (DAGL) from diacylglycerol. Both are produced from arachidonic acid-containing membrane phospholipids and are hydrolysed rapidly after release, anandamide by fatty acid amide hydrolase (FAAH), 2-AG mainly by monoacylglycerol lipase (MAGL). That rapid inactivation keeps endocannabinoid signalling spatially and temporally tight.

Anandamide is sometimes described as a partial agonist at CB1; 2-AG as a full agonist. The distinction matters for understanding intrinsic efficacy. A partial agonist occupies the receptor fully but produces a submaximal response even at saturation; relevant when thinking about how phytocannabinoids, which often have their own intrinsic efficacy profiles, might interact with endogenous tone.

Phytocannabinoids at the receptor: THC, CBD and the rest

Δ9-THC binds to both CB1 and CB2 as a partial agonist with relatively high affinity. Its psychoactive effects are attributed primarily to CB1 agonism in the CNS, particularly in circuits involving dopamine release in the mesolimbic pathway and modulation of GABA/glutamate balance. At the CB1 orthosteric site, THC's binding mode has been resolved by cryo-EM: the bicyclic terpenoid ring sits in the hydrophobic groove formed by transmembrane helices 3, 5, 6 and 7, with key interactions at residues Phe200 and Ser383 (in human CB1 numbering).

CBD is considerably more complicated. It has low affinity for CB1 and CB2 orthosteric sites under standard assay conditions, yet produces measurable effects through several other mechanisms: it acts as a negative allosteric modulator (NAM) at CB1, altering how other ligands bind and signal at the orthosteric site without activating it directly. It also has documented activity at TRPV1 channels, GPR55, GPR18, 5-HT1A receptors and as a FAAH inhibitor. The honest scientific position here is that CBD's receptor pharmacology is still being worked out, anyone who reduces it to "non-psychoactive CB1/CB2 ligand" is leaving a lot out.

Other phytocannabinoids are less well characterised but increasingly studied. CBG shows partial agonism at CB1 and CB2 with lower affinity than THC. THCV is particularly interesting: at low concentrations it behaves as a CB1 antagonist/inverse agonist, while at higher concentrations it can act as a partial agonist, a dose-dependent switch in functional profile that you don't see often in small-molecule pharmacology. Delta-8-THC has a similar binding geometry to Δ9-THC but with modestly reduced CB1 affinity, thought to relate to a single double-bond positional difference affecting the ligand's fit in the binding pocket.

Beyond CB1 and CB2: the extended receptor family

The "classical" two-receptor model has expanded. GPR55 was proposed as a third cannabinoid receptor; it responds to certain cannabinoids and some endogenous lipids, signals primarily through G12/13 proteins and RhoA rather than Gi, and is expressed in the dorsal root ganglia and gastrointestinal tract. Whether it belongs formally in the cannabinoid receptor family is still debated; the International Union of Basic and Clinical Pharmacology (IUPHAR) hasn't reclassified it as CB3.

TRPV1 (transient receptor potential vanilloid 1) deserves mention. It's a non-selective cation channel gated by heat, capsaicin, and, notably, anandamide at higher concentrations. TRPV1 activation by anandamide produces an effect often opposite in tone to CB1 activation, which may partly explain why anandamide's in vivo profile doesn't map cleanly onto CB1 agonism alone. The endocannabinoid system, viewed this way, is better described as a lipid signalling network than a two-receptor switch.

I'll admit I got the "ECS equals CB1 plus CB2" framing from my undergraduate biochemistry course, and it took a while reading the primary literature to appreciate how much was missing from that picture. The IUPHAR/BPS Guide to Pharmacology is genuinely useful for untangling the receptor classification mess if you want to go further.

Receptor regulation: tolerance, internalisation and sensitisation

Sustained receptor activation doesn't maintain a constant signal. CB1 receptors undergo rapid desensitisation via GRK (G protein-coupled receptor kinase)-mediated phosphorylation of the C-terminal tail, followed by β-arrestin recruitment and receptor internalisation into endosomes. This internalisation removes receptors from the cell surface and reduces signal amplitude; the cellular basis of acute tolerance.

Internalised receptors can be recycled back to the membrane (resensitisation) or trafficked to lysosomes for degradation (downregulation). Chronic high-level agonist exposure biases toward degradation, reducing total receptor protein. This molecular sequence, phosphorylation, arrestin binding, endocytosis, lysosomal degradation, is the mechanistic underpinning of why receptor availability changes with exposure history. It's also why the receptor pharmacology of someone who has had prolonged cannabinoid exposure may differ substantially from a naïve state, even before considering changes in gene expression.

From a research standpoint, the kinetics of CB1 resensitisation after agonist withdrawal have been studied in rodent models and cultured neurons; in vivo data in humans is more limited. This is worth keeping in mind when reading any claim about receptor dynamics in clinical populations; the cellular picture and the whole-organism picture don't always translate directly.

A note on signalling context

Receptor signalling never happens in isolation. The local cellular environment, which G proteins are expressed, what scaffolding proteins are present, the membrane lipid composition, receptor heterodimer formation, all shape how a given ligand's signal is decoded. CB1 receptors form heterodimers with D2 dopamine receptors and μ-opioid receptors, among others, and the signalling properties of these complexes differ from homomers. This context-dependence is part of why cannabinoid pharmacology in a cultured cell line doesn't always predict what happens in an intact brain circuit.

It's also why the entourage effect hypothesis; the idea that the full complement of cannabinoids and terpenes in a plant extract interact at multiple receptor and non-receptor targets to produce effects distinct from any single isolated component, has mechanistic plausibility even if its clinical evidence base remains thin. Mechanistic plausibility and clinical evidence are not the same thing, and I'd argue the literature conflates them more often than it should.

Sources

• Cannabinoid Receptors and the Endocannabinoid System: Signaling and Function in the Central Nervous System, NCBI/PMC
• Cannabinoid Receptors; IUPHAR/BPS Guide to Pharmacology
• Molecular Mechanisms of Cannabinoid Tolerance, NCBI/PMC
• Guidance on the Use of Medicinal Cannabis in Australia, TGA

, Elise Tran, Science writer; cannabinoids & the ECS

]]>