Cannabis does not introduce something foreign into your physiology — it borrows a signalling network your body already runs.
1. What the Endocannabinoid System Actually Is
The endocannabinoid system is a lipid-based signalling network found throughout the body. It is not an organ or a single pathway, but a toolkit cells use to adjust how loudly other cells are talking.
It has three parts. Receptors sit on cell membranes and receive the signal. Endogenous ligands — endocannabinoids — are molecules the body builds to activate them. Enzymes construct those ligands on the spot and then destroy them, usually within seconds. Unlike classical neurotransmitters, endocannabinoids are not stored in advance; they are made from membrane fats at the moment they are needed. The system behaves less like a switch and more like a brief, local correction.
2. The Receptors: Two Types, Two Different Jobs
Cannabinoid Receptor Type 1 and the Central Nervous System
Cannabinoid receptor type 1 (CB1) is among the most abundant G protein-coupled receptors in the mammalian brain. Kendall and Yudowski, in Frontiers in Cellular Neuroscience in 2017, describe high levels in the neocortex, hippocampus, basal ganglia and cerebellum, and note it is enriched at presynaptic and axonal compartments — on the terminal that releases neurotransmitter, not the one receiving it. That anatomy maps onto the familiar effects.
| Region rich in this receptor | Associated effect |
|---|---|
| Neocortex | Altered time perception and attention |
| Hippocampus | Short-term memory difficulty |
| Basal ganglia, cerebellum | Motor slowing, impaired balance |
| Hypothalamus | Changes in appetite |
Haspula and Clark, in a 2020 review in the International Journal of Molecular Sciences, report that CB1 expression is relatively sparse in the medullary respiratory control centres of the brainstem — the observation behind the point that cannabinoids do not suppress breathing as opioids can. That is a narrow claim about specific nuclei, not a claim of harmlessness.
Cannabinoid Receptor Type 2, Immune Cells and Peripheral Tissue
Cannabinoid receptor type 2 (CB2) sits predominantly in immune cells and tissues. Galiègue and colleagues, in the European Journal of Biochemistry in 1995, ranked expression across human leukocytes roughly as tonsillar B cells above natural killer cells, then monocytes and granulocytes, then T lymphocytes; a 2022 analysis by Simard and colleagues in Frontiers in Pharmacology found the strongest signals in eosinophils and B cells. In the brain, CB2 is chiefly associated with microglia — which is why it is discussed in the context of inflammation rather than intoxication. Activating it does not produce a high.
| Feature | Cannabinoid receptor type 1 | Cannabinoid receptor type 2 |
|---|---|---|
| Main location | Central nervous system, presynaptic terminals | Immune cells, spleen, tonsils, microglia |
| Functional theme | Dampening neurotransmitter release | Modulating immune signalling |
| Causes intoxication? | Yes | No |
3. The Endogenous Ligands and the Enzymes That Control Them
Two endocannabinoids are well established, both derived from arachidonic acid already present in cell membranes.
Anandamide — chemically arachidonoylethanolamide — was identified first, in the early 1990s; its name comes from the Sanskrit ananda, meaning bliss. Built by N-acyl phosphatidylethanolamine-specific phospholipase D, it acts as a partial agonist. 2-arachidonoylglycerol (2-AG) followed, generated by diacylglycerol lipase; Haspula and Clark describe it as a full agonist at both receptor types, present in brain tissue at substantially higher concentrations.
| Property | Anandamide | 2-arachidonoylglycerol |
|---|---|---|
| Built by | N-acyl phosphatidylethanolamine-specific phospholipase D | Diacylglycerol lipase |
| Broken down by | Fatty acid amide hydrolase | Monoacylglycerol lipase |
| Receptor activity | Partial agonist | Full agonist |
| Brain concentration | Lower | Substantially higher |
The Degrading Enzymes as Volume Control
Fatty acid amide hydrolase degrades anandamide; monoacylglycerol lipase degrades 2-arachidonoylglycerol. Because endocannabinoids are not stored, these enzymes effectively are the volume control: signal strength depends on how fast a ligand is made versus destroyed. That is why much research targets enzymes rather than receptors — slowing degradation raises signalling only where it already occurs. Whether that yields clinical benefit in humans remains open.
4. Retrograde Signalling: Messages That Travel Backwards
Made on Demand, Released Backwards
In an ordinary synapse the presynaptic neuron releases a transmitter that crosses the gap and acts on the postsynaptic neuron. Endocannabinoid signalling runs the other way: when the postsynaptic neuron is strongly activated, it builds endocannabinoids from its own membrane lipids and releases them backwards across the cleft, where they bind CB1 receptors on the presynaptic terminal and suppress further transmitter release. Castillo, Younts, Chávez and Hashimotodani, in a 2012 review in Neuron, call this the principal mode by which endocannabinoids mediate synaptic plasticity, while noting that non-retrograde mechanisms also exist.
Why This Matters for Understanding Effects
It is a feedback brake, not a driver: the signal comes from the cell being over-stimulated and reduces its own input. It is also local and brief — these lipids do not travel far and are degraded within seconds.
Plant cannabinoids break both properties at once. Tetrahydrocannabinol arriving through the bloodstream is not made on demand, not local, and not cleared within seconds; it reaches every CB1 receptor it can access, whether or not that synapse asked for a signal. The difference between endocannabinoid signalling and cannabis intoxication is less about the molecule than about where, when and for how long.
5. How Plant Cannabinoids Interact With This System
Tetrahydrocannabinol as a Partial Agonist
Tetrahydrocannabinol is described by Roger Pertwee in a 2008 review in the British Journal of Pharmacology as a partial agonist at both cannabinoid receptor types. It does not drive CB1 as hard as a full agonist could — but delivered systemically rather than synapse by synapse, it acts across the whole receptor population at once.
Cannabidiol: Honestly, Still Contested
What is reasonably supported about cannabidiol is negative: it has weak affinity for the main binding site of both cannabinoid receptors and is not a conventional CB1 agonist. Pertwee’s review noted it acting with unexpectedly high potency as an antagonist of cannabinoid receptor agonists. Laprairie and colleagues, in the British Journal of Pharmacology in 2015, characterised it as a negative allosteric modulator of cannabinoid receptor type 1 — binding a secondary site and damping other agonists rather than activating the receptor itself.
Beyond that the picture fragments. A 2023 review by León and colleagues in Molecules catalogues roughly fifty-six proposed molecular targets, among them serotonin 5-HT1A receptors, TRPV1 channels, GPR55 and peroxisome proliferator-activated receptor gamma. Those authors state plainly that cannabidiol’s pharmacology has not been fully elucidated, and note that many reported actions were shown at concentrations above those plausibly reached in the body. Cannabidiol’s mechanism is a live scientific controversy — a hypothesis space, not an established account.
| Claim | Evidence status |
|---|---|
| Tetrahydrocannabinol is a partial agonist at cannabinoid receptor type 1 | Well established |
| Cannabidiol is a negative allosteric modulator at that receptor | Supported |
| Cannabidiol acts via serotonin, TRPV1 and other targets | Proposed; relevance at real concentrations disputed |
| A single unifying mechanism for cannabidiol | Not established |
6. The Homeostasis Framing and Where It Breaks Down
The system is routinely introduced as the body’s “master regulator of balance.” As a teaching device this is defensible: it is inhibitory, widely distributed, activated by demand, and appears to buffer excursions in neural, immune and metabolic activity. As a claim it is much weaker. In physiology, homeostasis means a regulated variable with a set point and a correction mechanism — blood pH, core temperature, plasma glucose. This system has no single variable of that kind; it works through many local loops.
“Dampens excessive transmitter release at specific synapses” is a measured finding; “restores balance throughout the body” is a teaching metaphor; “supporting your endocannabinoid system will bring you back to health” is neither.
7. Clinical Endocannabinoid Deficiency: A Hypothesis Under Test
In 2001, and again in a 2016 paper in Cannabis and Cannabinoid Research, Ethan Russo proposed the theory of clinical endocannabinoid deficiency: that some treatment-resistant conditions — migraine, fibromyalgia and irritable bowel syndrome most prominently — might share an underlying insufficiency of endocannabinoid signalling. Russo frames it explicitly as a theory built on genetic overlap and comorbidity, noting that objective proof and trial data were lacking when he first put it forward.
What the Evidence Supports and What It Does Not
The strongest strand cited is cerebrospinal fluid measurement: statistically significant reductions in cerebrospinal fluid anandamide have been documented in people with chronic migraine compared with controls. For fibromyalgia and irritable bowel syndrome the evidence is more indirect. The theory has drawn methodological criticism, most directly in a 2020 paper titled Practical Considerations of Hypotheses and Evidence in Cannabis Pharmacotherapy: Refining Expectations of Clinical Endocannabinoid Deficiency, which argues for tightening the evidence standard applied to it.
The fair summary: this is a hypothesis, not an established finding. No named condition has been shown to be caused by endocannabinoid deficiency, and no conclusion about any particular intervention follows from it.
8. The Endocannabinoid System Beyond the Brain
The Gut, the Gut-Brain Axis and the Microbiome
CB1 is expressed on enteric neurons, and endocannabinoid signalling is implicated in gastrointestinal motility, secretion, visceral sensation and barrier function. Because gut and brain communicate continuously through neural, endocrine and immune routes, the system sits at a plausible junction in the gut-brain axis — though most mechanistic detail is from animal models.
The microbiome extension is newer and much shakier. A 2025 review by Wang, Guo, Mao and Chen in Frontiers in Cellular and Infection Microbiology surveys proposed links between gut bacteria and endocannabinoid signalling: bacterial production of endocannabinoid-like molecules, short-chain fatty acid effects on enzyme expression, and effects on barrier integrity. Those authors are direct about the limits — most of this work is observational and does not establish causality. It is a promising area honestly described as preliminary, not a basis for individual health conclusions.
Immune Tissue, Skin, Bone and Reproductive Tissue
Immune tissue. CB2 on leukocytes places endocannabinoid signalling inside immune regulation, including cytokine release and cell migration. Direction of effect varies with cell type and context, so “cannabinoids suppress inflammation” oversimplifies it.
Skin. Bíró and colleagues described a cutaneous endocannabinoid system in Trends in Pharmacological Sciences in 2009, involving keratinocytes, sebaceous gland cells, sensory nerve endings and resident immune cells.
Bone. Ehrenkranz and Levine, in the Journal of Clinical Endocrinology & Metabolism in 2019, describe CB2 as the predominant cannabinoid receptor in bone, with CB1 on sympathetic nerve terminals innervating it. They state that human studies are limited and the clinical relevance of cannabis effects on bone is unknown.
Reproductive tissue. Walker, Holloway and Raha, in the Journal of Ovarian Research in 2019, report cannabinoid receptors and endocannabinoid enzymes across ovary, endometrium, oviduct and myometrium, noting most knowledge comes from in vitro and animal work.
| Tissue system | Reasonably established | Not established |
|---|---|---|
| Immune system | Receptor expression on leukocytes | Consistent outcomes in humans |
| Gut microbiome | Correlations, plausible mechanisms | Causality; human generalisability |
| Skin, bone, reproductive tissue | Receptor and enzyme localisation | Mechanistic human evidence |
9. Tolerance and Receptor Downregulation With Heavy Use
Sustained CB1 activation does not leave the receptor population unchanged. Hirvonen and colleagues, in Molecular Psychiatry in 2012, used positron emission tomography in chronic daily cannabis smokers and found CB1 receptor density downregulated in cortical regions, in proportion to duration of use.
What Recovery Data Exists
The recovery finding is equally important and less often quoted: after roughly four weeks of continuously monitored abstinence on a secure research unit, CB1 receptor density returned to normal levels. A 2018 systematic review of human evidence by Marco Colizzi and Sagnik Bhattacharyya likewise concluded that tolerance develops with regular use and that abstinence is associated with at least partial reversal.
Qualifications belong with this: the study was small, involved heavy daily users, and found regionally selective rather than global change, and tolerance to different effects does not resolve at the same rate. What the data support is general — receptor-level adaptation to chronic heavy exposure is real, measurable and at least substantially reversible.
10. Why Individual Response Varies So Much
Baseline receptor density differs between people and between regions within a person, and is itself modified by prior exposure — which also shapes response through expectation, learned context and familiarity, all hard to separate from pharmacology.
Genetic variation is documented. The most studied example is a common missense variant in the gene encoding fatty acid amide hydrolase, designated C385A. Dincheva and colleagues, in Nature Communications in 2015, reported that it is associated with altered fronto-amygdala function in mice and humans, consistent with reduced enzyme activity and higher anandamide levels. A 2015 study by Boileau and colleagues found the same variant affects brain binding of a positron emission tomography tracer for the enzyme — confirmation that it changes enzyme availability in the living human brain.
Each is a group-level effect: none is measurable in ordinary practice, and none predicts what a given person will experience. Current science explains why people differ far better than it predicts how any one person will.
We would like to hear from the community. Which part of this system surprised you most — the backwards direction of the signalling, the breadth of peripheral tissue involvement, or how unresolved cannabidiol’s mechanism still is? If you work in physiology, pharmacology, nursing or laboratory science, corrections and additions are especially welcome.
This article is part of the Asiannabis Community educational series on medical cannabis. It is general information, not medical advice — treatment decisions belong with a licensed practitioner. Content is for educational purposes within jurisdictions where medical cannabis is legally permitted.