Endocannabinoids and the Nervous System (overview)

Your nervous system already makes its own cannabinoids

Most people first encounter the word "cannabinoid" on a product label or in a news headline about cannabis. Which is a shame, because the more interesting story starts inside the body, specifically at the synapse. The human nervous system synthesises its own cannabinoid-like molecules, uses dedicated receptors to read them, and then rapidly breaks them down. All of that was unknown until the early 1990s, and we're still working out what it means.

This piece is an overview of the endocannabinoid system (ECS) and how it operates in nervous tissue. It covers the two main endogenous ligands, the receptor landscape, and the basic signalling logic. Where something is hypothesis rather than settled science, I'll say so.

If you'd like a gentler starting point first, the endocannabinoid system explained for non-scientists covers the same ground with less jargon. For the history of how researchers got here, see the discovery of the endocannabinoid system.

The two principal endocannabinoids

Two molecules do most of the heavy lifting: anandamide (arachidonoylethanolamide, or AEA) and 2-arachidonoylglycerol (2-AG). Both are derived from arachidonic acid, an omega-6 fatty acid embedded in cell membranes. Neither is stored in vesicles the way classical neurotransmitters are. Instead, they're synthesised on demand: built from membrane precursors when the neuron receives a trigger, released immediately, and then degraded within seconds.

That last point deserves a moment. A conventional neurotransmitter such as glutamate is packaged in advance and released in a burst when the terminal fires. Endocannabinoids are lipids, and lipids dissolve in membranes rather than waiting in tidy packages. A cell that wants to send one has to make it from the membrane itself, which ties the signal tightly to what that particular neuron has just been doing.

Anandamide was characterised first, in 1992, by Raphael Mechoulam's group at the Hebrew University of Jerusalem, the same lab that had isolated THC almost thirty years earlier. Its name comes from the Sanskrit word ānanda, meaning bliss. A poetic choice, though whether that reflects a genuine function or just the naming enthusiasm of a research team is still debated. 2-AG was identified shortly after and turns out to be present in brain tissue at concentrations roughly 1,000 times higher than anandamide, which has shifted a lot of thinking about which of the two is the dominant endogenous ligand at CB1.

Both belong to the broader class called N-acylethanolamines in the case of anandamide, a family of fatty-acid-derived signalling lipids. Anandamide is a partial agonist at CB1; 2-AG is a full agonist at both CB1 and CB2. That distinction matters functionally and is worth keeping in mind as the field produces new research.

Agonists, ligands and other terms worth pinning down

Much of the confusion around this topic is vocabulary. An endogenous ligand is simply a molecule the body makes itself that binds to a receptor. "Agonist" describes a ligand that activates the receptor it binds. A full agonist produces the maximum response the receptor can give, while a partial agonist produces a lower ceiling even when every receptor is occupied. Binding strength (affinity) and the size of the response (efficacy) are separate properties, and mixing them up is one of the most common mistakes in popular writing.

The prefix matters too. "Endo" means made inside the body, "phyto" means from plants, and "synthetic" means made in a laboratory. A plant compound can act on the same receptors as a body-made one without being the same molecule, and the three groups are compared in endogenous, phytogenic and synthetic cannabinoids. When a headline says "cannabinoid", it is worth asking which of the three it means.

CB1 and CB2: where they sit in the nervous system

The two canonical cannabinoid receptors are G protein-coupled receptors (GPCRs); large transmembrane proteins that translate an extracellular chemical signal into intracellular change. They're related but not identical in their distribution or downstream effects. The general cannabinoid receptor entry gives the shared features, and cannabinoid receptor signalling follows what happens inside the cell after binding.

CB1 receptors are among the most abundant GPCRs in the mammalian central nervous system. They're found at high density in the basal ganglia, cerebellum, hippocampus, and cortex, with lower but meaningful expression in the peripheral nervous system. Functionally, CB1 activation generally suppresses neurotransmitter release: when CB1 is activated on a presynaptic terminal, it tends to reduce the probability that the next action potential will release glutamate or GABA into the synapse. This is the basis of retrograde signalling (more on that below).

CB2 receptors were initially described as peripheral receptors, found predominantly on immune cells, with lower CNS expression. More recent work has identified CB2 in microglia (the brain's resident immune cells) and in some neurons, though the extent and significance of neuronal CB2 expression is still being mapped. It's fair to call the precise CNS role of CB2 an active area of investigation rather than a settled question.

There are also non-CB1/CB2 targets. Anandamide is a known agonist at the TRPV1 receptor; a cation channel more commonly associated with heat and pain signalling. This "promiscuity" of endocannabinoid ligands is part of why the system is so hard to summarise neatly.

Retrograde signalling: the logic runs backwards

Here's the bit I find genuinely elegant. Classical synaptic signalling runs forward: a presynaptic neuron releases a transmitter, it crosses the cleft, hits a postsynaptic receptor, and the postsynaptic cell responds. The ECS in the CNS largely works in reverse.

When a postsynaptic neuron fires strongly, or when its metabotropic receptors are strongly activated, it synthesises endocannabinoids from its own membrane lipids and releases them back across the synapse toward the presynaptic terminal. The presynaptic CB1 receptors detect this, and the result is a reduction in further neurotransmitter release. It's a feedback brake. The postsynaptic cell is essentially telling the presynaptic one: "enough".

This retrograde mechanism has been studied extensively in the context of two phenomena: depolarisation-induced suppression of inhibition (DSI) and depolarisation-induced suppression of excitation (DSE). Both involve endocannabinoid release from the postsynaptic cell suppressing the upstream terminal. The research here is well-replicated; this is among the more solid foundations of ECS neuroscience. What remains hypothesis is the full functional significance of this modulation across different circuits and states.

The molecular sequence is satisfying in its own right: lipid synthesis at the membrane, diffusion across a hydrophobic cleft, GPCR activation, and reduced calcium influx at the presynaptic terminal. It upends the tidy assumption that synapses are one-way streets. Because the signal is local and brief, it adjusts one connection at a time rather than flooding a whole region, which is part of why researchers describe it as fine-tuning rather than switching activity on or off.

Synthesis, degradation, and the "on-demand" principle

Because endocannabinoids aren't stored, the rate of their action is largely controlled by synthesis and degradation enzymes. Anandamide is primarily degraded by fatty acid amide hydrolase (FAAH). 2-AG is primarily degraded by monoacylglycerol lipase (MAGL), with minor contributions from other hydrolases.

Both enzymes have attracted interest as research targets precisely because modulating degradation is a way to influence endocannabinoid tone without directly agonising or antagonising receptors. Understanding this degradation pathway is also relevant to why the signalling is so brief and localised; FAAH and MAGL act quickly, limiting the temporal and spatial reach of the signal.

The concept of endocannabinoid tone, the baseline level of ECS activity in a given tissue at a given time, is worth knowing. It is a useful shorthand, but it is not something that can simply be read off a single measurement, because levels change quickly and differ between tissues. Some researchers have proposed the hypothesis of clinical endocannabinoid deficiency, which suggests that insufficient ECS tone may underlie certain conditions. This is a hypothesis with some theoretical grounding but limited direct clinical evidence to date; it should be understood as speculative rather than established.

Repeated receptor activation is a related question, and the way receptors adapt over time is covered in cannabinoid tolerance and receptor downregulation.

Beyond the two receptors: the endocannabinoidome

The term endocannabinoidome has come into use to describe the expanded signalling network: not just AEA and 2-AG at CB1/CB2, but a broader family of lipid mediators acting at a range of receptors; including GPR55, GPR18, GPR119, TRPV channels, and peroxisome proliferator-activated receptors (PPARs). This framing captures the complexity better than "ECS" alone, though it also makes the system harder to study cleanly.

For context: tetrahydrocannabinol (THC), the principal psychoactive constituent of cannabis, produces its effects largely by acting as a partial agonist at CB1. It essentially mimics anandamide at the receptor level but is far more metabolically stable (anandamide is degraded by FAAH; THC is not). Cannabidiol (CBD) has a more complex pharmacological profile, with relatively low affinity for CB1/CB2 but meaningful activity at TRPV1, GPR55, and serotonin 5-HT1A receptors, among others. Its interactions with the ECS are indirect and still being characterised.

In Australia, the regulator treats cannabidiol products as a matter of therapeutic goods regulation, and the Therapeutic Goods Administration publishes consumer information on the subject (linked in the sources below). Pharmacology in a paper and the rules around a product on a shelf are different questions, and this piece deals only with the first.

Where the research is genuinely strong, and where it isn't

Honestly, one frustration I have with how the ECS gets written about, even in otherwise careful sources, is that the strength of evidence across different claims is flattened. The retrograde signalling mechanism at CB1? Very well-supported. The specific role of CB2 in neuroinflammation? Promising in animal models, but translating to human systems remains difficult. The endocannabinoidome as a therapeutic target? Conceptually interesting, early days. These distinctions matter.

A practical way to read any claim is to ask what kind of evidence sits behind it. Work in isolated cells or tissue slices can show that a mechanism exists. Animal studies show whether it plays out in a living nervous system, though species differences are real. Human evidence is the hardest to gather and the easiest to over-read. A sentence that moves from "receptors are present in this tissue" to "this explains a human outcome" has skipped several steps, and I'd argue a careful reader should notice each skipped step.

The ECS is genuinely involved in modulating nervous system activity. The receptor biology is real, the ligands are well-characterised, and the retrograde signalling logic is solid. What the system does at the level of whole-brain function, mood, memory formation, or any specific clinical outcome, that's where hypothesis and evidence get tangled, and precision is worth demanding. The wider picture beyond the brain is in the endocannabinoid system in whole-body physiology, and the Wikipedia entry on the endocannabinoid system offers a general reference point.

For anyone wanting to go deeper on individual components, the glossary entries for CB1, anandamide, 2-AG, and the TRPV1 receptor cover the receptor-level detail in more depth, and the cannabinoid science FAQ answers common questions in plain terms.

Sources

• Endocannabinoid system: An overview of its potential in current medical practice; National Center for Biotechnology Information (NCBI/NIH)
• Biochemistry, Endocannabinoids, StatPearls, NCBI Bookshelf
• Cannabidiol (CBD) products: information for consumers, Therapeutic Goods Administration (TGA)
• The Endocannabinoid System as an Emerging Target of Pharmacotherapy; Pharmacological Reviews, NCBI

Elise Tran, Science writer, cannabinoids & the ECS