The Endocannabinoid System as a Modulation Target

Made on demand, released backwards animal

The classical neurotransmitter model does not describe this system. Anandamide was isolated from porcine brain in 1992 as an endogenous ligand for the receptor that THC binds, and 2-arachidonoylglycerol from canine gut in 1995. Neither is stored in vesicles: both are lipids cleaved out of membrane phospholipid precursors at the moment they are needed, in response to postsynaptic depolarisation and rises in intracellular calcium, or to metabotropic receptor activation. Because they are made in the postsynaptic cell but act on receptors sitting on the presynaptic terminal, the signal travels backwards relative to the synapse. Wilson and Nicoll established this retrograde geometry at hippocampal synapses in 2001, and it is now the organising principle of the whole field: an endocannabinoid is a message from the receiving neuron back to the sending one, saying send less. Depolarisation-induced suppression of inhibition and of excitation are the canonical readouts.

• No storage pool means the rate-limiting step is synthesis, not release.
• Retrograde signalling means the effect is presynaptic: reduced neurotransmitter release.
• Because synthesis is activity-driven, the system is spatially and temporally local — it acts at the synapses that just fired, not brain-wide.

Sources: Devane WA 1992 · Mechoulam R 1995 · Wilson RI 2001 · Kano M 2009

Two ligands, two biosynthetic routes, two degradation routes animal

Anandamide and 2-AG are routinely spoken of as a pair, which obscures how separate they are. Anandamide is produced from N-acylphosphatidylethanolamine by a specific phospholipase D (NAPE-PLD), characterised molecularly by Okamoto and colleagues in 2004. 2-AG is produced from arachidonic-acid-containing diacylglycerol by sn-1-selective diacylglycerol lipases, of which DAGLα is the neuronal form, cloned by Bisogno and colleagues in 2003. They are then destroyed by different enzymes in different cellular compartments: anandamide by fatty acid amide hydrolase (FAAH), largely postsynaptic and intracellular on the endoplasmic reticulum, and 2-AG overwhelmingly by monoacylglycerol lipase (MAGL), largely presynaptic. They also differ at the receptor: 2-AG behaves as a reasonably efficacious agonist at both CB1 and CB2, while anandamide has lower efficacy at CB2 and additionally activates TRPV1, which no strictly cannabinoid account of it captures. Two ligands, four enzymes, two compartments — which is why an intervention aimed at one of them is not an intervention in "the endocannabinoid system" as a whole.

Sources: Okamoto Y 2004 · Bisogno T 2003 · Cravatt BF 1996 · Blankman JL 2007 · Howlett AC 2002 · Nomura DK 2011

Termination is hydrolysis, not vesicular recapture in vitro

A monoamine transmitter is cleared from the synapse largely by being pumped back into the presynaptic terminal and reloaded into vesicles. Endocannabinoid signalling has no equivalent step. The signal ends because the molecule is hydrolysed: anandamide to arachidonic acid and ethanolamine by FAAH, 2-AG to arachidonic acid and glycerol by MAGL. That is why enzyme inhibition, and not transporter blockade in the monoamine sense, is the natural pharmacological handle on this system, and why the entire "indirect agonist" strategy exists. The step that gets the lipid from the extracellular space to the intracellular enzyme is real and measurable but its mechanism remains genuinely unsettled — there is good evidence against a classical saturable transporter of the monoamine type, and competing accounts involving FAAH-maintained concentration gradients and intracellular fatty-acid-binding-protein shuttles. That argument has its own page on this shelf and is marked contested there.

Sources: Cravatt BF 1996 · Dinh TP 2002 · Blankman JL 2007 · Glaser ST 2003 · Chicca A 2012

Raising tone versus adding an agonist: the distinction this shelf is built on human data

These are not two routes to the same place. An exogenous agonist occupies receptors wherever it distributes, for as long as it is present, whether or not the synapse in question was signalling. Enzyme inhibition does something narrower: it prolongs and amplifies endocannabinoid signalling only at synapses that are already making endocannabinoid. The pattern of activity is set by the tissue, not by the drug. Three consequences follow, and they are the reason the distinction is pharmacological rather than rhetorical. First, efficacy. Δ9-THC is a partial agonist at CB1; a synthetic aminoalkylindole such as JWH-018 is a full agonist with several-fold higher affinity, and the clinical difference between those two facts is the difference between intoxication and the synthetic-cannabinoid toxidrome. Second, tolerance. Chronic exogenous agonism downregulates CB1: positron-emission tomography in chronic daily cannabis smokers showed reversible, regionally selective reduction of CB1 availability, recovering over roughly four weeks of abstinence. Third, and most importantly, indirect inhibition has a CEILING that direct agonism does not. You cannot raise endocannabinoid tone past the rate at which the tissue synthesises endocannabinoid. A full agonist has no such limit; its dose is the limit.

• Direct agonism: effect determined by the drug, ceiling set by the dose and the agonist efficacy.
• Indirect (enzyme) inhibition: effect determined by where and when the tissue is already signalling, ceiling set by synthesis rate.
• The ceiling is why the indirect strategy was attractive therapeutically — and its failure modes are a separate matter from its ceiling, which is what the MAGL and FAAH pages cover.

Sources: Showalter VM 1996 · Huffman JW 2005 · Trecki J 2015 · Hirvonen J 2012 · Howlett AC 2002

The ceiling is real, but it is not the same as safety human data

Two findings keep the ceiling argument honest. Chronic complete MAGL blockade in mice does not simply produce sustained cannabinoid effects; it produces functional antagonism of the system, with CB1 desensitisation and downregulation and loss of the acute behavioural effects — the ceiling turns into a cliff when the enzyme is fully and permanently inhibited rather than partially and reversibly. And in the FAAH programmes, the ceiling delivered exactly what it promised on the mechanism and nothing at all on the endpoint: PF-04457845 raised circulating fatty acid amides in humans and still failed to produce analgesia in osteoarthritic knee pain. Elevated endocannabinoid tone is a biomarker, not an outcome. The shelf states this on both enzyme pages because the temptation with a "work with your own system" framing is to treat the mechanism as though it were the result.

Sources: Schlosburg JE 2010 · Huggins JP 2012

Why the system is a target at all animal

The reason anyone builds inhibitors for these enzymes is that endocannabinoid signalling is mobilised by injury and inflammation and appears to be part of an endogenous protective response. 2-AG rises after closed head injury in mice and exogenous 2-AG administered after injury improved outcome in that model, which is the observation behind two decades of neuroprotection work. On the inflammatory side, MAGL sits at a junction rather than at an endpoint: hydrolysing 2-AG produces the arachidonic acid that feeds cyclooxygenase-derived prostaglandin synthesis, so blocking MAGL simultaneously raises a cannabinoid and lowers a prostaglandin. Those two facts, rather than any recreational consideration, are why the enzyme-inhibition literature exists and why its negative results matter.

Sources: Panikashvili D 2001 · Nomura DK 2011

How to read this shelf

The enzyme pages (magl, faah) cover the two hydrolases, their tool compounds and what went wrong with them in humans. The receptor pages (cb1, cb2) cover distribution, coupling and the affinity numbers that make partial-versus-full agonism a toxicological fact rather than a footnote. The reuptake page covers the contested transport step and the compounds that act on it cleanly. natural-inhibitors is the master table of botanical compounds by target and potency, with the in-vitro-to-human gap stated as its own section rather than buried. piperaceae follows the pepper-family thread, and entourage-effect and kava-potentiation are the two places where a popular claim has to be separated from what the evidence supports. No page on this shelf contains a preparation method, an extraction route or a dose; the formulation shelf handles the arithmetic of handling a compound already in hand.