THC and CBD: Similarities and differences between siblings

Abstract

Δ⁹-tetrahydrocannabinol (THC) and its sibling, cannabidiol (CBD), are produced by the same Cannabis plant and have similar chemical structures but differ dramatically in their mechanisms of action and effects on brain functions. Both THC and CBD exhibit promising therapeutic properties, however, impairments and increased incidence of mental health diseases are associated with acute and chronic THC use, respectively, and significant side effects are associated with chronic use of high dose CBD. This review covers recent molecular and preclinical discoveries concerning the distinct mechanisms of action and bioactivities of THC and CBD, and their impact on human behavior and diseases. These discoveries provide a foundation for the development of cannabinoid-based therapeutics for multiple devastating diseases and assure their safe use in the growing legal market of Cannabis-based products.

In brief:

This review by Nephi Stella discusses the recent molecular and preclinical breakthroughs made in our understanding of THC’s and CBD’s influence on human behaviors and neuropathological processes, and the key results that are fostering the development of transformative cannabinoid-based therapies.

1]. Introduction

The Cannabis plant synthesizes phytocannabinoids (phyto-CB), including the bioactive compounds THC and CBD (Figure 1). THC is often referred to as the principal “psychotropic” compound produced by Cannabis and CBD as the principal “non-psychotropic” compound. Since both compounds influence brain functions, this simple distinction is not valid, and this review addresses the effects of both. The recent legalization of medical and adult use of products that contain phyto-CBs by several countries, including the US, Canada, and the Netherlands, represents a historical shift in policy that was led by patient advocates who use Cannabis products medicinally, and by Cannabis industry advocates. Both groups sought to encourage a legal market that could improve well-being and reduce use of other harmful substances (for example opioids)^1,2. A milestone in the acceptance of CBD as a therapeutic modality was its FDA approval in the treatment of refractory epilepsies. Recent studies suggest promising therapeutic indications for other neurologic diseases where CBD also modulates neuronal function.

Figure 1: Cannabinoid compounds.

A] Cannabis plant. B] Phytocannabinoids THC and CBD (green) produced by the plant, C] THC metabolites 11-OH-THC and THC-COOH (blue) produced by P450 enzymes expressed in liver and brain and D] artificial cannabinoids, CP55940 and JWH-018 (active ingredient in the illegal market product, “Spice”) (red).

Concomitantly, public health professionals have expressed concerns about the increasing acceptance of a substance with significant side-effects, especially THC-associated mental health diseases^1,3. This concern is heightened by the escalation in THC concentrations found in Cannabis plants attributable to increased understanding of Cannabis genetics, improved cultivation methods, and optimization of extraction yields (Box 1)⁴. Taking Washington state between October 2014 and September 2016 as an example, the average levels of THC in extracts reached 75–80%, which is 3 fold more than Cannabis flowers (20.6%) during the same period^5,6. Furthermore, while traditional Cannabis flowers still accounted for most of the sales in Washington State at that time (i.e., 66.6%), the market share of THC extracts for inhalation increased by 150%⁵. Public health professionals in the US have also noted the increasing perception of THC-based products as “safe” recreational drugs, particularly in 16% of individuals ages 12–17 and in 52% of individuals ages 18–25 years (NIDA, News release in 8/22/22). These alarming statistics indicate increased use of products that contain high levels of THC, a scenario that may have important health consequences, particularly on brain function and behavior in vulnerable individuals such as adolescents.

Box 1. Inheritance of cannabinoid compounds synthesized by the plant.

THC concentration in plant: Best explained by both additive and dominance composite genetic effects

CBD concentration in plant: Best explained by cytoplasmic genomes and additive genes^225,330.

Enzymes that synthesize THC and CBD: single-nucleotide polymorphisms in the coding regions determine plant chemotype²²⁵.

Current understanding: Both genetic effects and cytogenetic contributions influence the THC:CBD ratios produced by the Cannabis plant³³⁰.

Outcome: Facilitated effective crop breeding to select cultivars that produce defined amounts of THC and CBD.

The research summarized in this review is discussed in the context of the known differences in the pharmacodynamic (PD) and pharmacokinetic (PK) profiles of THC and CBD. The body of work reviewed provides a foundation for translating research into therapies for a variety of diseases and conditions.

2]. Pharmacokinetics of THC: Differences among routes of administration

Humans administer THC-containing products by smoking, vaporizing and dabbing (a slang term that refers to the process of consuming THC products that come in a variety of textures, including resin); as well as by drinking and eating^6,7. When measuring THC bioactivity, studies need to indicate 1) the route of administrations, 2) how much THC was administered, and 3) the regimen of THC use (e.g., times a day, a week, or a month). Below is our current understanding of the PK profile of THC in humans and rodents, accounting for those three factors. The PK profile of a drug depends on four key physiological events: absorption, distribution, metabolism, and excretion, typically referred to as ADME. Thus, our understanding of how the drug’s concentration changes in selected tissues over time combined with knowledge of how that might correlate with the drug’s activity at selected protein targets (i.e., its PD) provides a clearer picture of its in vivo bioactivity.

2.1]. THC PK in humans.

Inhalation of Cannabis-based cigarettes has historically been the most common route of THC administration by humans. It leads to rapid THC absorption that peaks within minutes in blood and is not subject to first pass liver metabolism⁸. Considering that Cannabis-based cigarette currently average ≈20% of THC/total weight and that a cigarette weigh ≈1 g, then a full cigarette contains ≈200 mg THC⁶. Thus, a human adult of ≈70 Kg who smokes a 1/4 of a cigarette absorbs ≈0.3 mg/kg of THC (considering ≈50% pyrolytic loss because of burning the product). Human PK studies show that one inhalation of THC (0.3 mg/kg) results in vastly variable levels of THC in blood within the initial minutes, depending on the individual: 1.6–160 ng/ml of THC (≈ 5–500 nM)⁹. The hydrophobic nature of THC allows its distribution into most tissues, including the brain¹⁰. Thus, THC passes the blood brain barrier, briefly accumulates in brain parenchyma, and then slowly declines because of THC metabolism and excretion. THC undergoes extensive liver metabolism, and its clearance is restricted by fatty acid binding proteins¹¹. Of note, while over 80 metabolites of THC have been identified, only a handful remain bioactive, including THC’s first metabolite 11-OH-THC (more potent than Δ⁹-THC) and 11-nor-Δ⁹-THC-9-carboxylic acid (less potent than Δ⁹-THC)¹².

Humans are increasingly administering THC through eatables and drinks. In contrast to inhalation, oral THC absorption is slower and maximum THC concentrations are reached 1–3 h after dosing¹³. For example, oral consumption of 15 mg THC (≈200 μg/kg) results in ≈9 ng of THC/mL of blood (≈ 30 nM) after 3h¹⁴. Considering the current average amount of THC in edibles (10 mg/serving), its oral consumption by a human adult will result in ≈20 nM of THC in blood after 3 h. Orally administered THC and its bioactive metabolites decline slowly (half-life 20–60 h for THC), suggesting prolonged exposure and lingering bioactivity even after a single ingestion¹⁴. Thus, the bioactivity of THC fluctuates depending on its route of administration, rate and extent of absorption, metabolism, and excretion, all of which can significantly vary between individuals (Figure 2A–B)¹⁰.

Figure 2: THC metabolism, PK, bioactivity and regiments.

A] Model examples of the PK profiles of THC when inhaled (lungs) or consumed (oral). Examples of resulting behavioral responses associated with low dose THC producing the “high”/euphoria (green arrow), higher dose THC that impairs behavior such as reaction time and judgment (blue arrow), and even higher dose THC (red arrow) that might lead to Cannabis Use Disorder in vulnerable populations, and can be toxic and associated with physiological and psychological panic attack (commonly referred to as “to green-out”; symptoms include sweating, nausea, heart palpitations, hypervigilance, paranoia, and the fear or feeling that you might be dying or about to die). B] Model examples of the PK profile of THC when used to treat pain during the day using several administrations (red line) or to fall asleep with one administration (blue line). A] THC is either metabolized by CYP2C9 that produces 11-OH-THC, which is then metabolized by microsomal alcohol dehydrogenase, or CYP3A4 that produces 8-OH-THC. Conjugation represents the next metabolic step.

2.2]. THC PK in rodents.

As a preclinical model system for PK studies, rodents provide important information but also have key limitations, including a very active liver metabolism compared to humans. The PK and bioactivity of THC in rodents has been studied when administered by i.p., i.v. and s.c. injections, as well as by smoke exposure and oral gavage, routes of administration that allow for precise dosing yet are not used by humans^7,15,16. For example, i.p. injections into adult mice with THC (5 mg/kg) result in 600 pmol/ml of THC (≈ 2 nM) in blood within 15–30 min and 1000 pmol/g of THC in brain (≈3 nM) after 2h^17,18. However, it is important to emphasize that a portion of the THC injected via i.p. will undergo first pass liver metabolism and thus not all the injected THC will reach the brain. Obvious limitations of injections, smoke exposure and oral gavage are the absence of self-administration and the mild stress response by rodents from administration procedures.

Oral self-administration of THC by rodents is easily achieved when formulating THC in either gelatin or dough, and consumption results in THC brain concentrations that are bioactive^19–21. For example, consumption of 2–3 mg THC-gelatin during 1h results in 2–4 ng/ml THC in blood (≈ 6–12 nM)¹⁹. Of note, rodents self-administer less gelatin containing high dose THC compared to control-gelatin, suggesting an avoidance of the taste/odor of THC¹⁹. Inhalation of THC by rodents involves either burning Cannabis cigarettes or vaporizing THC extract in an airtight environment similar to an operant chamber ^18,22,23. THC vapor self-administration by rats is achieved by training them to nose poke for the vapor, which leads to increased motivation to take more THC vapor²⁴. For example, 30 min inhalation exposure of 25 mg/ml results in 40 pmol/ml (≈ 30 nM) in blood that then rapidly decays²⁴.

THC is metabolized by CYP2C3 and CYP3A4, P450 enzymes expressed in liver and brain tissues, which produce similar metabolites in rodents and humans (Figures 2C and 3A)¹⁰. When reaching the brain, THC rapidly crosses the blood brain barrier via ABC transporters P-glycoprotein and breast cancer resistance protein^25–27. A recent study showed an association between an ABCB1 polymorphism in humans and increased THC metabolism and frequency of use; yet there was no difference in the self-reported subjective high associated with smoking THC (12.5% in a cigarette) suggesting a nonlinear relation between THC metabolism and subjective high²⁸. In summary, the development of experimental approaches that better recapitulate human use of THC-containing products, combined with our growing understanding of the molecular mechanisms that control the overall PK profile of THC in rodents, provide initial mechanistic understanding of the link between the PK profile of THC and its bioactivity.

Figure 3: THC inactivation by CYP2C and biased signaling at CB₁R.

A] THC is metabolized to 11-OH-THC by the P450 enzyme CYP2C9. Fatty acid binding proteins (FABP) assist THC in being metabolized. B] THC activates CB₁R that couple to Gα_i proteins that inhibit adenylyl cyclase, Gbg proteins that regulate ion channels and b-arrestin that regulates kinase signaling. Different agonist will often preferentially modulate one of these signaling pathway. FABPassist THC in activating CB₁R.

3]. THC pharmacodynamics: Modulation of multiple targets

THC modulates that activity of multiple G protein-coupled receptors (GPCRs) and ion channels as shown by cell culture studies, and most of these targets mediate THC’s bioactivity in vivo as demonstrated by fully blocking THC’s bioactivity in rodent model systems using genetic deletion and/or antagonism. In this section, emphasis is on the concentrations of THC required to modulate these targets, especially concentrations of THC in the brain.

3.1]. Cannabinoid receptor 1 (CB₁R).

This GPCR was the first receptor shown to be modulated by THC and is the target that THC activates with the highest potency (≈30 nM) (Figure 4A)²⁹. CB₁R mediates THC’s psychotropic effects in humans as shown with the CB₁R antagonist, SR141617³⁰. CB₁R is involved in cell division and neuronal pathfinding during development and remains expressed by most neuronal and glial cell subtypes in the adult brain, although at remarkably different expression levels depending on the cell type. For example, CB₁R is expressed at high levels by GABAergic neurons, at intermediate levels by glutamatergic neurons, and at low levels by microglia³¹. This receptor signals through G_i/o and β-arrestin, and sometimes through G_s in neurons, microglia, and oligodendrocytes, and through G_q in astrocytes (Figure 3B)^32,33. However, coupling of CB1R to different signaling mechanisms may vary with overexpression in heterologous systems³⁴. CB₁R interacts with cytosolic CRIP1a and BiP proteins, and this interaction controls signaling biased toward second messenger pathways in a neuron population-selective manner^35,36. THC activates CB₁R as a partial agonist and modulates neuronal functions by reducing presynaptic neurotransmitter release (e.g., glutamate, GABA and acetylcholine), inducing synaptic plasticity and adjusting energy metabolism and cell phenotype through changes in gene expression³⁷. CB₁R expressed by astrocytes, oligodendrocytes and microglia controls energy metabolism and cell phenotype^38–40. Thus, THC influences the function of most neuronal and glial cells in the brain by modulating their CB₁R cell-specific signaling.

Figure 4: Molecular targets modulated by THC.

A] THC modulates CB₁R and CB₂R that are endogenously activated by 2-AG and AEA, and GLRA3 receptors that are endogenously activated by glycine, as demonstrated by in vivo genetic and antagonist experiments. B] THC modulates GPR55 that is endogenously activated by lysophosphatidyl inositol (LPI) and 5-HT3A receptors that are endogenously activated by serotonin, as suggested by cell culture experiments.

3.2]. Cannabinoid receptor 2 (CB₂R).

CB₂R is expressed at low levels in the healthy brain and predominantly by endothelial cells forming the blood brain barrier^41,42. Its expression is greatly upregulated under select pathological states, (e.g., activated microglia in the mouse model of experimental autoimmune encephalomyelitis)⁴³. While both genetic and pharmacological evidence indicate that CB₂R directly regulates neuronal functions, its expression by neurons remains inconclusive because of the absence of reagents such as selective antibodies⁴⁴. CB₂R signals through G_i/o and β-arrestin in microglia and oligodendrocytes, and sometimes through G_q in endothelial cells^38,45. Thus, THC’s bioactivity mediated by CB₂R includes glial cell activation and endothelial cell function, though it is undetermined whether modulation of neurons is direct or indirect through these cell types (Figure 4A).

3.3]. Endocannabinoid-dependent modulation of CB₁R/CB₂R.

CB₁R and CB₂R are endogenously activated by two signaling lipids, anandamide (AEA) and 2-arachidonoyl glycerol (2-AG), that are produced and inactivated by neuronal and glial cells⁴⁶ (Figure 5). These endocannabinoids (eCB) are produced on-demand in response to select stimuli that increase cellular activity (typically by increasing the calcium-dependent activity of membrane-associated lipases)⁴⁷. AEA is mainly produced by N-acetylphosphatidylethanolamine-hydrolyzing phospholipase D (NAPE-PLD), whereas 2-AG is mainly produced by diacyl glycerol lipase (DAGL)⁴⁸. Thus, AEA and 2-AG are released from plasma membranes and act in an autocrine and paracrine manner to modulate CB₁R/CR₂R signaling.

Figure 5: Endocannabinoid signaling.

A] Focus on a synapse between two neurons with adjacent a microglia and an astrocyte (tripartite synapse). B] Presynaptic release of neurotransmitters (gray dotted arrows) stimulates post-synaptic receptors and activate lipases involved in eCB production (light green arrow). Released eCB activate presynaptic CB₁R and inhibit neurotransmitter release (red line), CB₁R expressed by astrocytes and regulate their energy metabolism, CB₁R/CB₂R expressed by microglia and inhibit cytokine production. THC modulates these receptors. Therapeutic inhibitors that target eCB hydrolysis result in increasing local eCB levels and activity at their targets.

The amount of AEA and 2-AG that reach and activate CB₁R/CR₂R is regulated by at least five mechanisms: 1] enzymes that produce eCBs, 2] eCB transport from their synthesis to their target receptors, which involves fatty acid binding proteins, 3] the number of functional CB receptors at the membrane and their biased signaling, 4] the clearance of eCBs by uptake through the plasma membrane and 5] eCB inactivation by enzymatic hydrolysis. Specifically, presynaptic fatty acid amide hydrolase (FAAH) hydrolyzes AEA, whereas presynaptic monoacylglycerol lipase (MAGL) and post-synaptic α/β-hydrolase domain 6 (ABHD6) hydrolyze 2-AG^49,50. Importantly, these proteins are expressed by neuronal, glial, and endothelial cells; their contribution in controlling eCB signaling is being studied⁵¹. Thus, eCB signaling encompasses CB₁R/CB₂R, AEA/2-AG and the enzymes that produce and inactivate these two main eCBs, exhibit different potencies at CB₁R/CB₂R and provide parallel signaling systems to regulate multiple brain functions.

The PD differences between 2-AG and AEA include their potency and efficacy at CB₁R/CB₂R; 2-AG acts as a low potency full agonist compared to the high potency and partial agonist AEA. Considering that THC has a higher potency at CB₁R/CB₂R and acts as a partial agonist/antagonist, respectively, it may both partially activate free CB₁R/CB₂R and compete for eCB activity at CB₁R/CR₂R⁵². Thus, THC more likely mimics AEA’s bioactivity and reduces 2-AG’s bioactivity. A notable difference is that THC modulates CB₁R/CB₂R signaling in the entire brain as compared to the localized, activity-dependent, increases in AEA/2-AG levels and ensuing activation of CB₁R/CR₂R. Furthermore, CB₁R, and probably CB₂R, represent molecular hubs/coincidence detectors that respond to several endogenous signaling molecules through allosteric binding sites modulating their activity^53,54. Specifically, an allosteric binding site is a distinct protein domain from the orthosteric site that binds the primary endogenously produced ligand and modulates the protein’s activity. Thus, there are both positive and negative allosteric modulators. A remarkable example is pregnenolone, which acts as a negative allosteric modulator of CB₁R; its synthesis is CB₁R-dependent, an endogenous negative feedback loop mechanism thought to protect the brain from CB₁R overactivation⁵⁵. Significantly, this allosteric modulation site on CB₁R represents a very promising target for the treatment of several neurological and psychiatric diseases⁵⁶.

3.4]. Glycine Receptor.

The currents mediated by this ligand-gated chloride ion channels are enhanced by AEA and THC (EC₅₀s = 0.1–1 μM and 3 μM, respectively) and are thought to act through a positive allosteric mechanism (Figure 3)^57,58. Accordingly, the commonly studied THC-induced analgesia is absent in α3GlyR knockout mice and intact in CB₁R/CB₂R knockout mice⁵⁹. This research showed that a commonly studied THC bioactivity, here analgesia, may also be partially mediated by glycine receptors and that the selective targeting of this mechanism may produce specific analgesic benefit distinct from the analgesia mediated through CB₁R and CB₂R (see below). These results also uncover a new coincidence detector mechanism mediated by the synchronized release of glycine and AEA onto cells expressing glycine receptors. Such a coincidence detector mechanism likely contributes to locally inhibiting neuronal activity (Figure 4A).

3.5]. Other possible targets.

THC modulates the activity of at least two additional GPCRs when tested in cells in culture. THC (1–5 μM) is a partial agonist at post-synaptic GPR55 and reduces excitatory transmission (Figure 4B)^60,61. GPR55 is endogenously activated by the lipid signaling molecule, lysophosphatidic acid, and possibly the small peptide PACAP-27^62,63. GPR55 couples to Gq and β-arrestin and is expressed by neurons, astrocytes, and microglia^64–66. Thus, by only partially activating GPR55, THC might prevent endogenous activation of GPR55 and reduce excitatory transmission. Cell culture evidence suggests that AEA decreases and THC increases 5-HT_3AR currents (EC₅₀s ≈ 300 nM and 30 μM, respectively) (Figure 4B)^67,68.

In summary, the strong evidence above shows that the nanomolar concentrations of THC reached in the brain modulate the activities of multiple proteins, including CB₁R, CB₂R and GlyR, and possibly GPR55 and 5-HT_3AR. This poly-pharmacology profile of THC suggests that selective activation of each target will not fully recapitulate THC’s bioactivity, and that selectively modulating a target (excluding CB₁R) may provide specific therapeutic benefits and avoid THC’s associated psychotropic effects (see below).

4]. THC bioactivity: From humans to rodents and back

The landmark discovery of the chemical synthesis of THC and CBD in the 1960s enabled preclinical and human research on its bioactivity⁶⁹. Early studies aimed at comparing the bioactivity of injected THC versus smoked cannabis in humans established that both trigger similar somatic, perceptual, and cognitive changes. Specifically, both smoking a Cannabis cigarette containing THC (12 mg) or injecting THC i.v. (1–6 mg) trigger a rapid succession of symptoms, starting with numbness and tingling of the extremities, light-headedness, “floating” feelings, and loss of concentration, followed by euphoria, palpitation, sweating, tremulousness and weakness that lasts several hours⁷⁰. Subjects also report cognitive impairment described as difficulty in paying attention, difficulty in expressing oneself, mental confusion and loss of sense of time (e.g., one min feels like several min)^70,71 (Table 1). Below are the main behavioral changes commonly measured in rodents exposed to THC, and their relevance to THC’s bioactivity in humans.

Table 1: Psychotropic bioactivity of THC in humans.

Examples of changes in human somatic, perceptual, and cognitive functions triggered by acute THC use.

somatic	perceptual	cognitive
light-headedness	euphoria	introspective states
“floating” feelings	loss of time sense	rapid flow of thoughts
pulse rate increases	increased body	dreamy
palpitation	awareness	loss of concentration
sweating	distortions of vision	disrupted memory
tremulousness	decreased hearing	anxiety
weakness	decreased paying attention	incoordination
numbness	mental confusion	sleepiness
	dizziness	difficulty in thinking
	blurring	difficulty in speaking
	fatigue	difficulty in reading
		difficulty in remembering

4.1]. Cannabimimetic responses in rodents.

Early studies outlined the broad bioactivity of THC (1–30 mg/kg, i.p.) in rodents and showed progressive reduction in spontaneous locomotion; hypothermia; hypersensitivity to tactile and auditory stimuli; ataxia; and sedation, all of which were paralleled by increases in serotonin and noradrenalin levels in the brain^72,73. The first experimental approach commonly used to test the bioactivity of cannabinoids in rodents is referred to as the “classic tetrad behavior test,” which includes four easy-to-measure physiological and behavioral changes: hypothermia, hypolocomotion, analgesia, and catalepsy (Figure 6A)⁷⁴. Most, but not all, cannabimimetic responses are mediated by CB₁R expressed by select cell types, principally glutamatergic neurons and astrocytes (see below)^75–81. Furthermore, THC bioactivity, impairing effects, and therapeutic effects differ significantly depending on the age and sex of mice. For example, THC (5 mg/kg) reduces spontaneous locomotor activity and is anxiogenic in adult mice but not adolescents and exhibits stronger bioactivity in females^17,24,82. These behavioral measures in rodents have evident translational value in understanding THC’s bioactivity in humans.

Figure 6: Examples of THC bioactivity in rodents.

A] “Classic tetrad” behaviors triggered by CB₁R agonists. B] Recent experimental approaches to study THC self-administration. C] Behavioral changes induced by THC in rodents with translational relevance to humans.

Oral consumption and vapor inhalation of THC can be reliably self-administered by rodents (Figure 6B). Voluntary oral self-administration of THC by rats and mice results in significant acute CB₁R-dependent cannabimimetic responses^19–21,83. Rats trained to nose poke for THC vapor show high rates of responding and motivation to take vapor^22,84. Significantly, rats and mice can reliably discriminate between vehicle and THC injections through a CB₁R-dependent mechanism (e.g., i.p. injections of THC (10 mg/kg) versus vehicle injection in a 2-lever drug discrimination test for sucrose reward)⁸⁵.

THC chronic administration results in sex-specific tolerance to its bioactivity, principally because of receptor desensitization dependent on CB₁R phosphorylation by a G protein-coupled receptor kinase and CRIP1A interaction^36,86–88. In rodents, THC tolerance is associated with altered sleep during spontaneous cannabinoid withdrawal, a mechanism mediated by impaired dopamine function^89,90. Thus, rodents will self-administer THC, can discriminate its bioactivity from control treatment, and exhibit increasing tolerance to its bioactivity, providing direct translational value to understanding THC’s bioactivity in humans.

4.2]. Impairing effects triggered by THC.

Impairment is defined as a state or condition associated with marked reduction in an individual’s ability to control physical and mental functions. The importance of measuring impairment is critical. Motor vehicle accidents are a leading cause of mortality across the world. Accordingly, both simulated and real driving studies that performed a double-blind, placebo-controlled testing of THC administration (20 mg in a cigarette) show significant reduction in driving stability as measured by inappropriate line crossings^91,92. Can such impaired behavior be studied in rodents using simplified approaches?

Behaviors in rodents used to measure impairment from THC include locomotion, motor control, reaction time and the startle response to sudden noise. THC affects spontaneous locomotion of mice in an open field through a CB₁R-dependent mechanism that changes with dose: a low dose increases locomotion and a high dose reduces locomotion^93,94. CB₁R KO mice show worse motor coordination than WT mice as measured by the Rotarod performance test^95,96. Impaired motor coordination is thought to result from dysfunction of the cortico-striatal neuronal circuits⁹⁷. CB₁R is expressed at different levels in striatum: higher levels by GABAergic medium spiny neurons and parvalbumin-expressing interneurons, and lower levels by cortical neurons that project to the striatum^98–106. Evidence suggests that CB₁Rs expressed in cortico-striatal terminal and in medium spiny neurons control motor coordination skills. For example, brain slice electrophysiological studies show that cannabinoids dampen cortico-striatal activity through both CB₁R and CB₁R/D2 receptor heteromers located on cortical axon terminals abutting D2 receptor-expressing medium spiny neurons^98,106–110. CB₁R expressed by medium spinal neurons are involved in the initial learning of motor coordination skills measured on a Rotarod⁹⁶. Thus, rodent studies point to CB₁R expressed by the cortico-striatal pathway as playing a prominent role in the control of spontaneous locomotion and motor coordination and show that THC impairs this mechanism.

Startle is defined as a largely unconscious defensive response to sudden or threatening stimuli, such as sudden noise or sharp movement. In mice, THC (10 mg/kg) decreases the startle response as measured by the reflex response to loud acoustic stimulation via CB₁R- and 5HT2A-dependent mechanisms^111,112. Furthermore, THC (10 mg/kg) reduces reaction time tasks in rats as measured by the trial-unique, delayed nonmatching-to-location task, an index of executive function involving working memory and attention^111,113.

A recent human study showed that THC (36 mg, oral) impacts neuronal connectivity in prefrontal cortex (PFC), as measured by functional near infrared spectroscopy (fNIRS), a commonly used neural signature of impairment in humans¹¹⁴. CB₁R is expressed at different levels by distinct neuronal subpopulations in the PFC of rodents: higher levels by GABAergic projecting neurons and interneurons, and lower levels by projecting glutamatergic neurons¹¹⁵. Activation of CB₁R in rodent PFC slice electrophysiology recording shifts the balance between excitation and inhibition towards excitation, and THC administration to mice results in a remarkable up-regulation in CB₁R expression and down-regulation of BNDF in the PFC^116–118. Accordingly, the sensitive balance of glutamatergic and GABAergic transmission within the PFC is particularly vulnerable to THC that down-regulates CB₁R function^119–121. These rodent studies enhance our understanding of a molecular mechanism associated with THC triggered impairment that affects CB₁R signaling in the PFC.

Acute oral consumption of THC (20 mg/kg) impairs memory functions for hours in humans¹²². This finding was validated in a double blind trial that tested 7.5–15 mg oral THC on visual working memory¹²³. In mice, THC impacts spatial reference and working memory in a sex-dependent manner^124,125. Mechanistically, THC disrupts network neuronal spiking patterns and temporal firing synchrony in the hippocampus, a response that is mediated through CB₁R expressed in astrocytes, PKC and COX2 signaling in addition to hyperpolarization activated cyclic nucleotide-gated channels^126–131.

Thus, THC triggers a broad range of well-defined behavioral changes in rodents, some of which are considered close proxies of human behavior and cognitive tasks, including memory processing, motor coordination, and startle responses (Figures 6B–C)^132–135. Importantly, the bioactivity of THC in rodents shifts depending on the dose. For example, while THC (1 mg/kg) increases food consumption and locomotion and has anxiolytic effects, higher dose THC (5 mg/kg) reduces locomotion and produces anxiogenic and aversive effects^24,111,113. Such a biphasic response to THC is also apparent in humans: low doses of THC alter perception, produce euphoria, impair verbal fluency, and impair working memory while higher doses increase anxiety. Rapid-onset high doses trigger schizophrenia-like symptoms in some individuals¹³⁶. In summary, some of the cardinal behavioral impairments triggered by THC in humans can be studied in rodents, and thus have evident translational value.

4.3]. Mental health disorders associated with THC use: epidemiology and preclinical studies.

Chronic use of high amounts of THC impacts brain function and is associated with increased incidence of mental health disorders. I will highlight two predominant disorders: cannabis use disorder (CUD) and psychosis or schizophrenia¹³⁷. The impacts of THC during pregnancy and in the elderly have been recently reviewed and are not covered here^10,138–140.

Based on self-reporting surveys between 2016 and 2017 across the US, Cannabis dependence was the highest among young adults (16.7%)¹. The public health community is concerned that use of high dose THC-base products often increases during adolescence, may contribute to transitioning from experimental use to frequent use, and may increase impulsive/risky behaviors^141–145. CUD is characterized by impaired memory processing, decision-making, school/work performance, and social functioning. Chronic use of high concentration THC product is associated with progression to first CUD symptoms in adolescents^146,147. Adolescents with CUD also have predicted lower intelligence quotients and slower cognitive function as measured by full-scale IQ and reaction time testing¹⁴⁸. The largest online drug survey world-wide, the Global Drug Survey, along with the US National Comorbidity Survey database, indicate that acute Cannabis use can trigger psychotic symptoms as defined by the occurrence of hallucinations and/or paranoia in select individuals, including young adults who regularly use high amounts of Cannabis and individuals diagnosed with psychosis^149,150. Correspondingly, i.v. injection of THC (2.5 and 5 mg) produces a broad range of transient symptoms and cognitive deficits in healthy individuals that resemble some aspects of endogenous psychoses^136,151. Furthermore, daily use of Cannabis-based products with high THC increases the risk of developing psychotic disorders, including schizophrenia, and is related to earlier onset of symptoms compared to people who do not use Cannabis^152–155. Thus, daily use of products with high THC levels during adolescence may have a strong impact during adolescence itself and later in adulthood in some individuals, and, in particular, may increase the incidence of CUD, psychosis, and schizophrenia^153,156.

A recent flurry of results gathered in rodents outline the impact of THC use on adolescent rodent behavior and on later adulthood behavior. Administration of THC (3 mg/kg, i.p. daily for 3 weeks) to adolescent rats is associated with anxiety-like and impaired locomotor behavior in adulthood, indicated by increased repetitive and compulsive-like behaviors measured by the Nestlet shredding task¹⁵. Confirming and extending this result, administration of THC (0.3–3 mg/kg, twice daily for 10 days) to adolescent rats is associated with anxiety-like behavior in adulthood as measured by the elevated plus maze, increased spontaneous locomotion in open-field, and no change in attentional set-shifting performance¹¹⁶. Administration of adolescent mice with THC (3–8 mg/kg, daily for 3 weeks) impairs recognition memory, working memory, and novel object recognition in adulthood^15,157,158. THC exposure during adolescence does not influence adult nicotine self-administration, extinction, and reinstatement, nor does it influence conditioned place preference to THC itself, pre-pulse inhibition or attentional set-shifting performance in adulthood^116,159,160. In a rat model of anxiety-like behavior, acute intra-PFC infusions of THC produce anxiogenic effects while producing no observable impairments¹⁶¹. Finally, glial cells are also involved in the adolescent brain response to THC. For example, microglia-mediated phagocytosis of newborn cells during development is controlled by eCB signaling and sculpts sex differences in juvenile rodent social play¹⁶². A recent study based on analyzing microglia phenotype by RNA sequencing showed that daily low dose administration of THC to adolescent mice markedly changed microglia activation and function, and impaired the stress behavior caused by social defeat in repeated resident-intruder encounters, an experimental paradigm that produces a depression-like phenotype, anxiety, and social avoidance¹⁶³. Thus, the influence of THC on the developing brain appears to impact multiple neuronal and glial cell functions that affect adulthood behaviors^39,164.

Whether THC acts as a gateway drug, predisposing individuals to want to take ‘harder’ drugs such as opiates or psychostimulants, has been a longstanding debate. CB₁Rs are densely expressed in brain areas involved in reward, addiction, and executive function, including the amygdala, PFC, nucleus accumbens and VTA (Figure 7)¹⁶⁵. Because adolescence represents a vulnerable time of marked neuronal development, it is possible that THC use during this period changes an underlying neurophysiology that makes drugs that are already reinforcing even more rewarding. Several rodent studies proved this hypothesis correct. Adult rats show increased heroin self-administration when exposed to THC during adolescence (1.5 mg/kg, 8 injections over 21 days), a response that relies on increased pro-enkephalin expression in the nucleus accumbens^166–168. Adult rats also show increased cocaine self-administration when exposed to THC during adolescence (1 mg/kg, i.p., 18 days)¹⁶⁹. These effects of THC are likely mediated by CB₁R. Indeed, CB₁R is involved in the development of the rewarding properties of psychostimulants mediated by dopamine, including amphetamine and cocaine, as shown by CB₁R genetic deletion and SR1 treatment^170–173. The dopaminergic system that originates in the midbrain and projects to select areas of the forebrain to regulate motivated behaviors is particularly sensitive to chronic THC exposure, and its dysfunction is associated with drug addiction^19,174. THC exposure (2.5–10 mg/kg, daily for 10 days) during adolescence in rats affects the functionality of dopamine receptors and dysregulates glutamate receptor expression in adulthood^175–178. These results paint a grim picture in which THC exposure in rats during adolescence delays the maturation of the dopamine and glutamatergic neurons and impairs their ability to control motivated and reward-triggered behaviors.

Figure 7: CB₁R-dependent impact of THC on the developing brain during adolescence.

A] CB₁R are expressed by progenitor cells and their activation by THC can trigger apoptoptic cell death, and impair the development of radial glial cells and postmitotic neurons. B] THC activation of CB₁R expressed by mature neurons affects their phenotype and neurotransmitter release. C] CB₁R expressed by mature pyramidal neurons and inhibitory interneurons control the excitatory/inhibitory transmission balance in several brain areas, including prefrontal cortex (PFC), nucleus accumbens (Nac), ventral tegmentum area (VTA). THC also interacts with 5HT1A receptors in the dorsal raphe nucleus (DRN). D] PFC, Nac and VTA in human adolescents undergoes significant development and maturation into adulthood.

5]. THC therapeutic indications: Favorable evidence

Despite the negatives discussed above, THC shows promising therapeutic efficacy. Considering the multiple pathways that THC modulates, an understanding how each target might be involved in THC’s therapeutic response will enable the development of selective small molecules with improved therapeutic index compared to THC. Below I outline preclinical evidence for use of THC in the treatment of pain, sleep disorders, seizures, multiple sclerosis, and Huntington’s disease.

Probably the original and oldest therapeutic use of THC was for its analgesic properties in chronic pain^179,180. We now know that acute inhalation of THC (0.5–1 mg) produces greater analgesia compared to placebo in chronic pain patients and causes only mild side effects that resolve spontaneously¹⁸¹. Based on our understanding of THC’s molecular and cellular mechanism of action, preclinical studies are optimizing this analgesic activity^180,182. The antinociceptive efficacy of THC in rodent models of both pathological and injury-related persistent pain are well established; in addition, the expression pattern of most of its targets in pain pathways have been precisely mapped. Furthermore, the opioid-sparing effects and antinociceptive synergy with nonopioid analgesics have been described in detail, and co-administration of heroin and THC (50 mg/ml) by inhalation produces additive effects on thermal nociception in rats^180,183,184. Mechanistically, oral THC (≈ 3 mg/kg) reduces neuropathic pain in mouse injury via a CB₁R and through CB₂R in the mouse model of chemotherapy-induced neuropathy without the development of tolerance^20,83,185. Of particular relevance for the therapeutic relevance of this mechanism, rats with nerve injury self-medicate with CB₂R small molecule agonists to attenuate a neuropathic pain state¹⁸⁶. Thus, THC induces analgesia by interacting with at least 3 targets: CB₁R, CB₂R and GlyR, and if this depends on the drug or experimental paradigm used selective, emphasizing that the selective targeting these molecular components involved in pain sensation represents independent therapeutic approach. Dedicating resources to the optimization of cannabinoid-based analgesics is particularly relevant to the goal of reducing the use of opioid-mediated analgesics that are addictive and that represent a current major health problem¹⁸⁷.

The effect of THC on sleep quality has been explored since increased sleepiness is often reported following THC use. Low dose THC enhances subjective sleep quality, which is measured by the insomnia severity index via a clinically validated questionnaire, whereas high dose THC increases insomnia symptoms^188,189. One consistent finding is the short-term benefit of THC to treat sleep apnea, most likely due to its modulatory effect on serotonin transmission in key brain areas¹⁹⁰. However, chronic use of Cannabis is associated with rapid habituation to its sleep inducing and enhancing properties, and represents a primary reason for the lapse/relapse of using Cannabis¹⁹¹. Specifically, abrupt Cannabis use cessation among heavy users impairs sleep, may sabotage attempts to quit Cannabis use and raises the risk of relapse and emotional behavior (e.g., irritability and anhedonia)^192–194. One of the most intriguing results is found in individuals suffering from post-traumatic stress disorder who experience both emotional and behavioral symptoms from previous traumatic events. Here THC (5 mg/oral) improves subjective measures of sleep quality and reduces the frequency of nightmares in 70% of the subjects^195–197. Thus, THC may decrease sleep latency initially but could impair sleep quality long-term, depending on the individual^191,194. Studies on the effect of THC on sleep in rodents show that mice exhibit altered sleep during spontaneous THC withdrawal through a CB₁R-dependent mechanism and changes in striatal dopamine signaling^89,198. CB₁R-KO mice spend more time awake during the dark (active) period but not during the light (rest) period, thus enhancing the day-night variation of wake-sleep hours compared to wildtype mice¹⁹⁹.

There is also evidence that THC may be useful for the treatment of seizures, as shown in several rodent models of epilepsy. THC (1–10 mg/kg) reduces both electrically and chemically induced seizures in rats through a CB₁R-dependent mechanism^200,201. Accordingly, THC and CB₁R agonists reduce excitatory transmission and hippocampal overexcitation known to causes seizures^{127,202–204}. Of note, CB₁R expression is both upregulated and involved in the development of rodent models of seizures, suggesting increased therapeutic efficacy when selectively targeting this mechanism. However, considering THC’s side effects, few studies followed up on the therapeutic efficacy of THC as an antiepileptic drug, and the field moved to targeting the eCB signaling and the allosteric site of CB₁R^205,206. For example, inhibition of MAGL increases 2-AG levels and reduces seizures through a CB₁R-dependent mechanism in chemically induced seizures in rats^207,208. Treatment of a rat genetic model of childhood epilepsy with the small molecule allosteric CB₁R modulators, GAT211 and GAT229, reduces in vivo cortical spikes and wave discharges, a measure of absence seizures that are characterized in humans by staring blankly into space for a few seconds²⁰⁹.

Several clinical trials of THC for the treatment of multiple sclerosis (MS) showed limited therapeutic efficacy at reducing spasticity and pain^210,211. However, n a mouse model of MS (experimental autoimmune encephalomyelitis; EAE) activation of CB₁R expressed by neurons and CB₂R expressed by immune cells reduced disease severity^212–215. The therapeutic promise of targeting CB₂R is underscored by a 200-fold upregulation of its expression in microglia in the EAE mouse model⁴³. Thus, preclinical studies suggest that the mild therapeutic properties of THC will require optimization, for example by testing CB₁R/CB₂R allosteric modulators.

THC was tested therapeutically for Huntington’s disease (HD) as part of a clinical trial and showed no significant symptomatic effects when prescribed at 3 mg/kg administered via sublingual spray for 12-weeks²¹⁶. This lack of therapeutic efficacy should be interpreted in the context of the pronounced down-regulation of CB₁R expression in striatum, a predominant brain structure that undergoes neurodegeneration in HD¹⁰¹. CB₁R expression is also downregulated in multiple genetic mouse models of HD, and CB₁R genetic rescue prevents the loss of striatal synaptic contacts^95,217–219. THC exhibits only small therapeutic efficacy in mouse models of HD, indicating that either THC treatment requires optimization or that we need to consider targeting eCB signaling^220,221. For example, MAGL expressed in astrocytes controls mutant huntingtin-induced damage of striatal neurons as shown by the therapeutic efficacy of MAGL inhibition²²⁰. Remarkably, THC treatment is associated with significant neuroprotection in mouse models of HD through a CB₁R/CB₂R-independent mechanism of action; a result that is encouraging and prompts further investigations²²².

In summary, human trials and preclinical rodent models of chronic pain suggest promising therapeutic properties of THC that would benefit from optimization, but preclinical rodent models of insomnia, epilepsy, MS, and HD suggest that THC therapy would require significant optimization, most likely focusing on non-CB₁R-dependent mechanisms.

6]. CBD PK/PD: Similarities to and differences from THC

Since the legalization of CBD-containing products by several countries, including the US in 2018, there has been a striking increase in the human use of CBD-containing tinctures, drinks and edibles^223,224. Selective cross breeding of Cannabis has resulted in a plant producing high levels of CBD and low levels of THC (below 0.3%)²²⁵. In the US, this plant is now legally referred to as hemp. Concomitantly, peer-reviewed studies have reported behavioral and therapeutic effects of CBD in preclinical and human trials and have increased our understanding of the behavioral changes induced by CBD administered as a single agent, its mechanism of action and its therapeutic promise. Again, to measure CBD bioactivity in humans, studies need to provide information on route of administration, how much CBD was used, and its regimen of use. Our understanding of the PK profile of CBD, and the multiple protein activities that it modulates provides a clearer picture of its bioactivity in vivo.

6.1]. CBD PK in humans and rodents.

The PK profile of CBD in humans when administered by different routes of administrations has only recently been studied in detail. In humans, inhaled and oral CBD (10 mg) results in blood levels reaching maximal concentration values of 105 and 14 ng/ml, respectively (≈330 and 45 nM); values that are comparable to those obtained with inhaled and oral THC (10 mg)²²⁶. Oral CBD undergoes first hepatic metabolism that forms 7-COOH-CBD with a more pronounced and prolonged exposure profile than inhaled CBD²²⁶. Whether 7-COOH-CBD remains bioactive is unknown. The therapeutic potential of oral CBD formulations is limited by poor bioavailability and extensive first-pass hepatic metabolism. For example, following oral administration, 7-COOH-CBD blood concentration is ~40-fold higher than CBD²²⁷. Significantly, inhalation as a route of administration has several advantage compared to oral: it bypasses the PK variability attributed to irregular gastrointestinal absorption and first-pass hepatic metabolism, and efficiently delivers CBD into systemic circulation²²⁷. In rodents, i.p. and oral gavage delivery of CBD results in a PK profile that parallels the PK profile of THC^17,228. Because of the similarity of its chemical structure to THC, CBD is also metabolized by CYP3A4, and thus CBD competes for THC metabolism by this enzyme and augments THC levels by reducing its conversion to 11-OH-THC²²⁹. Of note, the benzodiazepine, clobazam, is also metabolized by CYP3A4, and thus CBD treatment augments the exposure of clobazam; and yet this molecular interaction does not significantly affect the bioactivities of CBD and clobazam (see below)^230,231. In summary, the PK profile of CBD delivered orally and i.p. parallels the PK profile of THC in both humans and rodents, but co-administration of CBD with THC and other select medications impacts their respective PK profiles¹⁴.

6.2]. CBD modulates the activities of multiple proteins.

CBD exhibits an even broader poly-pharmacology profile than THC, with only few common targets. The discussion below starts with those targets validated by genetic deletion or antagonism in vivo and ends with those that were reported using heterologous expression systems, systematically emphasizing the concentrations of CBD required to modulate these targets in the context of the concentrations it reaches in brain.

CBD modulates the activity of several GPCRs. Early studies established that CBD binds poorly to the orthosteric site of CB₁R, and yet it antagonizes CB₁R signaling (Figure 8)^29,232. Studies of CB₁R tertiary structure and signaling suggest that CBD is a negative allosteric modulator of CB₁R, most likely by binding to its N termini, one of the longest among class A GPCRs^53,233. Specifically, the current structure-based model of CB₁R transition between inactive and active state posits that CBD might bind to the N termini of CB₁R, resulting in a change in the 3 dimensional structure of the orthosteric binding site and in THC’s and 2-AG’s potencies²³³. In line with this model, CBD (0.1–1 μM) reduces the efficacies and potencies of both 2-AG and THC by activating CB₁R signaling in neurons in cell culture, and inhibits eCB-mediated synaptic plasticity without influencing excitatory transmission^234–236. While these results suggest that CBD inhibits CB₁R signaling by directly interacting with an allosteric binding site on this target, a likely mechanism for some of CBD’s therapeutic properties and diminishing the impairing consequences linked to CB₁R activation, direct demonstration of CBD binding to CB₁R is still needed (see below).

Figure 8: Molecular targets modulated by CBD.

A] CBD is a negative allosteric modulator of CB₁R that is endogenously activated by 2-AG and AEA, an antagonist of GPR55 that is endogenously activated by lysophosphatidyl inositol (LPI) and activate 5HT1A that is endogenously activated by serotonin. B] CBD is an agonist at TRPV1 that is activated by eicosanoids (Eico) and changes in H⁺, a positive allosteric modulator of GABA_AR that is activated by GABA and endogenously allosterically modulated by 2-AG, and a blocker of ENT-1 transporters that carry adenosine.

CBD (1–5 μM) is a partial agonist at GPR55, a GPCR involved in the regulation of cell division and neuronal pathfinding during development, and regulates neurotransmission in the adult brain (Figure 8)^237,238. By partially activating GPR55 and antagonizing its endogenous activation by lysophosphatidyl inositol, CBD most likely increases GABAergic inhibitory neurotransmission in the hippocampus and dampens hippocampal overexcitation in epilepsy^61,239–241 (Figure 9). In vitro experiments suggest that CBD antagonizes 5-HT_1A receptors; but only when reaching 100 μM, a surprising result considering that some of CBD’s bioactivity is mediated through this GPCR in vivo (see below)²⁴².

Figure 9: LPI-GPR55 Endogenous signaling.

LPI is produced by PLA1/PLA2, activates GPR55 and is inactivated by LysoPLA and LysoPLD. CBD antagonizes GPR55, as well as the antagonist CID160200465.

Cell culture evidence suggests that additional GPCRs are modulated by CBD. Computational and cell culture functional validation suggest that CBD (10 μM) antagonizes orexin 1 receptor of type 1 (OX1R), a key regulator in arousal and the sleep/wake cycle, as well as motivation and reward processes²⁴³. CBD acts as a negative allosteric modulate at mu- and delta-opioid receptors, though this effect occurs at 30–100 μM²⁴⁴. Considering that CBD reaches at maximum of 3 μM in the brain following 120 mg/kg i.p. injections, the involvement of these targets in mediating CBD’s bioactivity still requires in vivo validation²²⁸.

CBD modulates the activity of ion channels and protein transporters (Figure 8). Of relevance to the sedative response to high dose CBD, we now have a good understanding of the molecular mechanism by which CBD increases GABA_AR currents. Several endogenous mechanisms increase tonic GABA_AR currents through allosteric modulation, including neurosteroids and 2-AG, and thus contribute to enhancing inhibitory neurotransmission^49,245. Cell culture evidence shows that 2-AG (0.3–30 μM) and CBD (3–30 μM) allosterically increase GABA_AR currents by 4-fold though a mechanism that does not involve the benzodiazepine and neurosteroid binding sites^{230,246–248}. While the resolution of the GABA_AR crystal structure outlined an allosteric site in hydrophobic tails that is distinct from the site engaged by neurosteroids and possibly cholesterol, whether CBD acts by binding to this site has not been directly tested^249,250. CBD potentiation of tonic GABA_AR currents could also be responsible for dampening hyperexcitation and seizures and may cause sedation at high concentrations²⁵¹. These studies suggest that small molecules that directly target allosteric sites on GABA_AR engaged by 2-AG and possibly CBD, or by inhibiting 2-AG hydrolysis, (e.g., with ABHD6 inhibitors) represent promising therapeutic approaches to reduce enhanced excitatory transmission and treat select neurological diseases, including epilepsy⁴⁹.

CBD, acting through transient receptor potential cation channel subfamily V member 1 (TRPV1), is anxiolytic in rats, and analgesic in a rat model of acute inflammation (Figure 8)^252,253. Cell culture studies suggest that CBD activates and desensitizes TRPV1²⁵⁴. Computational approaches suggest that upregulation of TRPV channels might unmask CBD bioactivity at TPRV2, TRPV3, TRPV4, TRPA1, and TRPM8^255,256. While the molecular mechanism by which CBD might regulate TRP channels remains unknown, better understanding of its TRPV1-mediated bioactivity in vivo represents a growing area of study.

CBD is a competitive inhibitor of the equilibrative nucleoside transporter-1 (IC₅₀ < 200 nM), a primary inactivation mechanism for the signaling function of extracellular adenosine that is expressed by neurons and glia^257–259. Thus, inhibition of this transporter by CBD blocks adenosine reuptake by neurons and glia increases local adenosine levels and its ensuing activity at its receptors, A2a. Through this mechanism of action, CBD influences several behaviors, including reducing anxiety (see below). Cell culture studies suggest that CBD (3–10 μM) inhibits currents mediated by the voltage-gated sodium channel Nav1.4 and the voltage-dependent anion channel 1 in mitochondria^260–262.

Thus, CBD modulates the activity of several GPCRs, ion channels and transporters at concentrations that are within the range of the concentrations reached in the brain when treated in vivo. The involvement of most of these targets in the bioactivity CBD exhibits in vivo remain to be demonstrated but should guide the development of novel therapeutics.

7]. CBD bioactivity in humans and rodents and back

CBD has been considered non-psychotropic because, in humans, oral consumption of CBD up to 300 mg and i.v. injection of CBD up to 30 mg are both felt as being inactive^263,264. While recent human studies confirmed that even high doses of oral CBD do not cause THC-like or Cannabis-like effects²⁶⁵, additional studies showed that low and high dose CBD does influence human behaviors and thus CBD should be considered psychoactive. For example, inhaled CBD (12.5 mg) enhances verbal episodic memory in healthy young participants²⁶⁶.

CBD also influences rodent behavior when administered as a single agent and at low dose. CBD (10 mg/kg, i.p.) impairs memory reconsolidation in rats through a CB₁R-dependent mechanism and possibly by reducing Zif268/Egr1 expression, a transcription factor proxy for synaptic plasticity related to reconsolidation^267,268. CBD (20 mg/kg, i.p.) affects motor behavior in rats as indicated by reduction in vertical activity (i.e., rearing which is a context-sensitive behavior) through a 5-HT1A-dependent mechanism^269,270. Acute infusions of CBD (100 ng over 5 min) into the PFC in rats impairs attentional set-shifting and spatial working memory, without interfering with anxiety or sociability behaviors¹⁶¹. Mechanistically, single and repeated exposure to CBD (5–30 mg/kg, i.p.) differently modulate BDNF expression and signaling in rat cortex and striatum²⁷¹. Thus, rodent studies emphasize that CBD influences key brain functions, including motor behavior and memory reconsolidation, and emphasizes its psychoactivity at low dose.

7.1. CBD therapeutic indications.

The successful clinical trial demonstrating CBD’s antiseizure properties in young adults with refractory seizures represents yet another landmark in the study of its bioactivity. Specifically, a multinational, randomized, double-blind trial of adjunctive CBD (300 mg/kg, oral) versus placebo showed significant reduction in seizures in adolescent patients with Dravet syndrome and Lennox-Gastaut syndrome and caused only mild side-effects^240,272. The early finding that CBD metabolism competes with clobazam metabolism at CYP3A4 led to speculation that the antiseizure efficacy of CBD may simply reflect CBD augmenting clobazam exposure and intrinsic activity; however, human studies indicated that CBD reduces seizure frequency both with and without concomitant clobazam, and that these two antiepileptic treatments appear to produce additive benefit^231,273. Thus, Combination Therapy treatments using CBD and low dose clobazam represents a promising approach to control seizures while reducing the side effects linked to clobazam^{231,273–276}.

Rodent studies provided an initial mechanistic understanding of the anti-seizure qualities of CBD. CBD (30–100 mg/kg) reduces epileptiform and seizure responses as measured by electrophysiological recordings of rodents’ hippocampal slices and dampens seizures in rodent model systems of chemically induced seizures as well as in Scna1^+/− mice, a genetic mouse model of Dravet Syndrome^{239,277–279}. Furthermore, CBD potently inhibits CYP3A4 mediated metabolism of clobazam in Scn1a^+/− mice. CBD-clobazam combination treatment achieves greater antiseizure responses, but only when an anti-seizure dose of CBD is used²³⁰. Thus, in this preclinical mouse model of Dravet syndrome, CBD and clobazam produce additive effects, a dual mechanism likely to involve competition for metabolism and independent potentiation of GABA_AR, results that are guiding the design of future human clinical trials²³⁰. Of note, while CBD use as single agent at high concentrations does not produce side effects in the clinical trial setting, the effects of its use in combination with other medications and their mechanistic interactions remains largely unknown.

Two additional therapeutic opportunities with CBD may be in anxiety and sleep quality²⁸⁰. Early evidence in double-blind randomized design trials suggested that CBD (400–600 mg, oral, acute) decreases anxiety as measure by survey and functional neuroimaging (i.e., regional cerebral blood flow at rest) in naive patients with generalized social anxiety disorder^281,282. However, several trials challenged this result, including a randomized double-blind, placebo-controlled trial that tested acute oral CBD (600 mg) and measured emotional processing and neuroimaging²⁸³. Remarkably, lower dose CBD might represent a more promising approach as suggested by a single patient case reports and small retrospective study with CBD (18–25 mg, oral, chronic) reporting reduced sustained anxiety and improved sleep^284,285. Based on this premise, CBD treatment might help with anxiety and sleep, but significant optimization besides dose adjustment will be required. Preclinical evidence in rats might provide important clues as direct injection of CBD into the PFC induces antidepressant-like effect, and injection into the periaqueductal gray blocks panic-like response by activating 5-HT1A receptors^{270,286–288}.

CBD appears to have analgesic properties, at least when used at high dose. In humans, CBD (300 mg/oral/daily) prevents acute and transient chemotherapy-induced peripheral neuropathy, but CBD (30 mg/kg) does not affect pain linked to irritable bowel syndrome^289,290. In line with these human results, acute CBD in rodents significantly attenuates pain-associated behaviors in neuropathic pain models, but yields mixed results in inflammatory pain models¹⁸². CBD (0.3–30 mg/kg, i.p.) effectively reverses mechanical and thermal allodynia, hyperalgesia, and anxious behaviors in a rat neuropathic pain model, a response that involves CB₁R and TRPV1²⁹¹. Furthermore, the GRP55 antagonist, CID16020046, exhibits promising antinociceptive properties in a rat formalin test, suggesting the involvement of a CBD-GPR55 molecular interaction^61,292. Thus, CBD has evident analgesic properties that involve multiple protein targets (Figure 8).

What are the next most promising therapeutic indications for CBD? Monitoring practice-based evidence in patients can provide clues for new therapeutic approaches. The first striking example is autism spectrum disorder, which is characterized by persistent deficits in social communication; restricted and repetitive patterns of behavior, interests, or activities; and often intellectual disabilities^293,294. CBD (4–7 mg/kg/day for 6–9 months) is associated with some level of improvement in attention deficit, hyperactivity, communication deficits, social interaction deficits and cognitive deficits, as well as sleep, a prevalent co-morbidity of this syndrome²⁹⁵. CBD (600 mg/oral) alters regional low-frequency activity and functional connectivity in the brain of adults diagnosed with autism spectrum disorder as measured by fMRI signals²⁹⁶. A telling example provided by preclinical studies is that CBD bioactivity is also often biphasic, producing one set of behavioral changes at low doses and exhibiting distinctly different bioactivity at high doses. Specifically, CBD (5–20 mg/kg) improves autistic-like social interaction deficits in adult Scn1a^+/− mice, a preclinical model of Dravet Syndrome, and only at a higher dose (100 mg/kg) reduces seizures in adolescent Scn1a^+/− mice^206,239. Furthermore, CBD (5 mg/kg) attenuates aggressive behavior induced by social isolation in mice through CB₁R and 5-HT1A receptor-dependent mechanisms²⁹⁷.

The second example of CBD’s potential is for the treatment of drug abuse, especially opioid addiction²⁹⁸. Human trial testing of CBD showed that it reduces cue-induced craving and anxiety in drug-abstinent individuals with heroin use disorder²⁹⁹. Similarly, CBD shows promising therapeutic efficacy in several rodent models of addiction. CBD (5–20 mg/kg) reduces heroin-triggered anxiety-like behavior, motor activity, and somatic signs in mice³⁰⁰. Mechanistically, this treatment normalizes the altered expression profile of the mu-opioid receptor, proopiomelanocortin, and CB₁R expression in the VTA of mice exposed to spontaneous heroin withdrawal³⁰⁰. Thus, CBD has promising adjunctive management properties for opioid withdrawal³⁰¹. CBD (20 mg/kg) 1) attenuated the rewarding effects of cocaine in rats via CB₂R, 5-HT1A and TRPV1 receptor mechanisms, 2) decreased motivation for cocaine in a behavioral economics paradigm in mice (i.e., reduced cocaine intake when administered during the acquisition phase), and 3) normalized the expression of GluRA2 and pERK1/2 in amygdala^302,303. Mechanistically, CB₁R antagonism reverses the effects of CBD on cue- and stress-induced reinstatement of cocaine-seeking behavior in mice, suggesting that CBD acts by allosterically reducing CB₁R signaling and its role in the control of psychostimulant triggered hyperlocomotion^96,304. Accordingly, CBD (10 μg/5 μL) infused in the lateral ventricle during the drug conditioning phase impairs the rewarding effects of methamphetamine³⁰⁵. Finally, rodent studies suggest that CBD might also help with alcohol addiction. For example, CBD (10 mg/kg, acute) attenuates ethanol-induced place preference and reduces aggressivity in group-housed rats³⁰⁶. Together, this growing preclinical evidence suggests that CBD could potentially be added to the standard detoxification regimen and mitigate acute or protracted withdrawal-related symptoms for several drugs of abuse, including opioids, psychostimulants, and alcohol. Clearly, the therapeutic potential of CBD is only starting to be revealed and preclinical studies are poised to help identify new medical indications.

7.2]. Is CBD use associated with side effects?

It is noteworthy that CBD does not produce overt impairment as a single agent, and thus exhibits a broader therapeutic index than THC (even at relatively high doses). For example, the recent testing of CBD’s therapeutic efficacy in multiple human clinical trials reported only mild and transient side-effects in some patients (e.g., somnolence, decreased appetite, diarrhea, and fatigue)²⁶⁵. Inhaled CBD (400 mg) followed by a one-week washout period increases the duration and intensity of measures of mismatch negativity amplitude, an auditory event-related potential that occurs when a sequence of repetitive sounds is interrupted by an occasional sound that differs in frequency or duration³⁰⁷. Furthermore, an observational study on the impact of long-term consumption of oral CBD (40–60 mg/day for 30 days) suggested no increased prevalence of liver toxicity in these self-medicating patients³⁰⁸. Thus, these studies indicate that a handful of reversible side effects may occur in some individuals who take high doses of CBD (e.g., 400 mg/day for several weeks) and that using 10-fold lower doses (e.g., 40 mg/day for several weeks) may be safe while still significantly influencing behavior.

7.3]. Interactions between CBD and THC bioactivities.

As early as the 1970s, evidence suggested that CBD (40 mg, oral, acute) might interact with THC (20 mg, oral, acute) in humans. For example, CBD delays the onset and prolongs the psychotropic effects of THC^309,310. We now know that different doses and routes of administration of CBD have different effects on THC’s bioactivity. In humans, inhaled CBD (400 mg) attenuates the duration and intensity of THC (12 mg)-induced changes in mismatch negativity amplitude, the measure of auditory sensory perception³⁰⁷. However, inhaled CBD (125 mg) does not curb THC (125 mg)-induced impairment of driving and cognition as measured by simulated driving and cognitive performance tests, and, in some circumstances CBD may even exacerbate THC-induced impairment³¹¹. Furthermore, inhaled CBD (30 mg) does not protect against the acute adverse effects of cannabis as measured by delayed verbal recall, psychotic symptoms and subjective measures³¹². Oral CBD (600 mg) attenuates both THC (1.5 mg, i.v.)-elicited paranoid symptoms and memory impairments.

The ability of CBD to dampen specific THC’s bioactivities in rodents is well documented and depends on the dose administered. For example, in rats, CBD (3mg/kg, i.p.) mitigates the negative anxiogenic effects experienced from THC (1mg/kg, i.p.), a response that correlates with increase dopamine levels in the PFC³¹³. CBD (3 mg/kg, i.p.) prevents the THC (3 mg/kg, i.p.)-induced downregulation of BDNF function in the hippocampus³¹⁴. However, CBD fails to reverse hypothermia or locomotor suppression induced by THC (30 mg/kg, i.p.) in rats³¹⁵.

Can CBD help mitigate the detrimental effects of THC in the developing brain? Multiple preclinical studies suggest yes. In adolescent mice treated for 15 days (PND 28–48) with daily i.p. injections of THC (3 mg/kg), CBD (3 mg/kg) reduced the impact of THC¹⁵. Specifically, THC triggers immediate and long-term impairments in working memory measured by the novel object recognition task; increases adulthood anxiety measured on the elevated plus maze; and increases repetitive and compulsive-like behaviors measured with the Nestlet shredding task and marble burying¹⁵. All of these THC-induced behavioral abnormalities were dampened by the coadministration of CBD while CBD as a single agent did not generate behavioral outcomes¹⁵. Treatment of adolescent female rats (PND 35–45, 15 days, twice daily) with increasing doses of THC (2.5, 5 and 10 mg/kg) impairs emotional behaviors in adulthood as measured by the swim test, sucrose intake, palatable food intake, and elevated plus maze^300,316. However, CBD mitigates select long-term behavioral alterations and molecular changes in the PFC^300,316. Intra-PFC injection of CBD reverses the cognitive impairments induced by acute glutamatergic antagonism within the PFC, and blocks the anxiogenic properties of THC³¹⁶. These results suggest that CBD mitigates some of the long-term behavioral alterations induced by adolescent THC exposure and diminishes the long-term changes in PFC molecular components.

In summary, multiple, independent, preclinical studies have shown that the effects of THC use during adolescence and adulthood are lessened by CBD. The molecular mechanism by which CBD decreases THC’s effect on the brain remains to be established in vivo, but likely involves the negative allosteric modulation of CB₁R. Further studies are urgently needed to help develop therapeutics for such conditions as CUD and THC-triggered psychosis and schizophrenia.

8]. OUTLOOK

This review raises questions that point to future research directions. First, we need more, large, placebo-controlled studies to solidify the therapeutic properties of both THC and CBD, as well as for the interaction of their bioactivities using several ratios and for clinical indications suggested by preclinical studies. This commitment will result in the development of multiple new transformative therapeutic options for the treatment of devastating diseases that remain without standard care options. Second, we need better definitions of “psychotropic”, “non-psychotropic” and “impairing.” In particular, we need to better understand to what extent individuals can self-titrate. For example, individuals that smoke Cannabis containing THC, Cannabis containing THC-CBD or vaping THC concentrates self-titrate to reach comparable intoxicating levels³¹⁷. Remarkably, self-titration of THC is also measured in rodents^19,20. Accordingly, it is critical to develop reliable measures of acute THC and CBD bioactivity in preclinical models and humans when administered using common routes (inhaled and oral). Considering that 11-OH-Δ⁹-THC exhibits comparable activity at CB₁R and accumulation in the brain to THC, could this THC metabolite significantly contribute to THC’s bioactivity profile when administered orally? What is the role of P450 enzymes expressed in the brain that metabolize THC and CBD³¹⁸? How does the bioactivity of artificial cannabinoid agonists compare with THC? For example, what is the bioactivity of JWH-018, a high-affinity and full agonist at CB₁R, often infused in illegal products such as Spice and K2, that can be abused^176,319?

It remains possible that chronic use of high dose CBD may produce side effects in vulnerable individuals, including during development and when used in combination with other medications. For example, considering that CBD modulates the activity of proteins involved in neuronal pathfinding and maturation, such as CB₁R and GPR55, we still do not know what the impact of high dose CBD is on brain development? What medication interact pharmacologically with CBD? We have started to outline the pharmacological interactions between THC and other psychoactive drugs, such as opioids, but we still lack research on CBD’s pharmacological interactions. For example, considering that CBD allosterically modulates GABA_AR, does co-administration of benzodiazepine and CBD increase concentration difficulties, confusion, sedation, and motor incoordination? To answer these questions, we need to further define the single agent bioactivities of CBD at various doses and administered through distinct routes, determine its selectivity at specific targets, and consider whether CBD tissue exposure reaches concentrations that will significantly engage and modulate these targets in vivo.

THC and CBD as single agents do not represent the full bioactive profile of Cannabis which contains additional bioactive phyto-CBs and ingredients, such as terpenes³²⁰. Thus, another blooming area of research is the bioactivity, mechanism of action, and effects of THC analogues that are produced by Cannabis at low levels and that activate CB₁R similarly to THC, such as Δ⁸-THC and Δ⁹-tetrahydrocannabiphorol. These analogues can be concentrated via extraction procedures^321,322. For example, Δ⁸-THC (1–6 mg) triggers comparable but less potent behavioral changes in healthy human adults than Δ⁹-THC at similar dose^71,323. This research is particularly important considering the recent increase in the US of products that contain Δ⁸-THC instead of Δ⁹-THC³²⁴.

9]. Conclusions

Research on the bioactivity of THC and CBD started in the 1960s, initially focusing on THC, and more recently rediscovering CBD. We now have a foundational understanding of their distinct molecular mechanisms of action, and how they modify neuronal and glial cell functions in the brain over an individual’s lifetime. Key results, gathered in well-controlled human studies and validated in preclinical model systems, have set the stage for the optimization and development of novel, phyto-CB-based therapeutics for a broad array of devastating neurological and psychiatric diseases. This review summarizes impairing consequences and side effects alongside potential therapeutics, emphasizing the importance of establishing THC and CBD dosages and regimens based on bioactivity (Figure 10). Considering the wide differences and diversity in the molecular targets engaged by THC and CBD, phyto-CB-based therapeutics will require optimization for each medical indication, from sleep disorders, anxiety and drug abuse to refractory childhood epilepsy and autism.

Figure 10: THC and CBD bioactivity occurs along continuum.

THC exhibits promising therapeutic response for the treatment of chronic pain (green), , potential harm reduction properties in the context of addiction (gray), and triggers impairing effects and enhances the incidence of mental health disorders (red). CBD exhibits promising therapeutic response for the treatment of epileptic seizures, autism, sleep quality and chronic pain, potential harm reduction properties in the context of addiction, and triggers side effects when used at high dose.

Recent development of molecular tools and experimental approaches, validated in rodents, create an opportunity to provide practical and translatable results for humans. Molecular imaging technologies now enable the visualization of targets engaged by phyto-CBs, including PET ligands that target and image CB₁R in vivo and can be leveraged to study CB₁R pharmacological interaction with other psychoactive drugs, such as nicotine and alcohol^325,326. Biosensors that selectively measure changes in THC, CBD and eCB levels in tissue are paving the way to establish the precise real-time changes of these agents in different tissues and how that relates to their bioactivities^327–329.

Much of the Cannabis and cannabinoid research over the past 50 years was done in reverse: listening to users to understand bioactivity and to patients to recognize therapeutic benefits before carrying out the needed trials. A perfect illustration of such a research trajectory was the development of CBD-containing therapeutics for the treatment of epilepsies: patients and their families spurred the study and optimization of the anti-seizure properties of CBD, and this novel therapeutic approach was eventually proven in preclinical studies and human trials. The current thriving and innovative field of Cannabis and cannabinoid research is reversing this trajectory. We are poised to optimize cannabinoids, develop multiple lines of transformative medications, and guide their safe use in the growing legal market of Cannabis-based products.