NATRIAM · Research Library

Cannabinoid science, in plain words

Seven reading-length articles on the endocannabinoid system, cannabinoid receptors, the skin, what the research has actually looked at, and how to read a lab report. Written for curious adults, built from peer-reviewed sources, with every citation linked.

These pages are education, not medical advice, and they do not describe NATRIAM SOOTHE or any product. Each finding is labeled by its evidence: human, animal or in vitro.

Article 1 of 7

The endocannabinoid system, explained

Its three parts, where it is found in the body, how it was discovered in the early 1990s, and what researchers still do not know.

A system most of us have never heard of

Most people can name the body's familiar signaling systems: hormones, nerves, immunity. Far fewer have heard of the endocannabinoid system, or ECS. That is not surprising: it was only identified in the early 1990s.

The name can confuse. "Endo" means "from within." It is called "cannabinoid" because its parts were found by scientists studying how compounds from the cannabis plant act on the body, but the ECS itself is built by the body and is present whether or not a person has ever used cannabis (Lu and Mackie, 2016).

A note on evidence levels. Throughout this article, findings are labeled by where they come from: in vitro (isolated molecules or cells in a dish), animal (tissue from or experiments in animals), or human. Most of what is known about the ECS comes from the first two categories. A finding at one level does not transfer to another.

Three parts that work together

Reviews usually describe the ECS as having three parts (Lu and Mackie, 2016; Zou and Kumar, 2018).

Endocannabinoids. These are small fat-based molecules made by the body. The two best studied are anandamide and 2-arachidonoylglycerol, usually shortened to 2-AG. Unlike many messenger chemicals, they are not stored in advance; cells make them "on demand" from fats in the cell membrane, use them briefly, and clear them away (Lu and Mackie, 2016).

Receptors. A receptor is a protein on a cell that a molecule can attach to, like a key fitting a lock. The two main cannabinoid receptors are called CB1 and CB2. Reviews describe CB1 as the most widely expressed receptor of its family (the G protein-coupled receptors) in the brain (Zou and Kumar, 2018). CB2 is found mostly on cells of immune origin (Lu and Mackie, 2021).

Enzymes. Enzymes build or break down other molecules; two are central here. Fatty acid amide hydrolase, or FAAH, is the main enzyme that breaks down anandamide. Monoacylglycerol lipase, written MAGL or MGL, is the main enzyme that breaks down 2-AG (Lu and Mackie, 2016). Because these enzymes act quickly, endocannabinoid signals tend to be brief and local.

Where it is found in the body

CB1 receptors are especially concentrated in the nervous system, including brain regions involved in memory, movement and coordination, and they are also found in tissues such as the liver, fat tissue and skin (Lu and Mackie, 2021). CB2 receptors are found mainly on cells of the immune system, with smaller amounts elsewhere (Zou and Kumar, 2018). These distribution maps come largely from animal tissue and laboratory studies.

In the brain, much of the action is at synapses, the tiny gaps where one nerve cell signals the next. Endocannabinoids are unusual: in animal studies they often travel "backward" across the synapse, from the receiving cell to the sending cell, and turn down the signal being sent (Lu and Mackie, 2016).

What "homeostasis" means

Homeostasis is a textbook term for the body's habit of keeping internal conditions within a range. Think of a thermostat nudging a room back to its set temperature.

The ECS is often described in this vocabulary because of how it behaves in laboratory and animal studies: made on demand, acting locally, and mostly adjusting the volume of other signals rather than starting new ones. Reviews note that the ECS "influences, and is influenced by" many other signaling pathways and is not an isolated system (Lu and Mackie, 2021). That is a description of biology. It is not a claim that the system can be "supported," "balanced" or changed by any product, and no such conclusion follows from it.

How it was discovered

The story begins with the plant. In 1964, Yechiel Gaoni and Raphael Mechoulam in Israel isolated and described the structure of THC, the main psychoactive compound in cannabis (Gaoni and Mechoulam, 1964) [chemistry, in vitro]. For the next two decades, how THC acted on the body remained unclear.

In 1988, William Devane, Allyn Howlett and colleagues showed that rat brain tissue contained a specific binding site for cannabinoid compounds, pointing to a receptor (Devane et al., 1988) [animal tissue]. In 1990, Lisa Matsuda and colleagues cloned the gene for that receptor, now known as CB1 (Matsuda et al., 1990) [in vitro, using rat DNA expressed in cultured cells].

A receptor implies that the body makes its own key. In 1992, Devane, now in Mechoulam's laboratory, isolated the first endocannabinoid from pig brain and named it anandamide, after the Sanskrit word ananda (Devane et al., 1992) [animal tissue]. In 1993, Sean Munro and colleagues described a second receptor, CB2, identified in immune-type cells (Munro et al., 1993) [in vitro, cultured cells].

In 1995, two groups independently identified 2-AG as a second endocannabinoid, one in canine gut tissue and one in rat brain (Mechoulam et al., 1995; Sugiura et al., 1995) [animal tissue]. The enzymes came next: FAAH was characterized in 1996 (Cravatt et al., 1996) [in vitro], and MAGL was identified as the main brain enzyme clearing 2-AG in 2002 using rat brain tissue and cultured rat neurons (Dinh et al., 2002) [animal tissue and cultured cells].

None of these foundational studies was conducted in people. They established that the system exists; they do not, on their own, say what changing it would mean for a person.

What researchers still do not know

Even after thirty years, large gaps remain. Reviews point out that anandamide can be made by several pathways, and which ones matter most in the living brain is still unsettled (Lu and Mackie, 2016). How endocannabinoids cross the space between cells is not fully understood; reviewers write that identifying the proposed membrane transporter would greatly aid progress in the field (Lu and Mackie, 2021). The role of CB2 in endocannabinoid signaling at synapses is described as "largely elusive" (Zou and Kumar, 2018).

A 2018 review catalogued the many approaches pharmaceutical researchers have taken to study the system and the obstacles they have met, partly because the same molecules do several jobs at once (Di Marzo, 2018). That review concerns drug-discovery research, not consumer products.

CBD, a plant compound, does not fit neatly into this picture. It has negligible affinity for CB1 and CB2 (Zou and Kumar, 2018), and reviewers state that its molecular targets remain to be worked out (Lu and Mackie, 2021). Human studies of how CBD acts at the molecular level are few. The National Center for Complementary and Integrative Health describes research on cannabis and cannabinoids for most questions as being in its early stages (NCCIH, 2019). No conclusions about any product can be drawn from the biology described here.

Frequently asked questions

Does everyone have an endocannabinoid system? Yes. The system is built by the body itself and is present whether or not a person has ever been exposed to cannabis (Lu and Mackie, 2016).

Are endocannabinoids the same as cannabinoids from the plant? No. Endocannabinoids are made inside the body and have a different chemical structure from plant compounds like THC and CBD. They carry the plant's name because the receptors they act on were first identified through research on THC (Zou and Kumar, 2018).

Why is so much of the research in animals? The foundational discoveries used rat, pig and dog tissue because such experiments cannot ethically be done in people. Reviews emphasize that many questions remain open (Di Marzo, 2018), and that results from animal tissue do not automatically apply to humans.

References

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

Article 2 of 7

CB1 and CB2 receptors: what they are and where they are

The two cannabinoid receptors, when they were cloned, where each one sits, and how THC, CBD and the body's own cannabinoids interact with them differently.

Your body is full of tiny docking stations called receptors: proteins that sit in a cell's outer wall and wait for a particular molecule. When the right one docks, the receptor passes a message inside.

Two of these, CB1 and CB2, are the best-known parts of the endocannabinoid system. This article explains what they are, where they are found, and how different molecules interact with them. It describes biology, not products, and it does not describe effects in people.

Two discoveries, three years apart

In 1990, Lisa Matsuda and colleagues at the National Institutes of Health cloned the gene for a receptor in rat brain that responded to cannabinoids (Matsuda et al., 1990). Three years later, Sean Munro's group in Cambridge cloned a second receptor, CB2, found not in the brain but in immune cells of the spleen (Munro et al., 1993). The two share only about 44 percent of their amino acid sequence overall, with closer similarity in the membrane-spanning regions (Munro et al., 1993; Howlett et al., 2002).

What "G-protein-coupled receptor" means in plain words

Both belong to a large protein family called G-protein-coupled receptors, or GPCRs. Picture a ribbon threaded through the cell's outer membrane seven times: outside, its folds form a pocket where a molecule can dock; inside, it is attached to a partner protein called a G protein. When a molecule docks, the receptor shifts shape and nudges the G protein, which relays the message inside. It works like a doorbell: the button is outside, the wiring is inside.

For CB1 and CB2, the usual message turns down an enzyme called adenylyl cyclase, lowering a messenger chemical (cyclic AMP) inside the cell (Howlett et al., 2002).

Where CB1 is found

CB1 is the most abundant cannabinoid receptor and is found mainly in the central nervous system (Lu and Mackie, 2016). A 1990 autoradiography study of brain sections from several mammalian species, including human tissue, found binding densest in the outflow nuclei of the basal ganglia, the hippocampus and the cerebellum, and sparse in lower brainstem areas that control breathing and heartbeat (Herkenham et al., 1990).

CB1 sits mostly on nerve endings, where its signal dials down neurotransmitter release (Howlett et al., 2002). Smaller amounts have been measured in tissues such as the adrenal gland, heart and lung (Galiègue et al., 1995).

Where CB2 is found

CB2 is found mainly in immune tissue. A 1995 study that measured messenger RNA in human tissue samples found CB2 highest in the spleen and tonsils, ten to one hundred times above CB1 in those tissues, with B cells carrying the most among blood cells, and did not detect it in the brain (Galiègue et al., 1995). Immune cells also carry some CB1 (Howlett et al., 2002).

Both receptors are in the skin

In 2005, antibody staining of human skin sections found both CB1 and CB2 in nerve fiber bundles, mast cells, macrophages, epidermal keratinocytes, hair follicles, sebaceous glands and sweat glands, arranged in a complementary pattern in the epidermis and hair follicles (Ständer et al., 2005). This was a tissue-mapping study: it shows where the receptors are, not what happens when anything is applied to skin.

Later reviews describe a "c(ut)annabinoid" system: endocannabinoids, their receptors and the enzymes that make and break them down are all present in skin (Bíró et al., 2009; Tóth et al., 2019). Most of this knowledge comes from cell cultures and animal experiments; evidence in humans remains limited, and no conclusions about any product can be drawn from receptor mapping alone.

THC, CBD and your own endocannabinoids interact differently

Your body makes its own cannabinoids, chiefly anandamide and 2-arachidonoylglycerol (2-AG), which dock in the main pocket of CB1 and CB2 and switch them on (Lu and Mackie, 2016).

THC, the intoxicating compound in cannabis, docks in the same pocket as a partial agonist at both receptors: it presses the doorbell, but not all the way (Pertwee, 2008).

CBD is the odd one out: it has low affinity for the main pocket of both receptors, so it does not dock there well. Yet in tissue and cell experiments, it blocked CB1 and CB2 activators at concentrations well below its binding affinity would predict, and behaved as an inverse agonist at CB2 (Thomas et al., 2007; Pertwee, 2008). A 2015 study in cultured cells offered one explanation: CBD acted as a negative allosteric modulator of CB1, binding to a different spot on the receptor and making it less responsive to THC and 2-AG. Picture a hand resting on the doorbell that makes the button harder to press (Laprairie et al., 2015). These are in vitro findings; what they mean in a living person has not been established.

Other targets researchers discuss

Because CBD docks so weakly at CB1 and CB2, researchers have looked for other sites. Each of the findings below comes from laboratory cell systems, not from studies in people.

TRPV1 is the ion channel that responds to capsaicin, the hot ingredient in chili peppers. In a 2001 in vitro study, CBD activated it in cells engineered to express it, then desensitized it to capsaicin (Bisogno et al., 2001).

GPR55 is a formerly "orphan" receptor. In a 2007 study of cells transfected with GPR55, CBD did not activate the receptor but acted as an antagonist, blocking the response to a synthetic cannabinoid (Ryberg et al., 2007).

PPARs are receptors inside the cell nucleus that regulate genes. A 2016 review summarized evidence, mostly from reporter-gene assays and animal studies, that some cannabinoids activate PPAR alpha and gamma (O'Sullivan, 2016).

A 2010 international pharmacology committee concluded that none of these yet met the criteria for a third cannabinoid receptor (Pertwee et al., 2010).

Why this matters when you read about CBD

Receptor biology tells you where a molecule can dock and what signal that sends inside a cell. It does not tell you what happens when a person uses any product. Human evidence remains limited, and the findings above should not be read as evidence of any effect in people.

References

Frequently asked questions

Are CB1 and CB2 the only cannabinoid receptors? They are the only two officially classified. A 2010 expert review examined TRPV1, GPR55 and PPARs and found none yet met the criteria for a "CB3" label (Pertwee et al., 2010).

Does CBD bind to CB1 and CB2 the way THC does? No. THC docks in the main pocket as a partial agonist. CBD has low affinity for that pocket; in cell studies it acted as a negative allosteric modulator of CB1, binding elsewhere and dampening the receptor's response (Laprairie et al., 2015; Pertwee, 2008). These are laboratory findings, not findings in people.

Do skin cells have cannabinoid receptors? Yes. Antibody staining of human skin has found CB1 and CB2 on nerve fibers, immune cells, keratinocytes, hair follicles and glands (Ständer et al., 2005). What that means for any topical product has not been established in human studies.

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

Article 3 of 7

The skin has its own endocannabinoid system

Receptors and enzymes in skin cells, what laboratory, animal and human studies have measured when cannabinoids are applied to skin, and why topical research is hard.

Most people first hear of the endocannabinoid system in connection with the brain. Yet the skin, the body's largest organ, appears to carry a version of its own. This article maps what has been reported, how researchers study cannabinoids applied to skin, and what remains open. It describes published research only; it does not describe any product.

How to read the evidence labels. Each finding below is tagged by the kind of study it comes from: review (a paper summarizing other papers), human tissue (donated or biopsied human skin studied in the laboratory), in vitro (cultured cells), animal, or human (measurements in living people). Findings from cells, tissue samples or animals do not show what happens in people.

A signaling system, briefly

The endocannabinoid system has three parts: the body's own cannabinoid-like molecules, chiefly anandamide and 2-AG; the receptors they attach to, best known as CB1 and CB2; and enzymes that build these molecules when needed and break them down afterward.

A 2009 review in Trends in Pharmacological Sciences gathered evidence that skin cells contain each of these parts (Bíró et al., 2009; review). The same review listed, among its open questions, whether every member of the system is functionally expressed in human skin and its appendages. A 2020 dermatology review described the skin's endocannabinoid system as newly found (Sheriff et al., 2020; review).

Where the parts sit in the skin

Cannabinoid receptors have been reported in nearly every cell type in skin.

  • Keratinocytes, the cells of the outer layer, carry CB1 and CB2, with CB1 more prominent in the upper, more mature layers (Bíró et al., 2009; review of cultured-cell and tissue work).
  • Hair follicles express CB1 and, in organ culture, produce their own anandamide and 2-AG (Bíró et al., 2009; review).
  • Sebocytes, the oil-producing cells of sebaceous glands, carry both receptors: CB1 has been reported in the more differentiated central cells and CB2 mainly in the basal cells (Tóth et al., 2019; review).
  • Sensory nerve fibers and mast cells carry both. A 2005 immunohistochemistry study of human skin sections found CB1 and CB2 on nerve fiber bundles, mast cells, macrophages, keratinocytes, hair follicle epithelium, sebocytes and sweat glands (Ständer et al., 2005; human tissue). That study located the receptors; it did not test what they do.
  • Immune cells that live in or pass through the skin express CB2 (Bíró et al., 2009; review).

The enzymes are there too. A 2019 review in Molecules reported the building enzymes (NAPE-PLD and DAGL) and the breakdown enzymes (FAAH and MAGL) in a cultured human sebocyte line and in sebaceous glands in tissue sections, with one subtype (DAGLα) weaker in tissue (Tóth et al., 2019; review). Skin also carries related channels (TRPV1, TRPV3 and TRPV4) that respond to heat and to several plant compounds and are often discussed alongside the cannabinoid system (Tóth et al., 2019; review).

One caution: much of this map comes from cultured cells, tissue sections and mouse studies. The 2009 review's open question about functional expression in living human skin still stands (Bíró et al., 2009), and the 2019 review repeatedly points out where well-controlled human studies have not yet been done (Tóth et al., 2019).

What researchers study when a cannabinoid meets skin

When scientists apply a cannabinoid such as CBD to skin, they ask three questions. Does it get through the outer barrier? Does it stay in the skin or move into the bloodstream? And how does the formulation change the answer?

A 2022 review in Pharmaceutics separates "local (topical)" delivery, where the aim is to keep a molecule in the skin, from "transcutaneous (transdermal)" delivery, where the aim is to move it through the skin into general circulation (Mahmoudinoodezh et al., 2022; review).

CBD is a demanding molecule for either path. It dissolves readily in oil but barely in water. The 2022 review describes cannabinoids as a class as highly hydrophobic, with a log P of roughly 6 to 7 and water solubility of a few micrograms per milliliter. That lets them enter the oily outermost layer (the stratum corneum) but makes the water-rich layers beneath a bottleneck (Mahmoudinoodezh et al., 2022; review).

Laboratory work with donated human skin has measured this. In a 2004 diffusion study, CBD and cannabinol passed through human skin samples about ten times more readily than delta-8-THC, and adding 30 to 33 percent ethanol increased the flow of both delta-8-THC and CBD (Stinchcomb et al., 2004; human tissue, in vitro). A 2020 study using epidermis from a single donor compared seven carriers (paraffin, olive oil, propylene glycol, PEG 400, an ointment, a gel and a printed patch) and found that the carrier changed both how much CBD crossed and how much stayed in the epidermis (Casiraghi et al., 2020; human tissue, in vitro).

In a 2014 study of a cultured human sebocyte line, the change CBD produced in those cells' lipid synthesis was prevented by blocking the TRPV4 channel but not by blocking CB1 or CB2 (Oláh et al., 2014; in vitro, with an ex vivo human skin organ-culture arm). This is one reason the skin's system is described as broader than the two classical receptors. It is a cell-culture observation and says nothing about skin in living people.

What is known, and not known, about absorption

The most cited animal data come from a 2016 rat study, in which a CBD gel was rubbed into shaved back skin once a day for four days at four doses, and CBD was then measurable in blood. At the three lower doses, blood levels rose in a straight line with dose; at the highest dose they did not. The authors applied the gel to the back partly to limit the chance that the animals would ingest it (Hammell et al., 2016; animal, cited here only for its plasma measurements). Rat skin and human skin differ, so these numbers cannot be transferred to people.

Human data are thin. A 2018 systematic review of CBD pharmacokinetics in people found 24 studies, all oral, oromucosal, inhaled or intravenous, and stated that transdermal results had not been published in humans at that time (Millar et al., 2018; review). The 2022 review noted that most CBD skin-delivery data since then come from rats, mice, rabbits, pigs and dogs, that its clearest human skin-patch data were for delta-8-THC rather than CBD, and that a small clinical trial of a pharmaceutical transdermal CBD gel had been reported in people (Mahmoudinoodezh et al., 2022; review). A pharmaceutical gel engineered to cross the skin does not tell us what a cosmetic or everyday formulation does.

The U.S. Food and Drug Administration has said that additional information about the safety and effectiveness of cannabis and its constituents is needed (FDA, 2024; regulatory). How much CBD passes through intact human skin in everyday use has not been well measured, and no conclusions about any product can be drawn from these studies.

Why topical research is harder than oral

With a swallowed capsule, researchers can draw blood over time and build a clear curve. Skin is less tidy.

  • The target is inside the skin. Measuring tissue levels in living people usually means a biopsy or tape-stripping, so studies stay small.
  • Skin varies. Thickness, hydration, oiliness and age differ by person and body site.
  • The vehicle is part of the dose. Changing the carrier changes the result (Casiraghi et al., 2020; human tissue, in vitro).
  • Lab models are stand-ins. Skin from one donor, or from rodents, may not reflect the range of human skin (Casiraghi et al., 2020; Hammell et al., 2016).
  • Dose is hard to standardize. Amount, area and frequency all shift absorption.

These hurdles are why reviews keep calling for carefully designed human studies (Tóth et al., 2019; Mahmoudinoodezh et al., 2022).

Frequently asked questions

Does human skin have cannabinoid receptors? Yes. Human skin sections have shown CB1 and CB2 on nerve fibers, mast cells, macrophages, keratinocytes, hair follicle epithelium, sebocytes and sweat glands (Ständer et al., 2005; human tissue). What each receptor does in living skin is still being worked out.

Does CBD applied to the skin reach the bloodstream? In rats, a CBD gel produced measurable blood levels (Hammell et al., 2016; animal). In people, published measurements for everyday formulations are lacking (Millar et al., 2018; Mahmoudinoodezh et al., 2022; reviews); the answer likely depends on formulation, amount and skin site.

What kind of research would answer the open questions? Studies that apply defined amounts of CBD to human skin, then measure what stays in the skin and what reaches the blood, across sites and formulations. Reviews note that such studies are few (Tóth et al., 2019; Mahmoudinoodezh et al., 2022).

References

  1. Bíró T, Tóth BI, Haskó G, Paus R, Pacher P. (2009). The endocannabinoid system of the skin in health and disease: novel perspectives and therapeutic opportunities. Trends in Pharmacological Sciences, 30(8), 411–420. https://pmc.ncbi.nlm.nih.gov/articles/PMC2757311/ (Review.)
  2. Tóth KF, Ádám D, Bíró T, Oláh A. (2019). Cannabinoid signaling in the skin: therapeutic potential of the "c(ut)annabinoid" system. Molecules, 24(5), 918. https://pmc.ncbi.nlm.nih.gov/articles/PMC6429381/ (Review.)
  3. Sheriff T, Lin MJ, Dubin D, Khorasani H. (2020). The potential role of cannabinoids in dermatology. Journal of Dermatological Treatment, 31(8), 839–845. https://pubmed.ncbi.nlm.nih.gov/31599175/ (Review.)
  4. Ständer S, Schmelz M, Metze D, Luger T, Rukwied R. (2005). Distribution of cannabinoid receptor 1 (CB1) and 2 (CB2) on sensory nerve fibers and adnexal structures in human skin. Journal of Dermatological Science, 38(3), 177–188. https://pubmed.ncbi.nlm.nih.gov/15927811/ (Immunohistochemistry on human skin tissue; receptor location only.)
  5. Oláh A, Tóth BI, Borbíró I, et al. (2014). Cannabidiol exerts sebostatic and antiinflammatory effects on human sebocytes. Journal of Clinical Investigation, 124(9), 3713–3724. https://pmc.ncbi.nlm.nih.gov/articles/PMC4151231/ (In vitro cell line and ex vivo human skin organ culture; cited only for the receptor-pathway finding.)
  6. Stinchcomb AL, Valiveti S, Hammell DC, Ramsey DR. (2004). Human skin permeation of Delta8-tetrahydrocannabinol, cannabidiol and cannabinol. Journal of Pharmacy and Pharmacology, 56(3), 291–297. https://pubmed.ncbi.nlm.nih.gov/15025853/ (In vitro diffusion through donated human skin.)
  7. Casiraghi A, Musazzi UM, Centin G, Franzè S, Minghetti P. (2020). Topical administration of cannabidiol: influence of vehicle-related aspects on skin permeation process. Pharmaceuticals, 13(11), 337. https://pmc.ncbi.nlm.nih.gov/articles/PMC7690861/ (In vitro; human epidermis from a single donor.)
  8. Hammell DC, Zhang LP, Ma F, et al. (2016). Transdermal cannabidiol reduces inflammation and pain-related behaviours in a rat model of arthritis. European Journal of Pain, 20(6), 936–948. https://pmc.ncbi.nlm.nih.gov/articles/PMC4851925/ (Animal study; cited only for plasma CBD measurements after topical gel. Its behavioural findings in rats are not discussed here and do not apply to people.)
  9. Millar SA, Stone NL, Yates AS, O'Sullivan SE. (2018). A systematic review on the pharmacokinetics of cannabidiol in humans. Frontiers in Pharmacology, 9, 1365. https://pmc.ncbi.nlm.nih.gov/articles/PMC6275223/ (Systematic review.)
  10. Mahmoudinoodezh H, Telukutla SR, Bhangu SK, Bachari A, Cavalieri F, Mantri N. (2022). The transdermal delivery of therapeutic cannabinoids. Pharmaceutics, 14(2), 438. https://pmc.ncbi.nlm.nih.gov/articles/PMC8876728/ (Review.)
  11. U.S. Food and Drug Administration. (2024). FDA regulation of cannabis and cannabis-derived products, including cannabidiol (CBD). Content current as of July 16, 2024. https://www.fda.gov/news-events/public-health-focus/fda-regulation-cannabis-and-cannabis-derived-products-including-cannabidiol-cbd (Regulatory.)

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

Article 4 of 7

Cannabinoids and the inflammatory response: what the research has looked at

A neutral map of cell, animal and human studies on cannabinoids, CB2 receptors and inflammatory signaling, their limits, and where the FDA stands.

Inflammation is one of the body's oldest tools, and cannabinoids are among the most-studied plant chemicals. Where the two meet there is a large pile of laboratory papers and a much smaller pile of human studies. This article is a map of that literature, not an argument. Each study paragraph below is labelled with its evidence level so that no finding is read out of context.

What inflammation is

Inflammation is the immune system's response to something it reads as a threat: a germ, a splinter, damaged tissue. Immune cells release messenger substances that widen blood vessels and call in more immune cells, which is why an injured area turns red, warm and swollen. A short-lived response is called acute; one that stays switched on for months or years is called chronic (IQWiG, 2025).

Scientists track inflammation partly by measuring cytokines: small signaling proteins such as TNF-alpha, IL-1 beta and IL-6 that push the response up, and others such as IL-10 that tend to settle it.

Cannabinoid receptors and immune cells

The endocannabinoid system has two well-described receptors. CB1 is most abundant in the brain; CB2 sits mainly on cells of the immune and blood-forming systems (Pacher et al., 2006), which is why researchers have asked whether signaling through it changes immune cell behavior.

Evidence level: review of in vitro and animal work. A 2016 review summarized that CB2 is expressed on most immune cell types; that mice bred without CB2 showed stronger inflammatory responses in most, though not all, of the disease models examined (one sepsis model ran the other way); and that in cell cultures and animal models, synthetic CB2-activating compounds were mostly followed by lower cytokine release. Results with the body's own endocannabinoids were mixed, and the authors stated that CB2 had not yet been successfully targeted to treat any human disease (Turcotte et al., 2016).

Evidence level: mostly in vitro. Cannabidiol (CBD) binds only weakly to CB1 and CB2. Reviews describe it acting instead on other targets such as TRPV1 channels, PPAR-gamma and adenosine and serotonin receptors, mostly in laboratory settings (Atalay et al., 2020; Burstein, 2015).

Three kinds of evidence

Research on this topic comes in three layers:

  • In vitro: cells in a dish. Useful for mechanism; says nothing about a living person.
  • Animal: usually mice or rats with an induced condition; results often fail to repeat in humans.
  • Human: clinical trials, ideally randomized and placebo-controlled. The only layer that can speak to people.

Most of this literature lives in the first two layers.

The key reviews

Three narrative reviews are widely cited.

Evidence level: review, mostly in vitro and animal. Nagarkatti and colleagues (2009) surveyed cannabinoid effects on immune cells and catalogued cytokine changes in mouse macrophages and rodent models of autoimmune and liver conditions. They also noted that effects are not uniform: in some human cell lines THC lowered one cytokine while raising another. The human material in that review consisted of cell-line work, cells taken from patients, a 1994 self-report survey and small earlier clinical reports in multiple sclerosis; it did not include controlled trials of cannabinoids for inflammation (Nagarkatti et al., 2009).

Evidence level: review, mostly in vitro. Burstein (2015) reviewed CBD and its synthetic analogs, covering receptor binding, signaling and gene-expression changes connected to inflammation, built largely on laboratory findings (Burstein, 2015).

Evidence level: review, mostly in vitro and animal. Atalay and colleagues (2020) reviewed CBD's antioxidant and inflammation-related chemistry, drawing mainly on cell cultures and rodent models of organ injury and neuroinflammation; clinical studies are cited chiefly for safety and side-effect data rather than analyzed in depth. The authors also noted reports of raised liver enzymes (ALT and AST) after CBD treatment in some studies (Atalay et al., 2020).

What animal studies have measured

Evidence level: animal only. A 2021 systematic review pooled 26 in vivo studies, all in animals, mostly mice and rats. In most, CBD, cannabigerol or a THC-plus-CBD combination was followed by lower levels of TNF-alpha, IL-1 beta, IL-6 or interferon-gamma; THC alone was not, in the three studies that tested it. No study fully reported randomization and blinding, protocols were too varied for a meta-analysis, no human trials were included, and the authors cautioned that animal findings do not always translate to people (Henshaw et al., 2021).

The few human studies

Human studies are few and small.

Evidence level: human, pilot, n = 10, no placebo. A 2020 pilot gave 10 healthy adults a single 30 mg oral dose of CBD, then drew blood and stimulated their immune cells in the laboratory; TNF output was lower at 90 minutes than at baseline. The study was randomized and blinded between two CBD preparations but had no placebo group, and the authors presented it as pilot data for designing future trials (Hobbs et al., 2020).

Evidence level: human, randomized placebo-controlled, n = 20. A 2017 trial gave 20 adults with Crohn's disease, a long-term inflammatory bowel condition, 10 mg of oral CBD twice daily for eight weeks. Disease-activity scores fell by similar amounts in both groups, with no significant difference; the authors said this might reflect a true absence of effect, the low dose, the small sample or the lack of other cannabinoids (Naftali et al., 2017).

Evidence level: human, systematic review and meta-analysis, mostly observational. A 2026 meta-analysis of 46 studies with more than 54,000 participants found regular cannabis use associated with higher levels of both pro-inflammatory and anti-inflammatory markers at once. In the handful of randomized CBD trials it pooled, the estimate pointed to a slight rise in pro-inflammatory markers, with an uncertainty range that included no change. The authors described the overall pattern as consistent with immunomodulation rather than a uniform pro- or anti-inflammatory shift (Belvederi Murri et al., 2026).

A 2017 National Academies report devoted whole chapters to research gaps, regulatory barriers and methodological problems (NASEM, 2017).

Where the FDA stands

The U.S. Food and Drug Administration has approved one CBD-containing drug, Epidiolex, for seizures associated with Lennox-Gastaut syndrome, Dravet syndrome and tuberous sclerosis complex. It states that no other FDA-approved drug products contain CBD and has sent warning letters to companies marketing CBD products with therapeutic claims (FDA, 2024). The National Center for Complementary and Integrative Health adds that research on cannabinoids for most other conditions is at an early stage and that over-the-counter CBD products may contain more or less CBD than their labels state (NCCIH, 2019).

Why no conclusion about any product follows

Cell and animal work has mapped plausible pathways. Human trials are few, small, short and almost all used oral dosing, and in the largest pooled analysis the direction of effect is not even consistent. Cannabinoid signaling in skin cells has been reviewed in the same way: detailed cell-culture findings, sparse human data (Tóth et al., 2019). None of this tells anyone what a given consumer product does in a given person. The biology is interesting; evidence in humans remains limited.

Frequently asked questions

Is CB2 the only cannabinoid-related receptor on immune cells? No. CB1 and non-cannabinoid targets such as TRPV1 and PPAR-gamma also appear on some immune and skin cells, and CBD interacts mostly with those, according to cell-culture work (Atalay et al., 2020; Tóth et al., 2019).

Why do mouse results often fail to carry over to people? Mouse immune systems, doses, routes and disease models differ from human ones, and many animal studies are not randomized or blinded (Henshaw et al., 2021).

Has the FDA approved any CBD product for inflammatory conditions? No. The only FDA-approved CBD drug is Epidiolex, approved for specific seizure disorders (FDA, 2024).

References

  1. Nagarkatti P, Pandey R, Rieder SA, Hegde VL, Nagarkatti M. (2009). Cannabinoids as novel anti-inflammatory drugs. Future Medicinal Chemistry, 1(7), 1333–1349. https://pmc.ncbi.nlm.nih.gov/articles/PMC2828614/ [narrative review; mostly in vitro and animal data]
  2. Burstein S. (2015). Cannabidiol (CBD) and its analogs: a review of their effects on inflammation. Bioorganic & Medicinal Chemistry, 23(7), 1377–1385. https://pubmed.ncbi.nlm.nih.gov/25703248/ [narrative review; mostly laboratory data]
  3. Atalay S, Jarocka-Karpowicz I, Skrzydlewska E. (2020). Antioxidative and Anti-Inflammatory Properties of Cannabidiol. Antioxidants, 9(1), 21. https://pmc.ncbi.nlm.nih.gov/articles/PMC7023045/ [narrative review; mostly in vitro and animal data]
  4. Pacher P, Bátkai S, Kunos G. (2006). The Endocannabinoid System as an Emerging Target of Pharmacotherapy. Pharmacological Reviews, 58(3), 389–462. https://pmc.ncbi.nlm.nih.gov/articles/PMC2241751/ [review]
  5. Turcotte C, Blanchet MR, Laviolette M, Flamand N. (2016). The CB2 receptor and its role as a regulator of inflammation. Cellular and Molecular Life Sciences, 73(23), 4449–4470. https://pmc.ncbi.nlm.nih.gov/articles/PMC5075023/ [review; mostly in vitro and animal data]
  6. Henshaw FR, Dewsbury LS, Lim CK, Steiner GZ. (2021). The Effects of Cannabinoids on Pro- and Anti-Inflammatory Cytokines: A Systematic Review of In Vivo Studies. Cannabis and Cannabinoid Research, 6(3), 177–195. https://pmc.ncbi.nlm.nih.gov/articles/PMC8266561/ [systematic review; animal studies only]
  7. Hobbs JM, Vazquez AR, Remijan ND, et al. (2020). Evaluation of pharmacokinetics and acute anti-inflammatory potential of two oral cannabidiol preparations in healthy adults. Phytotherapy Research, 34(7), 1696–1703. https://pubmed.ncbi.nlm.nih.gov/32147925/ [human pilot, n = 10, randomized between two preparations, no placebo group]
  8. Naftali T, Mechulam R, Marii A, et al. (2017). Low-Dose Cannabidiol Is Safe but Not Effective in the Treatment for Crohn's Disease, a Randomized Controlled Trial. Digestive Diseases and Sciences, 62(6), 1615–1620. https://pubmed.ncbi.nlm.nih.gov/28349233/ [human randomized placebo-controlled trial, n = 20]
  9. Belvederi Murri M, Guglielmo R, Zizzi A, et al. (2026). Regular cannabinoid use and inflammatory biomarkers: Systematic review and hierarchical meta-analysis. Brain, Behavior, and Immunity, 135, 106517. https://doi.org/10.1016/j.bbi.2026.106517 [systematic review and meta-analysis of human studies, mostly observational]
  10. Tóth KF, Ádám D, Bíró T, Oláh A. (2019). Cannabinoid Signaling in the Skin: Therapeutic Potential of the "C(ut)annabinoid" System. Molecules, 24(5), 918. https://pmc.ncbi.nlm.nih.gov/articles/PMC6429381/ [review; mostly in vitro data]
  11. National Academies of Sciences, Engineering, and Medicine (NASEM). (2017). The Health Effects of Cannabis and Cannabinoids: The Current State of Evidence and Recommendations for Research. National Academies Press. https://doi.org/10.17226/24625 [consensus report]
  12. U.S. Food and Drug Administration (FDA). (2024). FDA Regulation of Cannabis and Cannabis-Derived Products, Including Cannabidiol (CBD). Content current as of July 16, 2024. https://www.fda.gov/news-events/public-health-focus/fda-regulation-cannabis-and-cannabis-derived-products-including-cannabidiol-cbd [regulatory]
  13. National Center for Complementary and Integrative Health (NCCIH). (2019). Cannabis (Marijuana) and Cannabinoids: What You Need To Know. https://www.nccih.nih.gov/health/cannabis-marijuana-and-cannabinoids-what-you-need-to-know [government page]
  14. Institute for Quality and Efficiency in Health Care (IQWiG). (2025). What is an inflammation? InformedHealth.org, updated April 11, 2025. https://www.informedhealth.org/what-is-an-inflammation.html [consumer health page]

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

Article 5 of 7

Full-spectrum, broad-spectrum and isolate: what the words mean

Definitions, the 2026 federal total-THC rule for finished hemp products, terpenes and minor cannabinoids, extraction methods, and why the entourage effect is still debated.

Three words keep appearing on hemp product labels: full-spectrum, broad-spectrum and isolate. They are shorthand for how much of the plant's chemistry ends up in the finished extract. Here is what each means, where the law draws its lines, and why one popular idea attached to them, the "entourage effect," is still an open question.

Three words, one plant

Hemp flowers make their cannabinoids and aromatic compounds in tiny glands called trichomes (Russo, 2011; review). When the flowers are processed, the maker decides how much of that mixture to keep.

Full-spectrum means the extract keeps the plant's broad chemical mix: cannabidiol (CBD) plus smaller amounts of other cannabinoids and terpenes and, by law, no more than a trace of delta-9 THC.

Broad-spectrum usually means a full-spectrum-style extract that has been further processed to bring THC down to non-detectable levels while keeping other cannabinoids and terpenes.

Isolate means CBD has been separated from everything else, typically as a white crystalline powder that is close to pure CBD.

One caution: none of these three terms is defined in the federal hemp statute or on the FDA's CBD page (7 U.S.C. § 1639o; FDA, 2024). They are industry conventions, so the only way to know what a given product contains is a third-party lab report.

Where the law draws the line

The 2018 Farm Bill defined hemp as Cannabis sativa L. and its extracts and cannabinoids "with a delta-9 tetrahydrocannabinol concentration of not more than 0.3 percent on a dry weight basis" (7 U.S.C. § 1639o). For the growing crop, USDA counts "total THC," meaning delta-9 THC plus the THC that would form when THCA is heated, using the formula Total THC = (0.877 × THCA) + THC (7 CFR § 990.1).

In November 2025, Congress rewrote that definition in Public Law 119-37. The amended statute keys on total THC including THCA, and it excludes from "hemp" any final hemp-derived cannabinoid product containing "greater than 0.4 milligrams combined total per container" of THC and other cannabinoids with similar effects, a list the statute asks FDA to publish. A container is the innermost package in direct contact with the product (7 U.S.C. § 1639o as amended; Mayl and Gardner, 2025). Cannabinoids made outside the plant, such as delta-8 THC converted from CBD, are excluded as well (Houck, 2026). The change was scheduled for November 12, 2026; the Congressional Research Service reports that a September 2026 appropriations law moved most of it to December 11, 2026, while the exclusion of synthetic cannabinoids still begins November 12 (CRS, 2026).

Once in force, that per-container cap applies no matter which spectrum word is on the label.

The "entourage effect" is a hypothesis

The phrase was coined in 1998 by Raphael Mechoulam's group to describe how fatty-acid compounds that were inactive on their own appeared to increase the measured activity of an endocannabinoid, 2-AG, in cell and mouse experiments (Ben-Shabat et al., 1998; in vitro and animal). It did not originally refer to plant extracts, and no humans were studied.

In 2011, Ethan Russo extended the idea to the plant, proposing that terpenes and minor cannabinoids might modify how THC and CBD act. He called it plausible but largely untested and asked for controlled experiments (Russo, 2011; narrative review). In 2019 he restated the case, arguing that whole-plant chemical varieties deserve study in their own right (Russo, 2019; review).

Other researchers are not convinced. A 2020 laboratory study, funded by a cannabis company, tested five common cannabis terpenes, alone and with THC or CBD, at human CB1 and CB2 receptors in cultured cells and found no change in binding or signaling, apart from a weak signal for beta-caryophyllene at CB2 (Finlay et al., 2020; in vitro). That is a cell-culture result, not a human one. A 2020 critical review argued that "entourage effect" has become a marketing phrase that outruns its evidence (Cogan, 2020; review). Controlled human comparisons of full-spectrum, broad-spectrum and isolate preparations are scarce, and nothing in this literature supports a claim about what any product does in a person.

Terpenes and minor cannabinoids, as chemistry

Terpenes are the aromatic oils plants make. More than 200 volatile compounds have been reported across cannabis varieties, and the main ones are familiar from elsewhere: limonene from citrus peel, alpha-pinene from conifers, linalool from lavender, plus myrcene and beta-caryophyllene (Sommano et al., 2020; Russo, 2011; reviews).

The minor cannabinoids share one chemical ancestor. The plant first builds cannabigerolic acid (CBGA), then enzymes convert it into the acid forms of THC, CBD and CBC (Gülck and Møller, 2020; Russo, 2011; reviews). Heat removes a carboxyl group, turning CBDA into CBD and THCA into THC. CBG is the neutral form of that parent molecule, usually present at low levels. CBC is made mostly early in the plant's life. CBN is not made by the plant at all; it forms as THC oxidizes with age (Russo, 2011).

How the extract is made

Two solvents dominate. Supercritical CO2 extraction pressurizes carbon dioxide above about 31 °C and 74 bar until it behaves like a liquid solvent; afterward the CO2 turns back into gas and leaves no residue, and the low temperature spares heat-sensitive compounds (Lazarjani et al., 2021; narrative review). Ethanol extraction is simpler and gives high yields, but it also pulls out chlorophyll, which adds a green tinge and off flavor, and stripping it out with activated charcoal cut cannabinoid content by roughly half in one report (Lazarjani et al., 2021). Either method can produce a full-spectrum oil; further refining steps take that oil toward broad-spectrum or isolate. The reviewed literature does not rank one method above the other (Lazarjani et al., 2021).

What the label cannot tell you

Independent tests keep finding gaps between label and contents. In a 2017 JAMA letter, 84 CBD products bought online were analyzed: about 31 percent were labeled accurately for CBD, and THC was detected in about 21 percent (Bonn-Miller et al., 2017; product survey). A 2022 study of 80 CBD oils found 54 percent within 10 percent of the labeled amount (Johnson et al., 2022; product survey). A current, batch-specific certificate of analysis is the only document that turns "full-spectrum" into actual numbers.

FAQ

Does "full-spectrum" mean a product contains THC? It can contain a trace amount. Under current federal law the plant material must test at or below 0.3 percent delta-9 THC, and the amended definition adds a 0.4 mg total THC per-container limit for finished products once it takes effect.

Is the entourage effect proven? No. It is a hypothesis with laboratory support on some points and laboratory evidence against others. Controlled human studies comparing full-spectrum, broad-spectrum and isolate preparations remain scarce.

Does the extraction method decide which "spectrum" a product is? Not by itself. CO2 and ethanol can both yield a full-spectrum extract; the spectrum depends on the refining steps that follow, such as THC removal or crystallization into isolate.

References

  1. 7 U.S.C. § 1639o, Definitions ("hemp"), as amended by Pub. L. 119-37, Div. B, § 781 (2025). Legal Information Institute, Cornell Law School. https://www.law.cornell.edu/uscode/text/7/1639o
  2. 7 CFR § 990.1, Meaning of terms (USDA Domestic Hemp Production Program). Legal Information Institute, Cornell Law School. https://www.law.cornell.edu/cfr/text/7/990.1
  3. Mayl S, Gardner J. New federal restrictions on hemp and hemp-derived products: Top points. DLA Piper, November 18, 2025. https://www.dlapiper.com/insights/publications/2025/11/new-federal-restrictions-on-hemp-and-hemp-derived-products
  4. Houck LK. Changes to "Hemp" Definition Set to Close the Farm Bill Loophole. FDA Law Blog, Hyman, Phelps & McNamara, January 21, 2026. https://www.thefdalawblog.com/2026/01/changes-to-hemp-definition-set-to-close-the-farm-bill-loophole/
  5. Congressional Research Service. Changes to the Statutory Definition of Hemp and Implications for Agricultural Policy (IF13136), updated September 28, 2026. https://www.everycrsreport.com/files/2026-09-28_IF13136_aa36fad7d69289127cbbbafef54e2aede431b2c4.html
  6. U.S. Food and Drug Administration. FDA Regulation of Cannabis and Cannabis-Derived Products, Including Cannabidiol (CBD). Updated July 16, 2024. https://www.fda.gov/news-events/public-health-focus/fda-regulation-cannabis-and-cannabis-derived-products-including-cannabidiol-cbd
  7. Ben-Shabat S, Fride E, Sheskin T, Tamiri T, Rhee MH, Vogel Z, Bisogno T, De Petrocellis L, Di Marzo V, Mechoulam R. An entourage effect: inactive endogenous fatty acid glycerol esters enhance 2-arachidonoyl-glycerol cannabinoid activity. Eur J Pharmacol. 1998;353(1):23-31. https://pubmed.ncbi.nlm.nih.gov/9721036/
  8. Russo EB. Taming THC: potential cannabis synergy and phytocannabinoid-terpenoid entourage effects. Br J Pharmacol. 2011;163(7):1344-1364. https://pmc.ncbi.nlm.nih.gov/articles/PMC3165946/
  9. Russo EB. The Case for the Entourage Effect and Conventional Breeding of Clinical Cannabis: No "Strain," No Gain. Front Plant Sci. 2019;9:1969. https://pmc.ncbi.nlm.nih.gov/articles/PMC6334252/
  10. Finlay DB, Sircombe KJ, Nimick M, Jones C, Glass M. Terpenoids From Cannabis Do Not Mediate an Entourage Effect by Acting at Cannabinoid Receptors. Front Pharmacol. 2020;11:359. https://pmc.ncbi.nlm.nih.gov/articles/PMC7109307/
  11. Cogan PS. The 'entourage effect' or 'hodge-podge hashish': the questionable rebranding, marketing, and expectations of cannabis polypharmacy. Expert Rev Clin Pharmacol. 2020;13(8):835-845. https://pubmed.ncbi.nlm.nih.gov/32116073/
  12. Sommano SR, Chittasupho C, Ruksiriwanich W, Jantrawut P. The Cannabis Terpenes. Molecules. 2020;25(24):5792. https://pmc.ncbi.nlm.nih.gov/articles/PMC7763918/
  13. Gülck T, Møller BL. Phytocannabinoids: Origins and Biosynthesis. Trends Plant Sci. 2020;25(10):985-1004. https://pubmed.ncbi.nlm.nih.gov/32646718/
  14. Lazarjani MP, Young O, Kebede L, Seyfoddin A. Processing and extraction methods of medicinal cannabis: a narrative review. J Cannabis Res. 2021;3:32. https://pmc.ncbi.nlm.nih.gov/articles/PMC8290527/
  15. Bonn-Miller MO, Loflin MJE, Thomas BF, Marcu JP, Hyke T, Vandrey R. Labeling Accuracy of Cannabidiol Extracts Sold Online. JAMA. 2017;318(17):1708-1709. https://pmc.ncbi.nlm.nih.gov/articles/PMC5818782/
  16. Johnson E, Kilgore M, Babalonis S. Label accuracy of unregulated cannabidiol (CBD) products: measured concentration vs. label claim. J Cannabis Res. 2022;4:28. https://pmc.ncbi.nlm.nih.gov/articles/PMC9169299/

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

Article 6 of 7

How to read a Certificate of Analysis

Potency units, the total-THC formula, ND versus LOD versus LOQ, contaminant panels, extract versus finished-product reports, and how to verify the lab and the batch.

A Certificate of Analysis, or COA, is the document that shows what a laboratory actually measured in a batch, rather than what a label says. Here is how to read one.

What a COA is

A COA is a report from a testing laboratory listing the quantities of the substances it measured in a specific sample (Utah Admin. Code R66-35-2, 2024). It is tied to one dated batch, identified by a lot number.

Label checks show why it matters. Of 84 CBD extracts bought online, only about 31 percent were labeled within 10 percent of their measured CBD, and THC was detected in 21 percent (Bonn-Miller et al., 2017). Of 105 hemp-derived topicals, 24 percent of those stating a CBD amount were accurately labeled, and THC was detected in 35 percent, including four labeled "THC-free" (Spindle et al., 2022).

Who issues a COA

Credible COAs come from third-party laboratories accredited to ISO/IEC 17025, the international standard for the competence of testing and calibration laboratories (ISO, 2017). Accreditation means an outside body has assessed the lab and issued a scope listing which tests it may run; it is a finding of competence, not a guarantee that any single result is correct (NIST, n.d.). "Third-party" means the lab has no stake in the grower, processor or brand.

The potency panel

This table lists cannabinoids such as CBD, CBDa, delta-9 THC and THCa in two units that say the same thing: one percent equals 10 mg/g, so 0.23 percent is 2.3 mg/g and 48.99 percent is 489.9 mg/g.

Raw hemp contains mostly THCa, an acid that converts to delta-9 THC when heated (Dussy et al., 2005). Federal hemp rules therefore define total THC as delta-9 THC plus 0.877 times THCa (7 CFR 990.1). The 0.877 is simply the ratio of the two molecules' weights, about 314 to 358, which accounts for the carboxyl group THCa loses when it converts. Under federal law, hemp is cannabis with not more than 0.3 percent delta-9 THC on a dry-weight basis (FDA, 2024; 7 CFR 990.1).

ND, LOD and LOQ

"ND" means not detected, which is not the same as zero. The LOD, or limit of detection, is the lowest concentration a method can reliably distinguish from a blank sample. The LOQ, or limit of quantitation, is higher: the lowest concentration the lab can measure with acceptable accuracy and precision (Armbruster and Pry, 2008). A result between the two is printed as "<LOQ": present, but not measurable with confidence. Good COAs list both limits for every analyte.

The contaminant panels

A full-panel COA compares each measured value against an action limit, usually set by a state, and reports pass or fail.

  • Residual solvents: ethanol, hexane, butane and similar solvents left over from extraction.
  • Heavy metals: lead, cadmium, mercury and arsenic. Cannabis can take up metals such as lead, cadmium and mercury from soil and fertilizer (Montoya et al., 2020).
  • Pesticides: dozens of compounds, many fat-soluble and therefore liable to concentrate during extraction (Montoya et al., 2020).
  • Microbials: yeast, mold and bacteria counts, and organisms such as Salmonella and E. coli.
  • Mycotoxins: toxins produced by certain molds, such as aflatoxins (Montoya et al., 2020); state panels commonly list ochratoxin A as well.

Utah requires all of these, plus the cannabinoid profile, on the product in its final form (Utah Admin. Code R66-35-4, 2024).

Extract COA versus finished-product COA

An extract COA describes the concentrated raw ingredient; a finished-product COA describes what is in the container you buy. Cannabinoid numbers drop sharply because the extract is diluted many times over. Contaminants can run the other way: whatever was in the plant is liable to end up concentrated in the extract (Hazekamp, 2018; Montoya et al., 2020).

A worked example

Three figures from an actual hemp-extract COA, used only as a reading exercise; they describe an ingredient, not any finished product.

  • CBD: 48.99 percent, or 489.9 mg/g. Nearly half the extract's weight is CBD, a sure sign of an extract.
  • Delta-9 THC: 0.23 percent, or 2.3 mg/g.
  • Total THC: 2.32 mg/g, or 0.232 percent.

Check the arithmetic: total THC minus delta-9 THC is 0.02 mg/g, so this extract contains almost no THCa, and both figures sit below the 0.3 percent threshold. The finished-product COA will show far lower numbers per gram, and that is the one that should match the label.

How to verify a COA

  • Lab name: look it up and confirm ISO/IEC 17025 accreditation covering potency and contaminants.
  • Sample ID and batch number: they should match your package. Utah requires the batch number, dates received and completed, and the method for each test on the COA (Utah Admin. Code R66-35-4, 2024).
  • Matrix: "topical" or "distillate" tells you which form was tested.
  • QR code: it should land on the certificate for your batch, not a general page.
  • Completeness: a potency-only report is not a full panel.

What state rules require on the label

Utah requires a cannabinoid product label to carry a scannable barcode, QR code or web address "with an easily located certificate of analysis for the batch identified," plus the batch number (Utah Admin. Code R66-35-5, 2024). Florida requires a scannable barcode or QR code linked to the batch COA from an independent laboratory, plus the batch number (Fla. Stat. 581.217, 2024). Texas requires a batch number, batch date and a URL to the COA, which may be delivered as a QR code (Tex. Health & Safety Code 443.205). Wording varies, but the label should lead you to your batch's certificate. Because FDA has not evaluated unapproved CBD products (FDA, 2024), the third-party COA is the main quality signal consumers have.

References

FAQ

What does "ND" on a COA mean? Not detected: the laboratory did not find the compound above its limit of detection. That is not the same as zero, so check the LOD beside the result.

Why are the CBD numbers on an extract COA so much higher than on a product label? An extract is a concentrated ingredient; a finished product contains a small amount of it. Only the finished-product COA should be compared with the label.

How can I tell whether a COA belongs to my batch? Match the batch or lot number on the package with the one on the certificate, check the dates, and confirm the lab's accreditation. If the QR code leads to a general page, ask the seller for the batch report.

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

Article 7 of 7

A short history of hemp and cannabinoid science

From colonial fiber crop to the isolation of CBD and THC, receptor discovery, the 2018 Farm Bill and the 2026 total-THC rule.

Two stories that grew up together

Hemp is one of the oldest crops in the American story. Cannabinoid science is one of the youngest branches of chemistry and pharmacology. For a long time the two ran side by side, then the law pulled them apart for most of the twentieth century, and only in the past decade have they begun to meet again.

This article walks through that history in plain language: plants, molecules, laboratories and laws. It is a map of what was discovered and when. It is not about what any product does, and it contains no claims about effects in people.

Hemp as a crop: fiber and seed

For most of its history, hemp meant rope, sailcloth, twine and paper. Farmers in the American colonies grew it widely from the 1600s into the mid-1800s, and it supplied both fine and coarse fabrics (Congressional Research Service, 2018). The stalk gave fiber; the seed gave oil and food.

The seed remains a food ingredient today. In December 2018, the U.S. Food and Drug Administration completed its evaluation of three notices concluding that hulled hemp seed, hemp seed protein powder and hemp seed oil are generally recognized as safe for use in human food. These are seed-derived foods, which FDA notes contain only trace amounts of cannabinoids, and the conclusion says nothing about CBD as an ingredient (FDA, 2023).

One point of botany matters throughout. Hemp and marijuana are the same species, Cannabis sativa L. The difference in U.S. law is chemical, not botanical: hemp varieties contain very little THC, the component responsible for intoxication (NCCIH, 2019).

Two legal turns: 1937 and 1970

The Marihuana Tax Act of 1937 brought the first federal restrictions on cannabis and, in practice, on hemp as well. Farmers had to hold a federal registration and a special tax stamp, which effectively capped further expansion of the crop (Congressional Research Service, 2018; Bridgeman and Abazia, 2017).

The Second World War briefly reversed course. The government encouraged hemp for fiber and oil, and the U.S. Department of Agriculture produced a short film titled "Hemp for Victory" to show farmers how to grow and harvest it (USDA National Agricultural Library, c. 1942-1945). In 1943, U.S. production exceeded 150 million pounds on about 146,200 harvested acres. After the war the crop faded, and no production was recorded after the late 1950s (Congressional Research Service, 2018).

The second turn came in 1970. The Controlled Substances Act adopted the 1937 definition and placed every variety of the plant, low-THC or high-THC, in Schedule I, without distinguishing between them (Congressional Research Service, 2018). Reviews of the period note that this made research on the plant practically very difficult for decades (Crocq, 2020; Bridgeman and Abazia, 2017).

Chemists find the molecules

Even as the law tightened, the chemistry moved forward.

In 1940, Roger Adams and colleagues at the University of Illinois isolated cannabidiol, now known as CBD, from an extract of Minnesota wild hemp and proposed a partial structure (Adams et al., 1940). The complete structure and three-dimensional shape were worked out in 1963 by Raphael Mechoulam and Yehiel Shvo in Israel (Mechoulam and Shvo, 1963).

A year later, Yechiel Gaoni and Mechoulam isolated delta-9-tetrahydrocannabinol, or THC, determined its structure and achieved a partial synthesis (Gaoni and Mechoulam, 1964). A cruder THC-containing mixture had been extracted from the plant as early as 1942, but the 1964 paper settled the structure (Pertwee, 2006).

These were chemistry papers. They describe what the molecules are, not what they do in the body.

The body's own system

For another quarter century, no one knew how these molecules acted in the body. That changed in a burst of laboratory discoveries between 1988 and 1995. Each is labeled below by the kind of evidence it is.

In 1988, William Devane, Allyn Howlett and colleagues identified a specific binding site for cannabinoids in rat brain membranes, the first direct evidence of a cannabinoid receptor (animal tissue, in vitro; Devane et al., 1988). In 1990, a team at the National Institutes of Health cloned the gene for that receptor, now called CB1, from rat tissue (molecular cloning, in vitro; Matsuda et al., 1990). In 1993, Sean Munro's group in Cambridge described a second receptor, CB2, cloned from a human cell line and found mainly in immune tissue rather than brain (molecular cloning, in vitro; Munro et al., 1993).

If the body has receptors, it should have its own molecules that fit them. In 1992, Devane, Mechoulam and colleagues isolated such a molecule from pig brain and named it anandamide, by reference to the Sanskrit word ananda, meaning bliss (animal tissue; Devane et al., 1992; Crocq, 2020). The name is an etymology chosen by the discoverers, not a description of what the molecule does in people. In 1995, Mechoulam's group identified a second one, 2-arachidonoylglycerol, in canine gut tissue (animal tissue; Mechoulam et al., 1995).

Together, these receptors and the body's own ligands are now called the endocannabinoid system. A fair reading is also a modest one: these were laboratory and animal-tissue studies that established that the system exists. None of them tested a plant extract or a product in people, and how the system behaves in everyday human life is still being mapped.

Three recent milestones

On June 25, 2018, FDA approved Epidiolex, an oral solution of purified, pharmaceutical-grade CBD, as a prescription drug for seizures associated with Lennox-Gastaut syndrome or Dravet syndrome, in patients two years of age and older at the time of that approval; the label has since been expanded (FDA, 2018; FDA, 2023). That approval rests on the manufacturer's own randomized controlled trials, reviewed by FDA, and applies to that prescription medicine alone. It says nothing about CBD in other forms, and no hemp-derived product sold outside a pharmacy shares that status (FDA, 2023).

On December 20, 2018, the Agriculture Improvement Act of 2018, usually called the 2018 Farm Bill, removed hemp from the Controlled Substances Act definition of marijuana. It defined hemp as Cannabis sativa L. and its derivatives with no more than 0.3 percent delta-9 THC on a dry-weight basis (Public Law 115-334, 2018). FDA kept its authority over foods, dietary supplements, drugs and cosmetics, and has stated that CBD may not be added to food or sold as a dietary supplement (FDA, 2023).

The newest turn is still ahead. Section 781 of a federal appropriations law signed on November 12, 2025, rewrites the definition of hemp, effective November 12, 2026 (Public Law 119-37, 2025). Under the new language, the 0.3 percent threshold applies to total THC, including THCA, rather than delta-9 THC alone; final hemp-derived cannabinoid products are excluded from the definition if they contain more than 0.4 milligrams of total THC per container; and cannabinoids synthesized outside the plant are excluded. FDA is directed to publish lists of naturally occurring cannabinoids and guidance on the meaning of "container" (Wisconsin Legislative Council, 2026).

What this history does and does not tell us

This is a history of chemistry and law, not of clinical evidence. The receptors and ligands were characterized in rat, pig and dog tissue and in cultured cells. The 2018 drug approval concerns one prescription product and two rare conditions. Nothing in this article reports a human study of a hemp-derived or topical product, and no conclusion about any product's effects can be drawn from the story told here. Other articles in this library map individual topics in more depth, with the same evidence-level labels.

References

Adams R, Hunt M, Clark JH. Structure of cannabidiol, a product isolated from the marihuana extract of Minnesota wild hemp. I. Journal of the American Chemical Society. 1940;62(1):196-200. https://doi.org/10.1021/ja01858a058

Mechoulam R, Shvo Y. Hashish I. The structure of cannabidiol. Tetrahedron. 1963;19(12):2073-2078. https://doi.org/10.1016/0040-4020(63)85022-X

Gaoni Y, Mechoulam R. Isolation, structure, and partial synthesis of an active constituent of hashish. Journal of the American Chemical Society. 1964;86(8):1646-1647. https://doi.org/10.1021/ja01062a046

Devane WA, Dysarz FA, Johnson MR, Melvin LS, Howlett AC. Determination and characterization of a cannabinoid receptor in rat brain. Molecular Pharmacology. 1988;34(5):605-613. https://pubmed.ncbi.nlm.nih.gov/2848184/

Matsuda LA, Lolait SJ, Brownstein MJ, Young AC, Bonner TI. Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature. 1990;346(6284):561-564. https://pubmed.ncbi.nlm.nih.gov/2165569/

Munro S, Thomas KL, Abu-Shaar M. Molecular characterization of a peripheral receptor for cannabinoids. Nature. 1993;365(6441):61-65. https://pubmed.ncbi.nlm.nih.gov/7689702/

Devane WA, Hanus L, Breuer A, Pertwee RG, Stevenson LA, Griffin G, Gibson D, Mandelbaum A, Etinger A, Mechoulam R. Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science. 1992;258(5090):1946-1949. https://pubmed.ncbi.nlm.nih.gov/1470919/

Mechoulam R, Ben-Shabat S, Hanus L, Ligumsky M, Kaminski NE, Schatz AR, Gopher A, Almog S, Martin BR, Compton DR. Identification of an endogenous 2-monoglyceride, present in canine gut, that binds to cannabinoid receptors. Biochemical Pharmacology. 1995;50(1):83-90. https://pubmed.ncbi.nlm.nih.gov/7605349/

Pertwee RG. Cannabinoid pharmacology: the first 66 years. British Journal of Pharmacology. 2006;147(Suppl 1):S163-S171. https://pmc.ncbi.nlm.nih.gov/articles/PMC1760722/

Crocq MA. History of cannabis and the endocannabinoid system. Dialogues in Clinical Neuroscience. 2020;22(3):223-228. https://pmc.ncbi.nlm.nih.gov/articles/PMC7605027/

Bridgeman MB, Abazia DT. Medicinal cannabis: history, pharmacology, and implications for the acute care setting. P&T. 2017;42(3):180-188. https://pmc.ncbi.nlm.nih.gov/articles/PMC5312634/

Congressional Research Service. Hemp as an agricultural commodity. Report RL32725, updated July 9, 2018. https://www.everycrsreport.com/reports/RL32725.html

USDA National Agricultural Library, Special Collections. Hemp for Victory [16 mm film, 14 min]. Circa 1942-1945. https://archivesspace.nal.usda.gov/repositories/4/digital_objects/7814

U.S. Food and Drug Administration. NDA 210365 approval letter: Epidiolex (cannabidiol) oral solution. June 25, 2018. https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2018/210365Orig1s000ltr.pdf

Agriculture Improvement Act of 2018, Public Law 115-334, enacted December 20, 2018. https://www.govinfo.gov/app/details/PLAW-115publ334

U.S. Food and Drug Administration. FDA regulation of cannabis and cannabis-derived products, including cannabidiol (CBD). 2023. https://www.fda.gov/news-events/public-health-focus/fda-regulation-cannabis-and-cannabis-derived-products-including-cannabidiol-cbd

Continuing Appropriations, Agriculture, Legislative Branch, Military Construction and Veterans Affairs, and Extensions Act, 2026, Public Law 119-37, enacted November 12, 2025. https://www.govinfo.gov/app/details/PLAW-119publ37

Otis A. Change to federal definition of "hemp." Wisconsin Legislative Council Issue Brief, May 2026. https://docs.legis.wisconsin.gov/misc/lc/issue_briefs/2026/alcohol_beverages_and_cannabis/ib_federal_hemp_legislation_ao_2026_05_19

National Center for Complementary and Integrative Health. Cannabis (marijuana) and cannabinoids: what you need to know. 2019. https://www.nccih.nih.gov/health/cannabis-marijuana-and-cannabinoids-what-you-need-to-know

Frequently asked questions

Are hemp and marijuana different plants? No. Both are Cannabis sativa L. U.S. law separates them by THC content: hemp must stay at or below 0.3 percent, a threshold that shifts from delta-9 THC to total THC in November 2026.

Who first isolated CBD? Roger Adams and colleagues isolated it from Minnesota wild hemp in 1940. Raphael Mechoulam and Yehiel Shvo published its full structure in 1963, and Gaoni and Mechoulam isolated THC the following year.

When were the cannabinoid receptors discovered? A binding site was identified in rat brain membranes in 1988, the CB1 receptor was cloned from rat tissue in 1990, and CB2 was cloned in 1993. The body's own ligands, anandamide and 2-AG, followed in 1992 and 1995, in pig brain and canine gut tissue. All of these were laboratory findings, not human studies.

This article is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question.

Back to the library

This library is for general education only. It is not medical advice and does not describe NATRIAM SOOTHE or any other product. NATRIAM SOOTHE is not intended to diagnose, treat, cure or prevent any disease. Talk to a healthcare professional about any health question. Sources are linked in each article; summaries reflect the cited papers as published.