Posted on

Can i purchase cbd oil immune metabolic support for animals

Immune Responses Regulated by Cannabidiol

Department of Basic Sciences, Center for Environmental Health Sciences, College of Veterinary Medicine, Mississippi State University, Mississippi State, Mississippi.

Barbara L.F. Kaplan

Department of Basic Sciences, Center for Environmental Health Sciences, College of Veterinary Medicine, Mississippi State University, Mississippi State, Mississippi.

Department of Basic Sciences, Center for Environmental Health Sciences, College of Veterinary Medicine, Mississippi State University, Mississippi State, Mississippi.

* Address correspondence to: Barbara L.F. Kaplan, Department of Basic Sciences, Center for Environmental Health Sciences, College of Veterinary Medicine, Mississippi State University, 240 Wise Center Drive, Mississippi State, MS 39762 [email protected]

This Open Access article is distributed under the terms of the Creative Commons Attribution Noncommercial License ( which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and the source are cited.


Introduction: Cannabidiol (CBD) as Epidiolex ® (GW Pharmaceuticals) was recently approved by the U.S. Food and Drug Administration (FDA) to treat rare forms of epilepsy in patients 2 years of age and older. Together with the increased societal acceptance of recreational cannabis and CBD oil for putative medical use in many states, the exposure to CBD is increasing, even though all of its biological effects are not understood. Once such example is the ability of CBD to be anti-inflammatory and immune suppressive, so the purpose of this review is to summarize effects and mechanisms of CBD in the immune system. It includes a consideration of reports identifying receptors through which CBD acts, since the “CBD receptor,” if a single one exists, has not been definitively identified for the myriad immune system effects. The review then provides a summary of in vivo and in vitro effects in the immune system, in autoimmune models, with a focus on experimental autoimmune encephalomyelitis, and ends with identification of knowledge gaps.

Conclusion: Overall, the data overwhelmingly support the notion that CBD is immune suppressive and that the mechanisms involve direct suppression of activation of various immune cell types, induction of apoptosis, and promotion of regulatory cells, which, in turn, control other immune cell targets.

Cannabidiol History and Therapeutic Uses

Cannabidiol (CBD) is a plant-derived cannabinoid that has structural similarity to the primary psychotropic congener in cannabis, Δ 9 -tetrahydrocannabinol (THC). While CBD was initially isolated in the 1940s, its structure was not elucidated until the 1960s. 1,2 Unlike THC, CBD is bicyclic, comprised a terpene and an aromatic ring, and is a pentyl side chain. 1 It exists as two enantiomers, and it is (−)CBD 3 that is one of the major constituents found in Cannabis sp., and will be the focus of this review. For many years, THC and CBD were designated as psychoactive and nonpsychoactive, respectively, owing to the fact that THC produces the euphoric high associated with cannabis use, while CBD does not. However, since we know that CBD produces biological effects in the central nervous system (CNS), perhaps it is better defined as psychoactive, but not psychotropic, since it is active in the CNS without producing the euphoric high.

Perhaps it was the association of the euphoric high with THC that provided the initial focus on THC as opposed to CBD for potential medical use, since THC was originally identified as the active component of the plant. 4 However, in recent years, researchers have begun to explore CBD more as a therapeutic addition or alternative to THC. In the United States, oral THC (dronabinol, Marinol ® ) was first approved in 1985 by the Food and Drug Administration (FDA) to treat nausea and vomiting associated with chemotherapy. In 1992, dronabinol was also approved to treat cachexia in AIDS patients. 5 The next major advancement in cannabinoid pharmaceuticals was not until the mid-2000s when Sativex ® (nabiximols), a combination of THC and CBD as an oromucosal spray, was approved in Canada and the EU for neuropathic pain in multiple sclerosis (MS) and intractable cancer pain. 6 There are several reasons why combining THC and CBD in a single therapeutic could have value. 6 First, additional therapeutic benefit might be gained from hitting multiple targets; for example, if THC alleviates pain and CBD alleviates anxiety, 7–16 the combination therapy could be quite effective for chronic pain sufferers. Second, for disease states in which both THC and CBD are efficacious, a combination might allow for lower doses of THC, thereby potentially decreasing the psychotropic effects of THC. Third, there are some studies suggesting pharmacokinetic interactions between CBD and THC in which CBD treatment increases THC levels, 17–20 thereby allowing longer duration of effects of THC. Sativex ® has been evaluated in several clinical trials for spasticity associated with MS, neuropathic pain, and other conditions. 21–37

The latest approved cannabinoid pharmaceutical in the United States is CBD as Epidiolex ® . It was approved by the U.S. FDA in 2018 for epilepsy in children, in particular, for Dravet Syndrome and Lennox-Gastaut Syndrome. 38–42 CBD is also being investigated for its effectiveness in other diseases, including Tuberous Sclerosis, a genetic condition that causes growth of benign tumors all over the body, 43,44 schizophrenia, 45 and refractory epileptic encephalopathy. 46

In addition to the federally approved uses of CBD as Epidolex ® , CBD, usually as CBD oil, is widely used for putative medical benefit in several states, and is certainly used in states in which cannabis has been decriminalized, or legalized, for recreational use. 47 There are reports that CBD and other cannabinoids are beneficial for sleep, anxiety, pain, post-traumatic stress disorder, schizophrenia, neurodegenerative disorders, and immune-mediated diseases. 48 Often these conditions are self-diagnosed and self-treated, so there can be issues with dosing, other drug interactions, and characterization of CBD safety and efficacy.

Overall, it is clear that exposures to CBD are increasing. 47,49–51 It is also clear that CBD possesses therapeutic benefit, and in some cases, the beneficial effects of CBD are for diseases for which other available treatments have not been efficacious. 52 Together, these observations demonstrate the critical need to continue research on CBD, and therefore the goal of this review is to provide a summary of the effects and mechanisms by which CBD alters immune function. The review will include an evaluation of the role for various receptors through which CBD acts in the immune system. There will also be a description of CBD effects in animal and human immune responses, a characterization of mechanisms by which CBD mediates immune effects, and identification of knowledge gaps regarding CBD’s actions in the immune system.

Identification of CBD Receptors and Other Targets

Upon identification of the cannabinoid receptors, CBD was determined to exhibit low affinity for CB1 53 and CB2 receptors. 54 Consistent with this, we showed CBD-induced suppression of cytokine production in mouse splenocytes in both wild-type and double cannabinoid receptor knockout mice (Cnr1 −/− /Cnr2/− mice). 55 Another study demonstrated that ophthalmic administration of CBD following corneal inflammation reduced neutrophils in both wild-type and CB2 receptor knockout mice. 56 CBD-mediated suppression of anti-CD3-mediated proliferation of T cells also occurred in both wild-type and CB2 receptor knockout splenocytes. 57 However, there are a few reports using inflammatory stimuli in which CBD’s actions have been attributed to either CB1 or CB2 receptors ( Table 1 ). In a sepsis model induced with bacterial lipopolysaccharide (LPS), CBD-mediated inhibition of gastric emptying was reversed with the CB1 receptor antagonist, AM251. 58 Similarly, CBD inhibited interleukin (IL)-1 in a hypoxia-ischemia brain insult model and this effect was reversed with the CB2 receptor antagonist, AM630. 59 Use of ovalbumin to induce an asthma-like disease in mice demonstrated that some cytokines and chemokines induced in the lungs of mice that were suppressed by CBD (IL-4, IL-5, IL-13, and eotaxin) were differentially regulated by CB receptors. 60 Specifically, CBD-induced suppression of IL-5 was reversed in the presence of the CB2 receptor antagonist in bronchoalveolar lavage fluid and lung tissue, but there was no clear receptor dependence identified for CBD’s suppression of IL-4, IL-13, or eotaxin. 60 Thus, several studies do suggest a possible role for cannabinoid receptors in CBD-mediated suppression of inflammatory effects. It should also be noted that there are several reports suggesting that CBD acts as an allosteric modulator of CB1 or CB2 receptors, 61–64 although the role for CB1 or CB2 receptor allosteric modulation by CBD in immune function has not yet been determined.

Table 1.

Receptors Identified in Mediating Cannabidiol Immune Effects

Receptor Activity References
CB1 Agonist 58
CB2 Agonist 59,60
FAAH Inhibition 58,65–67,84,157,160
TRPV1 Agonist 65,66,74,82–88,105,148,194
Adenosine A2A Agonist 89–91,125,164
PPAR-γ Activation 92–94,96–98,136
5-HT1a Agonist 59
GPR55 Antagonist 109,110

FAAH, fatty acid amide hydrolase; PPAR-γ, peroxisome proliferator-activated receptor gamma; TRPV1, transient receptor potential vanilloid 1.

Another mechanism by which CBD acts is through inhibition of fatty acid amide hydrolase (FAAH), 65–67 suggesting that some of CBD’s effects are mediated by anandamide elevation since FAAH is responsible for the breakdown of anandamide. 65,66 Anandamide is an endogenous cannabinoid that exhibits affinity for CB1 and CB2 receptors. 68,69 A recent study suggested that the mechanism by which CBD elevates anandamide involves CBD interaction with fatty acid binding proteins, which prevents anandamide binding to these proteins to block anandamide transport to FAAH. 67 Since anandamide exhibits affinity for CB1 and CB2 receptors, and oxidation products of anandamide through cyclooxygenase or cytochrome P450 enzymes produce metabolites that also exhibit affinity for CB1 and CB2 receptors, 70,71 anandamide or its metabolites could account for some of the reports that CBD acts through CB1 and/or CB2 receptors. 58,61–64,72–84

Actions of CBD in immune function might also be mediated by the transient receptor potential V1, known as the vanilloid receptor (TRPV1), which was found to be activated by CBD. 65 Specifically, CBD was found to increase intracellular calcium in HEK cells transfected with TRPV1, and the CBD-induced increase in calcium was blocked by the TRPV1 antagonist, capsazepine. 65,66 Follow-up studies demonstrated that CBD desensitizes TRPV1 following activation. 85 Other studies have suggested that CBD acts through TRPV1 in the immune system ( Table 1 ). CBD can induce myeloid-derived suppressor cells (MDSCs), a type of regulatory cell, in the liver, and this effect is lost in TRPV1 knockout mice. 86 Specifically, regarding inflammation, CBD attenuated thermal hyperalgesia in response to carrageenan injections or in a neuropathic pain model in a capsazepine-dependent manner. 87,88 CBD suppression of cytokines in inflamed primary human colonic tissue was attenuated by the TRPV1 antagonist, SB366791. 82 SB366791 was also effective in reversing CBD’s suppression of rolling and adherent leukocytes in the sodium monoiodoacetate model of osteoarthritis in rats. 83 Together, these data suggest that TRPV1 is a critical receptor through which CBD acts in the immune system.

There have been several critical articles in which adenosine A2A receptors have been shown to mediate CBD’s effects in the immune system. 89–91 CBD was shown to inhibit microglial cell proliferation, which was associated with inhibition of adenosine uptake into cells. 89 The studies also demonstrated that CBD suppression of tumor necrosis factor-alpha (TNF-α) could be reversed using an adenosine A2A receptor antagonist, and CBD-induced suppression of LPS-stimulated TNF-α was not observed in adenosine A2A receptor knockout mice. 89 The role for adenosine A2A receptor in CBD-mediated neuroprotection or suppression of neuroinflammation was demonstrated in a model of hypoxia-ischemia in newborn mouse brains. 90 CBD inhibited adenosine uptake into rat microglial cells and CBD enhanced adenosine’s ability to inhibit TNF-α, which was prevented by the adenosine A2A receptor antagonist, ZM241385. 91 These studies show that CBD acts through the adenosine A2A receptor, especially in microglial cells.

CBD’s effects have also been shown to be mediated by peroxisome proliferator-activated receptor gamma (PPAR-γ) using PPAR-γ antagonists in models of β amyloid neuroinflammation, 92 apoptosis, 93,94 dinitrobenzene sulfonic acid (DNBS)-induced colitis, 95 human ulcerative colitis, 96 LPS activation of microglial cells, 97 and hypoxia-ischemia model of neuroinflammation. 98

There are several reports that CBD acts through the serotonin 5-HT1a receptor ( Table 1 ). Although most of the evidence for the involvement of this receptor comes from the attenuation of CBD’s effects using the 5-HT1a antagonist, WAY100635, early studies demonstrated that CBD displaced binding of the 5-HT1a agonist, 8-OH-DPAT, in membranes from CHO cells expressing the human 5-HT1a receptor. 99 Few of the CBD-mediated effects acting through the serotonin 5-HT1a receptor have been reported in immune cells, but immune cells do express 5-HT1a. 100–103 One study showed that IL-1 produced in the brain in response to hypoxia-ischemia insult was inhibited by CBD, and reversed with the 5-HT1a receptor antagonist, WAY100635. 59

Studies have suggested that CBD might act through other receptors, including other TRP receptors, 66,85,104–107 or opioid receptors. 108 There is also evidence that CBD acts through blockade of GPR55, 109 and specifically that CBD modestly antagonized proinflammatory effects in human innate cells following GPR55 activation. 110 Thus, together, the current data support that immune effects of CBD are mediated through activation of CB1, CB2, TRPV1, adenosine A2A, and PPAR-γ receptors, blockade of GPR55 receptors, and FAAH inhibition.

CBD Immune System Effects and Mechanisms

Immunity is maintained through various cell types acting together to provide protection against foreign invaders, and simultaneously avoid reactions against self-proteins. Thus, an appropriate immune response requires a regulated balance between robust reactions against non-self, but limited or no reactions against self. Cell types include neutrophils, macrophages, and other myeloid cells comprising the innate immune system, which reacts quickly to destroy pathogens. In the event that an innate response is insufficient, certain innate cells can activate the adaptive immune response, comprised predominantly of T and B cells. T cells can then provide signals that recruit and activate other immune cells, or directly lyse or induce apoptosis of infected cells. T cells can also help stimulate B cells, which produce antibodies to neutralize pathogens and/or enhance destruction of the pathogens. Communication between the various cell types, and therefore the innate and adaptive immune responses, is mediated by expressed or secreted proteins called cytokines or chemokines. Inflammation is the process commonly associated with the innate immune response since pathogen destruction can also cause tissue damage, although T cells certainly are proinflammatory as well. In fact, many cell types, regardless of whether they are immune cells, produce proinflammatory cytokines in response to inflammation.

The effects of CBD on immune responses can involve innate or adaptive responses. In assessing these responses, various cell types and their functions have been examined. For instance, a common end-point to examine regardless of cell type is cytokine or chemokine production. Typical proinflammatory cytokines include IL-1α, IL-1β, IL-6, TNF-α and IL-17A, while IL-10 is considered anti-inflammatory. Some cytokines are produced by specific T cell subsets; for instance, the Th1 subset produces interferon-gamma (IFN-γ) and promotes cell-mediated cytotoxicity, while the Th2 subset produces IL-4 and promotes B cell responses. Other end-points that might provide clues of disruption of immune competence are nitric oxide or myeloperoxidase (MPO) production from innate cells, as these are often released during pathogen destruction. Thus, the effects of CBD on immune function are presented by cell type, outlining known mechanisms by which CBD alters various end-points. Tables 2–4 include the studies described in the text (and others) and are organized by experimental approach. As indicated above, inflammation can induce proinflammatory cytokine production in nonimmune cells, so there are also a few of those examples included in the tables.

Table 2.

Cannabidiol-Induced Immune Suppression by Cell Type in Human Cells In Vitro

Cell type End-point(s) References
PBMCs ↓rosette formation 138a
PBMCs ↓cytokines 111,112a
Human cell lines b ↓cytokines 186
HL-60 b ↑apoptosis 113
Jurkat and MOLT-4 T cells b ↑apoptosis 80a
Human coronary artery endothelial cells ↓adhesion molecules, migration, transcription factors, nitrative stress 119a
Jurkat T cells b ↓cytokines, transcription factors 55a
Human neutrophils ↓migration 195
PBMCs ↓indoleamine-2,3-dioxygenase (IDO),
THP-1 cells b ↓IDO 142
PBMCs ↑apoptosis 114a
Human intestine ↓proteins and nitric oxide 96
Human liver sinusoidal endothelial cells ↓adhesion molecules 118
Human gingival mesenchymal stem cells ↓inflammatory genes 79
Caco-2 cells b ↓phosphoproteins 82a
Primary colonic explants ↓cytokines 82a
Human neutrophils ↓ROS 185
Human PBMCs ↓proliferation and cytokines 146a
HaCaT human keratinocytes b ↓cytokines 84
Human monocytes ↑apoptosis 115a
Human plasmacytoid dendritic cells ↓CD83 expression in HIV + dendritic cells 134a

ROS, reactive oxygen species.

Table 3.

Cannabidiol-Induced Immune Suppression by Animal Cell Type In Vitro

Cell type End-point(s) References
B6C3F1 female splenocytes ↓IL-2 196
EL-4 T cells a ↑apoptosis 80b
Mouse EOC-20 microglial cells a ↓proliferation 89b
BALB/c male splenocytes ↓IL-4 and IFN-γ 140b
B6C3F1 female splenocytes ↓IL-2 and IFN-γ 55b
BALB/c male thymocytes and EL-4 T cells a ↑apoptosis 150b
BALB/c male splenocytes ↑apoptosis 151b
Sprague-Dawley rat microglial cells c ↓adenosine uptake, 91b
BV-2 cells a ↓cytokines, ↓NF-κB activation 147b
Mouse brain slices c ↓cytokines 90b
Rat male astroglial cells ↓gliosis 92b
C57BL/6 male Kupffer cells ↓TNF-α 118
BALB/c microglial cells c ↑apoptosis 156b
BV-2 cells a ↓oxidative stress, ↓Ccl2 159
MOG-specific female T cells ↓IL-17A and IL-6 144b
Mouse brain endothelial cells a ↓VCAM-1 and leukocyte adhesion 164b
Rat astrocytes c ↓Ccl2 164b
RAW cells a ↓TNF-α 148
MOG-specific female T cells ↓cytokines 143b
Rat male splenocytes and mesenteric lymph nodes ↓proliferation and cytokines 146b
Primary mouse male and female microglial cells ↓activation 97b
BV-2 cells a Alteration of circadian rhythm-associated genes 197
BV-2 cells a alteration of miRNAs 161b
C57BL/6 or BALB/c female splenocytes ↓proliferation and cytokines 57

c Sex not stated for cells derived from animals (or in the case of primary microglial cell isolates, not determined in newborn animals).

IFN-γ, interferon-gamma; IL, interleukin; miRNA, microRNA; MOG, myelin oligodendrocyte glycoprotein; NF-κB, nuclear factor-κB; TNF-α, tumor necrosis factor-alpha; VCAM-1, vascular cell adhesion molecule-1.

Table 4.

Cannabidiol-Induced Immune Suppression in Animals In Vivo

Model Disease model Route, dose range, and duration/frequency a Major effects Reference
Male CD-1 mice sRBC i.p. Modest ↓antibody production 155
25 mg/kg
4 days
Male DBA/2 mice Collagen-induced arthritis i.p. or oral ↓disease, ↓TNF-α and IFN-γ 139b
2.5–20 mg/kg for i.p.
5–50 mg/kg for oral
10 days
Male ICR mice Carrageenan-induced inflammation ethosome (CBD in ethosomal gel) ↓inflammation 198
100 mg of ethosomal CBD (3%)
Male Wistar rats Carrageenan-induced inflammation Oral ↓disease, ↓prostaglandin (PGE2) 199
5–40 mg/kg
3 days
Female NOD mice Diabetes i.p. ↓disease incidence, ↓IL-12, TNF-α and IFN-γ, ↑IL-4 123
5 mg/kg/day
10–20 injections
Female C57BL/6 mice EL-4 leukemia growth i.p. ↑apoptosis of tumor cells 80b
12.5 or 25 mg/kg once
Male Wistar rats Sciatic nerve pain or CFA-induced inflammation Oral ↓pain, ↓TNF-α, ↓prostaglandin (PGE2) 88
2.5–20 mg/kg
7 days
Male Sprague-Dawley rats Ischemia-reperfusion injury (myocardial) i.p. Modest ↓infarct size, ↓TNF-α 200
5 mg/kg
C57BL/6J mice c Aβ inflammation i.p. ↓IL-1β, ↓iNOS 117
2.5 or 10 mg/kg
7 days
Male BALB/c mice Ovalbumin (asthma) i.p. ↓serum antibodies, ↓IL-2, IL-4, and IFN-γ 140b
5–20 mg/kg
Male ddY mice Focal cerebral ischemia i.p. ↓infarct size, ↓neutrophil MPO activity 129b
3 mg/kg
various times surrounding occlusion
Female NOD mice Diabetes i.p. ↓disease incidence, ↓IL-6 and IL-12, ↑IL-4 and IL-10 124b
5 mg/kg/day
5 injections per week for 4 weeks
Female B6C3F1 mice sRBC Oral Modest ↓antibody production 55b
25–100 mg/kg/day
5 days
Male ICR mice DNBS colitis i.p. ↓inflammation, ↓colon weight:length ratio, ↓iNOS, IL-1β, ↑IL-10 95b
1–10 mg/kg
6 days
Male Wistar rats None i.p. ↓blood leukocytes and lymphocytes, ↓B, T and CTL cells, ↑NK and NKT cells 201
2.5 or 5 mg/kg
14 days
Male CD-1 mice Diabetes i.p. or i.n. ↓diabetic pain, ↓density of microglial cells 81b
0.1–2 mg/kg i.n.
1–20 mg/kg i.p.
3 months
Male C57BL/6 mice Streptozotocin-induced diabetes i.p. ↓disease, ↓TNF-α, NF-κB activity, ICAM-1, VCAM-1, iNOS, p-p38, p-JNK, ↑p-AKT 120b
1–20 mg/kg
11 weeks
Male Wistar rats TNBS colitis i.p. Modest ↓disease, ↓colonic contractions, ↓neutrophil MPO activity 130
5–20 mg/kg
Male Wistar rats Cecal ligation and puncture i.p. ↑disease survival 184
2.5–10 mg/kg
once or up to 9 days
Female Sabra mice Hepatic encephalopathy (bile duct ligation) i.p. Improved disease-associated cognitive impairments, ↓TNF-α 202
5 mg/kg
4 weeks
Male BALB/c mice Ovalbumin (footpad) i.p. ↓footpad swelling, ↓TNF-α and IFN-γ, ↑IL-10 188
1–10 mg/kg
5 days
Male Swiss OFI mice LPS i.p. i.p. ↓mast cell infiltration, macrophage activation marker, ↓TNF-α 96
10 mg/kg
Female C57BL/6 mice Experimental autoimmune hepatitis i.p. ↓hepatic inflammation, ↓IL-2, TNF-α, IFN-γ, IL-6, IL-17A, IL-12, MCP-1 (CCL-2), and eotaxin, ↑MDSCs 86b
10–50 mg/kg
Male C57BL/6 mice Ischemia reperfusion injury (liver) i.p. ↓hepatic inflammation, ↓MIP-1α, ICAM, MIP-2, TNF-α, NF-κB activity, ICAM-1, iNOS, p-p38, p-JNK 118
3 or 10 mg/kg
C57BL/6 mice c LPS i.v. i.v. ↓vasodilation, leukocyte margination, and extravasation, ↓COX-2, TNF-α, and iNOS 121
1 or 3 mg/kg
Male C57BL/6 mice LPS-induced pulmonary inflammation i.p. ↓BALF lymphocytes, macrophages, and neutrophils, ↓TNF-α, IL-6, MCP-1 (CCL-2), and MIP-2 125b
0.3–80 mg/kg
Male Wistar rats Meningitis (Streptococcus pneumoniae) i.p. Improved disease-associated cognitive impairments, ↓TNF-α 203
2.5–10 mg/kg
once or up to 9 days
C57BL/6 mice c Cerulein (pancreatitis) i.p. ↓disease, ↓TNF-α and IL-6, ↓neutrophil MPO 128b
0.5 mg/kg
Newborn pigs c Hypoxia-ischemic brain injury i.v. neuroprotection, ↓IL-1 59b
1 mg/kg
Male Wistar rats Ovalbumin (asthma) i.p. ↓TNF-α, IL-6, IL-4, IL-5, and IL-13 127b
5 mg/kg
Male C57BL/6 mice LPS-induced pulmonary inflammation i.p. ↓inflammation, ↓BALF lymphocytes, macrophages, and neutrophils, ↓TNF-α, IL-6, MCP-1 (CCL-2), and MIP-2 132
20–80 mg/kg
Female C57BL/6 mice None i.p. ↑MDSCs 136b
20 mg/kg
Female C57BL/6 mice Malaria (Plasmodium berghei) i.p. ↓IL-6 and TNF-α 204
30 mg/kg
3–5 days
Male Sprague Dawley rats Freund’s Adjuvant (osteoarthritis) Transdermal ↓inflammation, ↓TNF-α 205
0.6–63.2 mg/day
4 days
Male ICR mice DNBS Colitis i.p. or oral d ↓colon weight:length ratio, ↓neutrophil MPO 131
5–30 mg/kg for i.p. 10–60 mg/kg oral
3 days
Female NOD mice Type 1 diabetes i.p. ↓disease 206
5 mg/kg
5 injections/week for 10 weeks
Male A/J mice Experimental autoimmune myocarditis i.p. ↓disease, ↓lymphocyte populations in heart, ↓IL-6, IFN-γ, IL-1β, and MCP-1 (CCL-2) 126b
10 mg/kg
46 days
Male Wistar rats Middle cerebral artery occlusion i.c.v. ↓infarct size 149
50–200 ng/rat
5 days
Male Wistar rats Middle cerebral artery occlusion i.c.v. ↓infarct size, ↓TNF-α 207
50–200 ng/rat
5 days
Male Wistar rats Sodium monoiodoacetate (osteoarthritis) Intra-arterial ↓pain, ↓rolling and adherent leukocytes, ↓joint nerve demyelination 83b
100–300 μg/rat
multiple doses
Female C57BL/6 mice Alcoholic liver disease i.p. ↓liver damage, ↓neutrophils, ↓TNF-α, MIP-1, IFN-γ, IL-1β, and MCP-1 (CCL-2) 185
5 or 10 mg/kg
11 days
Male and female dogs Osteoarthritis Oral e ↓pain 208
2 and 8 mg/kg
every 12 h for 4 weeks
Male Wistar rats Ulcerative tongue lesion i.p. ↓inflammation 209
5 or 10 mg/kg
3 or 7 days
Female C57BL/6 mice Spinal cord contusion i.p. ↓spinal cord CD4 T cells, ↓IL-23A, IL-23R, IFN-γ, CXCL9, CLCL11, NOS2, and IL-10 189
1.5 mg/kg
1 and 24 h after injury, on day 3, then twice/week up to 10 weeks
Male Sprague-Dawley rats Carrageenan-induced inflammation Oral ↓hyperalgesia 210
100 or 10,000 μg/kg
Male Swiss mice Haloperidol-induced inflammation i.p ↓IL-1β and TNF-α, ↑IL-10 97b
60 mg/kg
twice/day up to 21 days
Male BALB/c mice Corneal inflammation Topical (ophthalmic) ↓pain, ↓neutrophils 56b
3% or 5%
Male ICR mice Ischemia-reperfusion injury (kidney) i.p. ↓kidney injury, ↓TH17 cells, ↑Tregs and Treg17 cells 152b
10 mg/kg
Female C57BL/6 and BALB/c mice Syngeneic or allogeneic bone marrow transplant i.p. ↓lymphocyte recovery 57
5 mg/kg
every other day for 2 weeks
BALB/c mice Ovalbumin (asthma) i.p. ↓airway resistance; ↓IL-4, IL-5, IL-13, and eotaxin 60b
5 or 10 mg/kg
three times at time of ovalbumin challenge

CBD, Cannabidiol; DNBS, dinitrobenzene sulfonic acid; iNOS, inducible nitric oxide synthase; i.n. intranasal; i.p., intraperitoneal; JNK, c-jun N-terminal kinase; LPS, lipopolysaccharide; MDSCs, myeloid-derived suppressor cells; MPO, myeloperoxidase; sRBC, sheep red blood cell; TNBS, 2,4,6-trinitrobenzene sulfonic acid; Treg, regulatory T cell.

CBD effects and mechanisms of immune suppression in innate cells

One of earliest effects reported with CBD was in human mononuclear cells, 111,112 in which TNF-α, IFN-γ, and IL-1α were all suppressed (0.01–20 μg/mL CBD or 0.03–64 μM CBD). Later studies focused on human monocytic cells revealed that CBD can induce apoptosis in either HL-60 (1–8 μg/mL CBD or 3.2–26 μM CBD) 113 or primary human monocytic cells (1–16 μM CBD). 114,115 Macrophages are also targets, although they have been studied more commonly in animal models. Peritoneal macrophages were used early on to demonstrate that CBD (3 μg/mL or 10 μM) targets nitric oxide, 116 and this has also been a well-studied target of suppression by CBD in many tissues and cell types. The mechanism by which CBD suppressed nitric oxide involves suppression of endothelial 87 or inducible nitric oxide synthase (iNOS) 58,95,117–121 in response to various inflammatory stimuli. iNOS is known to be regulated by the transcription factor nuclear factor-κB (NF-κB), 122 which is comprised of p65 and other proteins, and becomes active after degradation of the inhibitory protein, IκB. Decreased expression of iNOS by CBD correlated with stimulation of the inhibitory IκBα protein and inhibition of NF-κB p65 protein expression. 119,120 Using peritoneal macrophages from diabetic mice stimulated ex vivo with LPS revealed that macrophages isolated from CBD-treated mice did not produce as much TNF-α or IL-6 as macrophages isolated from vehicle-treated mice. 123,124 A direct effect of CBD decreasing macrophage numbers in the bronchoalveolar lavage fluid was shown following intranasal LPS administration to induce pulmonary inflammation. 125 There was also decreased expression of F4/80 (a marker of macrophages) mRNA expression by CBD in heart tissue in experimental autoimmune myocarditis. 126 Although this study identified CBD only affecting F4/80 mRNA expression as opposed to F4/80 cell surface staining, it does suggest a novel target (i.e., heart tissue) of CBD in a relatively understudied autoimmune model.

IL-6 is a proinflammatory cytokine produced by many cell types, predominantly innate cells. Many studies have shown that circulating IL-6 is readily inhibited by CBD in inflammatory models, including diabetes, 124 asthma, 127 pancreatitis, 128 and hepatitis. 86 CBD treatment in vivo resulted in lower IL-6 production in peritoneal macrophages stimulated ex vivo with LPS, 124 in the pancreas in acute pancreatitis, 128 and in bronchoalveolar lavage fluid in LPS-induced pulmonary inflammation. 125

There have been some reports that CBD alters neutrophil function. Compromised MPO activity by CBD has been studied in several tissues, including brain, 129 colon, 130,131 lung, 125,128,132 and pancreas. 128 Interestingly, in the pulmonary inflammation studies with LPS, neutrophil cell counts in the bronchoalveolar lavage fluid were also decreased by CBD compared to LPS. 125,132 Together, the results suggest that CBD’s mechanism for neutrophil suppression involves both decreased numbers of neutrophils and compromised MPO activity.

There are two recent studies focused on CpG stimulation of IFN-α production from human plasmacytoid dendritic cells. 133,134 While these studies are focused primarily on THC and other CB2 agonists, CBD was also used (1–10 μM) and did not affect IFN-α production. 133,134 It was interesting, however, that CBD suppressed the CD83 dendritic cell activation marker on dendritic cells derived from HIV + , but not healthy, individuals. 134 Reduction in dendritic cell CD83 signaling can compromise T cell function, 135 although additional studies using CBD in human dendritic cells and T cells are needed to establish the consequences of CBD-induced reduction in CD83 on HIV + dendritic cells.

Another mechanism by which CBD controls immune function is induction of regulatory cells. MDSC are innate, myeloid cells that possess the ability to control immune responses. Hegde et al. demonstrated that CBD induced CD11b + Gr-1 + MDSCs in the liver in a mouse hepatitis model. 86 Importantly, the isolated MDSCs were functional, that is, they suppressed proliferation of responder T cells ex vivo and improved liver function when administered before hepatitis induction. 86 CBD-induced MDSCs from the peritoneal cavity were able to attenuate inflammation in response to LPS. 136 In the experimental autoimmune encephalomyelitis (EAE) model, CBD induced MDSCs in the peritoneal cavity, but decreased the infiltration of MDSCs in the spinal cord and brain. 137 CBD-induced MDSCs from the peritoneal cavity were able to attenuate responder T cell proliferation ex vivo and attenuate EAE disease when administered in vivo. 137

CBD effects and mechanisms of immune suppression in lymphocytes

The area in which most of the effects of CBD in the immune system have been studied is T cells. Early studies examining rosette formation in response to sheep red blood cells (sRBCs) (generally considered to be a T cell response) revealed that CBD (1 and 100 μM) reduced this response. 138 Phytohemagglutinin (PHA)-stimulated IFN-γ production in T cells has also been shown to be inhibited by CBD (0.01–20 μg/mL or 0.03–64 μM). 111,112 Other studies have provided further evidence that T cell-produced IFN-γ is a critical target of CBD suppression. CBD inhibited IFN-γ production from lymph node cells isolated from arthritic mice stimulated ex vivo with collagen, 139 and from splenocytes isolated from NOD mice stimulated ex vivo with ConA. 123,124 IFN-γ production from splenocytes isolated from untreated mice was suppressed by CBD following ex vivo stimulation with phorbol 12-myristate 13-acetate/ionomycin (PMA/Io). 140 In the latter study, a 1-h exposure of CBD to the mice was meant to mimic the time for CBD distribution before receiving antigen sensitization with ovalbumin to induce asthma-like disease. 140 Thus, CBD’s ability to compromise various cytokines at the time of antigen sensitization might suggest that CBD affects primary activation of T cells, as has been suggested as part of the mechanism for other cannabinoids, such as THC. 141 Indeed, we have shown that a 30-min pre-treatment with CBD (0.1–20 μM) suppressed IFN-γ production in mouse splenocytes in response to PMA/Io or anti-CD3/CD28. 55 In these studies, it was shown that the mechanism by which CBD suppressed IFN-γ occurred at the level of transcription and that two important transcription factors for IFN-γ, activator protein-1 (AP-1) and nuclear factor of activated T cells (NFAT), were inhibited by CBD, suggesting a transcriptional mechanism for suppression. 55 CBD-induced suppression (0.1–10 μg/mL or 0.3–32 μM) of Ifng mRNA expression was shown using PHA-stimulated human PBMCs. 142 Given the many reports that IFN-γ seems to be a sensitive target of suppression by CBD, it was surprising that Ifng mRNA was not affected by CBD (5 μM) using encephalitogenic T cells stimulated by antigen-presenting cells (APCs) and myelin oligodendrocyte glycoprotein peptide (MOG35–55) in vitro. 143 However, CBD did inhibit expression of IFN-γ receptor 1 and CBD increased several IFN-γ-responsive genes known to attenuate T cell proliferation. 143 Overall, the data reveal that an important part of CBD’s action in the immune system is its ability to affect IFN-γ in multiple ways. Not only did CBD directly suppress IFN-γ production through a transcriptional mechanism under several conditions 55,142 but also suppressed IFN-γ receptor expression, and increased IFN-γ-induced genes that subsequently attenuate other immune targets. 143

A few other T cell-derived cytokines have been shown to be targets of CBD. As noted above, IL-6 is a critical target of CBD in many cells and tissues, 82,84,86,97,125–128,132 many of which are innate cells. However, IL-6 was also suppressed by CBD (5 μM) using encephalitogenic T cells stimulated by APCs and MOG35–55 in vitro, 144 and “IL-6 signaling” as a critical pathway suppressed by CBD. 143 Interestingly, “IL-17 signaling” was also identified as a critical pathway suppressed by CBD (5 μM) in T cells in vitro. 143 It should be noted that IL-6 promotes the differentiation of TH17 cells, 145 so the simultaneous suppression of IL-6 and IL-17A by CBD is consistent with CBD suppressing TH17 cell differentiation. Indeed, CBD (1–20 μg/mL or 3.2–64 μM) suppressed IL-17A production in human CD3 + T cells (derived from healthy patients or patients with MS or nonseminomatous germ cell tumors) stimulated ex vivo with PMA/Io. 146 Taken together with the data described in innate cells above, it is clear that CBD’s action in inflammation and immune function involves suppression of cytokine production from many different cell types.

The ability of CBD to suppress transcription factors such as NFAT, AP-1, and NF-κB likely accounts for its widespread suppression of many cytokines. 74,82,118–120,147–149 Some of the studies suggest that CBD increased, or perhaps stabilized, expression of IκB as part of the mechanism by which it suppresses NF-κB. 119,120,147 CBD (4 μM) stimulated IκB-α expression in high glucose-treated human coronary artery endothelial cells. 119 CBD induced expression of IκB-α in heart tissue from diabetic mice in vivo 120 and in LPS-stimulated microglial cells in vitro (CBD 1–10 μM). 147 It is interesting that NF-κB activity has not yet been identified as a target in T cells, suggesting that CBD-mediated suppression of NF-κB plays a bigger role in mediating anti-inflammatory effects in non-T cells.

Certainly, some of the dysregulation of these transcription factors is the result of suppression of various kinases upstream of their activation. Extracellular signal-regulated kinase (ERK), c-jun N-terminal kinase (JNK), and p38 MAPKs have all been identified as targets of suppression by CBD in various cell types. 74,80–82,118,120 Of these reports, one was conducted in human T cells. 80 In these studies, CBD (5 μM) was shown to suppress expression of total and phosphorylated p38 at the 16-h timepoint following CBD treatment. The authors also showed that the CBD-mediated inhibition of phosphorylated p38 was reversed by SR1445328 or tocopherol, suggesting that CBD acts through CB2 and that the mechanism of suppression involves reactive oxygen species (ROS) production. 80

Although well studied in cancer cell lines and primary tumor tissue, CBD-mediated apoptosis is also a contributor to the immune suppressive mechanism. Initially CBD-induced apoptosis in T cells was described in Jurkat and MOLT4 human T cells. 80 In the same study, McKallip et al. observed increased apoptosis of mouse lymphoma cells injected into, and then recovered from, the peritoneal cavity of mice that were treated with CBD. 80 Since then, there has been a series of studies characterizing the mechanisms by which CBD induced apoptosis in mouse immune cells. CBD (1–16 μM) was shown to induce apoptosis in mouse thymocytes and EL-4 T cells. 150 The same group demonstrated that CBD (1–16 μM) induced apoptosis in mouse splenocytes, including assessment of CBD-induced apoptosis by cell type (B220 + B cells and CD4 + and CD8 + T cells). 151 In both studies, CBD increased ROS, and CBD-mediated apoptosis was attenuated by N-acetylcysteine. 150,151 Wu et al. further demonstrated that the CBD increased ROS-activated caspase-8 to mediate apoptosis. 151 In follow-up studies in human monocytes, Wu et al. noted that CBD (1–16 μM) readily induced apoptosis, but that the effect of CBD on apoptosis was lost if the monocytes were pre-cultured for 72 h. 114 The authors suggest that the differential responsiveness to CBD was due to an increase in antioxidant capacity in cultured cells, which is a thought consistent with the mechanism by which CBD induced apoptosis in mouse lymphocytes. 150,151 CBD-induced apoptosis (1–16 μM CBD) in human monocytes was due to a cascade of intracellular events, including opening of the mitochondrial permeability transition pore, depolarization of the mitochondrial membrane potential, oxidation of a lipid in the mitochondrial inner membrane, and mitochondrial ROS generation, leading to cytochrome C release. 115 Thus, this latest study demonstrates a critical role of the mitochondria in CBD-induced apoptosis.

Another important mechanism by which CBD acts to control immune responses is through regulatory T cell (Treg) induction. In the ConA model of hepatitis, CBD modestly enhanced Tregs in the liver as quantified by CD4 + Foxp3 + cells. 86 A confirmation of in vivo induced Tregs by CBD was noted in an ischemia-reperfusion injury model in the kidney, in which CBD returned the disease-induced reduction in CD3 + Foxp3 + cells to baseline. 152 Interestingly, in the ischemia-reperfusion kidney model, CBD also induced “TReg17 cells,” which were defined as CD3 + Foxp3 + CCR6 + STAT3 + . 152 It has been suggested that Treg17 cells help control a TH17 response. In vitro, CBD (5 μM) induced a CD69 + LAG + population in CD4 + CD25 − cells, which were identified as one type of regulatory cell, and induced Il10 mRNA expression. 153 We showed in vitro that CBD (1–15 μM) induced functional CD4 + CD25 + Foxp3 + T cells under conditions of suboptimal stimulation and that Il10 mRNA expression was induced. 154

There are only a few studies in which B cells are identified as targets of CBD. CBD given at 25 mg/kg by intraperitoneal (i.p.) injection modestly reduced the sRBC-induced plaque-forming cells, which is a measure of antibody production. 155 We conducted a similar study using oral administration of CBD and also found modest inhibition of antibody production. 55 Other studies have shown that CBD robustly inhibited the sRBC-induced antibody production in vitro, 55 suppressed ovalbumin-induced IgM, IgG1, and IgG2a in an in vivo asthma model, 140 and reduced expression of activation markers such as major histocompatibility complex II, CD25, and CD69, on B cells. 153 CBD has also been shown to induce apoptosis in B cells. 151 Overall, the results suggest that B cells can be targets of suppression by CBD.

CBD-induced neuroprotection by suppression of microglial cell activation

There is no doubt that many of the mechanisms already identified for innate cells and lymphocytes also account for CBD’s ability to decrease microglial cell activation. CBD (1–16 μM) induced apoptosis in microglial cells, 156 which was dependent on activation of caspase 8 and 9, and was reversed in the presence of an agent that depletes cholesterol and disrupts lipid rafts. 156 These results suggest that CBD-induced apoptosis is dependent on lipid raft formation, 156 and indeed, this observation was confirmed by another group in BV-2 microglial cells. 157

BV-2 microglial cells have been used as a model in several articles, in which detailed transcriptional effects of CBD have been evaluated. 147,157–159 The mechanisms contributing to CBD (10 μM)-mediated suppression of LPS-stimulated cytokine production in microglial cells includes decreased activation of the Toll/IL-1 receptor domain-containing adapter-inducing IFN-β (TRIF)/IFN-β/signal transducer and activator of transcription (STAT) signaling pathway. 147 CBD suppressed LPS-stimulated NF-κB activation, and induced LPS-stimulated STAT3 activation, which has been shown to suppress NF-κB activation. 147 CBD (10 μM) was shown to affect several genes involved in lipid metabolism in unstimulated BV-2 cells, 157 which might account for CBD’s ability to increase anandamide 58,65–67,84,157,160 or could account for CBD’s dependence on lipid raft formation to induce apoptosis 156,157 Follow-up studies examining CBD’s effects in unstimulated BV-2 cells demonstrated that CBD (10 μM) alters zinc homeostasis, oxidative stress, and glutathione levels in microglial cells. 158,159 A recent study demonstrated that CBD alters microRNA (miRNA) expression, 161 and two of the CBD miRNA targets identified are discussed. First, CBD downregulated miR146-a, which acts as a negative regulator of inflammation, in both resting and LPS-stimulated cells, thereby contributing to CBD’s ability to downregulate proinflammatory cytokines. 161 Second, CBD upregulated miR-34a, which has several roles in cell survival, such as cell cycle, apoptosis, and differentiation. 161 These results show that CBD-induced alterations in miRNA expression are involved in the mechanism by which CBD suppresses immune function.

In vivo, CBD has been shown to decrease microglial accumulation in the spinal cord in diabetic mice, 81 which might contribute to attenuation of neuropathic pain, and CBD decreased haloperidol-induced activation of reactive microglial cells. 97 CBD’s suppression of TNF-α production from microglial cells in vitro was mediated by A2A adenosine receptors in EOC-20 mouse microglial cells (0.5–5 μm) 89 or rat retinal microglial cells (1 μM). 91

CBD Effects in Autoimmune Disease Models

EAE and MS

The immunosuppressive and neuroprotective mechanisms of CBD make it an ideal therapeutic candidate for MS, a neurodegenerative autoimmune disease of the CNS that affects ∼2.5 million people worldwide. The average age of onset is around 30 years, and symptoms can vary greatly for each patient based on the lesion locations within the CNS. 162 Two models frequently used in the laboratory environment to study MS are the EAE and Theiler’s murine encephalomyelitis virus (TMEV) models, and an increasing number of studies have shown promising results with CBD using these models ( Table 5 ). In 2011, Kozela et al. successfully demonstrated that CBD (5 mg/kg i.p.) administered at the onset of disease attenuated clinical disease, microglial activation, and T cell infiltration into the CNS in EAE, and that CBD reduced T cell proliferation in vitro. 163 CBD showed similar effects in the TMEV model, in which Mecha et al. demonstrated that CBD (5 mg/kg i.p.) administered for the first 10 days following disease onset reduced clinical disease and neuroinflammation by decreasing microglial activation and immune cell trafficking signals in the CNS. 164 Use of MOG35–55-specific T cells isolated from EAE mice in vitro has also been extremely vital to determining how CBD might be affecting T cells in these and other disease models. As outlined above, in the T cell section, in vitro CBD treatment of MOG35–55-specific T cells co-cultured with APCs with CBD suppressed IL-17A and IL-6 production, suggesting CBD suppressed TH17 development; however, production of Il10 mRNA was potentiated with CBD treatment, suggesting that CBD may have multiple suppressive mechanisms. 144 In vitro treatment of MOG35–55-specific T cells with CBD induced a Treg with a CD4 + CD25 LAG3 + CD69 + phenotype, promoted upregulation of anergy-associated genes, such as Lag3, Erg2, and Il10, and altered the balance between STAT3 and STAT5 activation. 153 In another study, CBD administered during disease onset increased the number of functional MDSCs present within the peritoneal cavity, decreased neuroinflammation, and reduced IL-17A and IFN-γ in the serum. 137 When splenocytes from these mice were restimulated ex vivo, the CBD-treated mice had significantly decreased levels of IL-17A and IFN-γ, and increased levels of IL-10 in the supernatants. 137 Finally, a recent study using an adoptive transfer EAE model showed a reduction in neuroinflammation, demyelination, and axonal damage with CBD treatment during disease onset. 165 Adoptive transfer EAE is a variation of the EAE model induced by transfer of encephalitogenic T cells into naive mice, which allows experiments performed with this model to focus more on the T cell-specific mechanisms of pathogenesis in the EAE model. From the accumulation of data, it is obvious that multiple immune cell types, proinflammatory and anti-inflammatory, within the EAE model are modulated by CBD, but overall, CBD appears to downregulate proinflammatory pathways and upregulate anti-inflammatory pathways in the EAE model.

Table 5.

Cannabidiol Effects in Experimental Autoimmune Encephalomyelitis

Model Approach Dosage/concentration Effects Reference
EAE in ABH In vivo In vivo: 0.5–25 mg/kg i.p. No effects 211
EAE in C57BL/6 In vivo and in vitro In vivo: 5 mg/kg i.p. in vitro: 1, 5, and 10 μM in vivo: ↓disease severity, ↓T cell infiltration into the CNS, ↓microglial activation, ↓axonal damage in vitro: ↓T cell proliferation 163a
TMEV in SJL/J In vivo and in vitro In vivo: 5 mg/kg i.p. in vitro: 1 and 5 μM in vivo:↓disease severity, ↓leukocyte infiltration into the CNS, ↓microglial activation, ↓CCL2 (MCP-1), ↓CCL5, ↓IL-1β ↓TNF-α in vitro: ↓sVCAM-1 production from endothelial cells, ↓leukocyte adhesion, ↓CCL2 (MCP-1) 164a
MOG35–55-specific T cells from EAE mice In vitro In vitro: 0.1, 1, and 5 μM in vitro: ↓IL-17A, ↓IL-6, ↑IL-10 144a
MOG35–55-specific T cells from EAE mice In vitro In vitro: 5 μM in vitro:↓IL-17A, ↓IL-6, ↑IL-10, ↑EGR2, ↑CD4 + CD25 − CD69 + LAG3 + phenotype, ↑STAT5/↓STAT3, ↓B cell activity, ↑Nfatc1, ↑Casp4, ↑Cdkn1a, ↑Icos, ↑Fas 153a
EAE in C57BL/6 In vivo In vivo: 5 mg/kg i.p. in vivo: ↓disease severity, ↓leukocyte invasion, ↓demyelination, ↓TNF-α, ↓IFN-γ, ↓IL-17A 212
EAE in C57BL/6 In vivo In vivo: 10 mg/kg i.p. in vivo: ↓disease severity, ↓FAS ligand, ↓ERK phosphorylation, ↓Caspase-3 activity, ↓Bax/↑Bcl-2, ↓p53-p21 activation, ↓apobody formation 166a
MOG35–55-specific T cells from EAE mice In vitro In vitro: 5 μM in vitro: ↓IL-1β, ↓IL-3, ↓Xcl1 mRNA, ↓IL-12a mRNA, ↑Dusp6 mRNA, ↑Btla mRNA, ↑Lag3 mRNA, ↑Irf4 mRNA, ↑IL-10 mRNA 143a,b
EAE in C57BL/6 In vivo In vivo: 10 mg/kg i.p. in vivo: ↓disease severity, ↓leukocyte infiltration, ↑PI3k/Akt/mTOR phosphorylation, ↑S6k phosphorylation, ↑BDNF expression, ↑PPAR-γ, ↓IFN-γ, ↓IL-17A, ↓JNK activity, ↓p38 MAP kinase activity 167a
Adoptive Transfer EAE in C57BL/6 In vivo and in vitro In vivo: 5–50 mg/kg i.p in vitro: 1, 5 & 10 μM in vivo: ↓disease severity, ↓leukocyte invasion, ↓demyelination, ↓axonal damage, ↓microglial activation, ↓CB2 receptor expression in CNS, ↓GPR55 receptor expression in CNS in vitro: ↓Cell viability, ↓IL-6, ↑apoptosis, ↑ROS 165a
EAE in C57BL/6 In vivo In vivo: 20 mg/kg i.p in vivo: ↓disease severity, ↓leukocyte invasion, ↓IL-17A, ↓IFN-γ, ↓RORγT, ↓T-bet, ↑IL-10, ↑MDSC ex vivo: ↓IL-17A, ↓IFN-γ, ↑IL-10 137a

CNS, central nervous system; EAE, experimental autoimmune encephalomyelitis; ERK, extracellular signal-regulated kinase; STAT, signal transducer and activator of transcription; TMEV, Theiler’s murine encephalomyelitis virus.

In addition to its immunomodulatory effects, CBD’s neuroprotective properties in the EAE model also indicate its therapeutic potential in MS. CBD has been shown to decrease the activation of proapoptotic proteins, such as caspase-3 and Bax, 166 and to counteract the effects of EAE on the PI3K/Akt/mTOR pathway, JNK, and p38 MAP kinases in the CNS of EAE mice. 167 Interestingly, the study from Giacoppo et al. found the PI3k/Akt/mTOR pathway was upregulated in neural tissues when EAE mice were treated with CBD. 167 However, Kozela et al. 153 observed a reduction in the activation of Akt in vitro in MOG35–55-reactive T cells, which might suggest a differential role for CBD’s effects on the PI3K/Akt/mTOR pathway in various cell types.

Despite the growing number of studies involving the neuroprotective and immunosuppressive effects of CBD, the majority of human studies involving cannabinoids and MS have been focused on the use of THC:CBD mixtures, with a particular focus on Sativex. Clinical studies that have been performed have shown that Sativex has beneficial effects on spasticity, mobility, bladder function, and pain in MS patients, and is well tolerated 22,25,28,31,168–175 ; however, there has been little focus on the neuroprotective and immunosuppressive effects of THC:CBD mixtures in MS, and so it is difficult to say at this point if the successful results observed with CBD in the animal models of MS will be observed in MS patients. For a more complete review on the effects of Sativex in MS, see Zettl et al. 176

Other autoimmune disease states

CBD has been shown to attenuate experimental autoimmune hepatitis, 86 experimental autoimmune myocarditis, 126 and autoimmune diabetes 123,124 in mice. There are few studies done with CBD only in human autoimmune diseases. In human patients, CBD at 20 mg/kg did not reduce clinical Crohn’s disease. 177 However, CBD is effective at attenuating intestinal inflammation in other models of human inflammatory bowel disease, 82,96 so it is possible that CBD could be effective at higher doses. Indeed, CBD as Epidolex for epilepsy in children is being used as high as 20 mg/kg, but CBD doses as high as 300 mg/kg have been evaluated, and have not exhibited significant adverse effects. 178

CBD Immune Enhancement Effects

Much of the data support the fact that CBD is immune suppressive and anti-inflammatory; however, there have been a few reports over the years that CBD has produced some immune enhancing effects ( Table 6 ). The potential for CBD, and other cannabinoids, to produce immune enhancing effects has been attributed to differences in hormetic (i.e., biphasic) responses depending on CBD concentration/dose, cell culture conditions, including serum presence and/or percent, immune stimulant, and magnitude of cellular activation in response to the immune stimulant. Indeed, studies from our laboratory and others have shown that CBD either enhanced or suppressed cytokine production (IL-2 and IFN-γ) in response to relatively low or high degree of immune stimulation, respectively. 154,179,180 The mechanism for the differential responsiveness likely involves alterations in intracellular calcium, as CBD increases intracellular calcium in mouse splenocytes regardless of the increase of intracellular calcium produced by the immune stimulant. 179 In addition, the differential cytokine production was correlated with nuclear expression of the NFAT transcription factor, 179 which is calcium responsive. Interestingly, CBD’s ability to increase intracellular calcium also likely accounts for some of the other enhancing effects, including stimulation of neutrophil degranulation, 181 chemotaxis, 182 and mast cell/basophil activation. 183

The Benefits Of CBD Oil For Dogs

The good news is that it can help with many of your dog’s health issues from allergies to cancer. The bad news is that the CBD industry for pets is still unregulated. That means the majority of pet owners might be getting ripped off.

So today I want to talk about all the good things CBD oil can do for your dogs. Then I’ll show you how to find the best product for your dog and talk about how to give it.

What Does CBD Oil Do For Dogs?

There’s a messenger system in your dog’s body called the endocannabinoid system. It helps regulate sleep, appetite, pain, the immune system and more. CBD impacts the activity of the messengers in this system and stimulates the nervous, digestive and immune systems, as well as the brain. And it can do this because the endocannabinoids in CBD are very similar to the ones found in your dog’s body.

That’s why the benefits of CBD can be deep and significant. And why CBD oil is the fastest-growing healthy plant in the world!

6 Ways CBD Oil Can Help Your Dog

Let’s take a look at common conditions where CBD can help dogs. And after I’ll talk about which CBD oil you should buy and general dosing information.

1. Dogs With Joint Problems

If your dog has joint pain, your vet might prescribe NSAIDs or other pain meds like Gabapentin. But NSAIDs can cause deterioration in joints and soft tissues … and they can damage your dog’s liver. Gabapentin can also cause kidney damage. Plus, it’s not all that effective.

CBD is a natural anti-inflammatory that doesn’t carry the same risk of side effects as drugs. It works by binding to CB1 receptors in the brain. These receptors stimulate the immune system to reduce inflammation. CB1 receptors also change the way the brain responds to pain.

CBD also binds to CB2 receptors found in the nervous and immune systems. When this happens, the body may produce more cannabinoids naturally. This helps reduce inflammation even more and reduce the pain associated with it.

In fact, researchers at Cornell University found that dogs taking CBD for arthritis were more active and showed a decrease in pain.

Some of the common people buy CBD Oil for dogs as an anti-inflammatory for joint problems include:

  • Arthritis
  • Hip and elbow dysplasia
  • Sprains and strains
  • Torn ligaments (CCL)

2. Dogs With Cancer

Sadly, 50% of adult dogs will get cancer. Cancer is a massive health challenge for dogs, especially if they undergo chemotherapy or radiation.

Cancer researchers are always looking for new ways to treat cancer and release the pain and nausea that can go with it. And CBD has been extensively researched as a cancer-fighting substance.

A study in mice showed that CBD slowed the growth of mammary cancer cells. And in 2018, researchers found that CBD increased survival time in mice with pancreatic cancer. Other animal studies show CBD oil has cancer-fighting abilities and can slow the growth of tumors.

In another study, cancer cells became more sensitive to treatment with CBD. That means CBD can increase the effectiveness of conventional cancer treatments.

CBD also kills cancer cells by blocking their ability to produce energy. And it can stimulate the immune system to produce killer cells that cause death in cancer cells.

Researchers also found that CBD blocks a cannabinoid receptor called GPR55. This is important because GPR55 increased the growth rate of cancer cells in mice.

CBD oil can also help with nausea associated with many cancer treatments. And studies have shown CBD can significantly reduce cancer-related pain.

3. Dogs With Seizures And Epilepsy

It’s estimated that about 5% of dogs suffer from seizures. They can be terrifying for both dogs and their humans … and they can cause anxiety.

Most vets treat epilepsy and seizures with antiepileptic drugs. Common options are phenobarbital or potassium bromide. But these drugs are extremely harmful to your dog’s liver and other organs. And even if the drugs don’t cause unmanageable side effects, they don’t always work …

So researchers at Colorado State University got excited when they studied CBD as a treatment for epilspsy in dogs. A whopping 89% of dogs that received the CBD had a reduction in seizures.

In human trials, CBD even worked in patients with drug-resistant epilepsy. In one study, 7 out of 8 patients saw a marked improvement within 4 to 5 months.

CBD reduces the frequency and severity of seizures because of how it interacts with the endocannabinoid system. It’s believed that abnormal electric charges of the neurons in the nervous system cause seizures. But CBD can bind to receptors in the brain … researchers speculate this can improve the functioning of the nervous system.

4. Dogs With Anxiety

Anxiety is a common reason dog owners turn to CBD. Anxiety can appear in different forms, including:

  • Noise phobia
  • Separation anxiety
  • Aggression
  • Fear

Of course, there are anti-anxiety drugs available … but CBD is being studied for anxiety because it doesn’t carry dangerous side effects.

Most human users of CBD take it for pain, anxiety and depression. Over a third of these users report that CBD worked “very well by itself.” CBD has even helped manage anxiety and insomnia in children with post-traumatic stress disorder (PTSD). And animal studies show its antidepressant effects aren’t just for people.

CBD can work quickly given directly by mouth when your dog gets stressed. It usually only takes 5 to 20 minutes to work. But CBD appears to be most beneficial for anxiety when given over a period of time. So if your dog is prone to stress, a daily dose might work best.

A 2012 study looked at stress in rats exposed to cats. The rats given repeated doses of CBD had less anxiety than those given a single dose.

Researchers aren’t certain how CBD relieves stress and anxiety, but it’s thought that it can help regulate serotonin. Serotonin is a hormone that regulates mood, social behavior, digestion, sleep and appetite.=

5. Dogs In Pain

Probably the most promising research on CBD is that done on pain. From nerve pain to arthritis, it works well … without the harmful side effects of pain medications.

CBD binds to both CB1 and CB2 receptors in the brain and nervous system and this helps change the way your dog’s brain perceives pain. Plus, CBD can help manage the other symptoms that accompany pain, such as sleeplessness and nausea.

CBD can also help manage acute pain from injuries.

6. Dogs With Allergies

Allergies are on the rise in dogs. And they’re difficult to treat … so, sadly, allergies are a common reason dogs are euthanized. Skin conditions in general are one of the most frequent reasons for vet visits.

The endocannabinoid system is also found in the skin … and that’s good news for dogs with allergies. It means CBD can help relieve dry and itchy skin. And it can promote the growth of new healthy skin cells.

You can give CBD internally for allergies, or use it externally for hot spots or interdigital cysts.

Now that you know a bit more about WHY you would give your dog CBD oil to your dog, let’s about HOW to choose a good quality product.

How To Choose The Best CBD Oil For Your Dog

CBD (Cannabidiol) is a naturally found substance in cannabis and hemp. Both deliver amazing health benefits … but there are differences.

Cannabis (marijuana) contains a relatively large amount of THC (tetrahydrocannabinol). THC is what causes the psychoactive activities of cannabis. It’s why marijuana can give a “high” or “buzz.”

CBD oil made from hemp contains much lower amounts of THC. To sell hemp legally, it must contain less than 0.3% THC. So while your dog can still enjoy the calmness and reduction in anxiety that CBD provides, he won’t get high. And that’s important … because you might enjoy the high, but your dog definitely doesn’t!

Your dog will also get the same pain-relieving and immune-supporting benefits from hemp CBD.

But not all hemp CBD products are the same …

1. Look For A Full Or Broad Spectrum Hemp

Check the label of your CBD product to make sure it’s full spectrum or broad spectrum.

This means your dog’s CBD oil contains not just CBD, but other important cannabinoids that occur naturally in full-spectrum hemp. This includes CBC (Cannabichromene) and CBG (Cannabigerol).

Researchers have looked at CBC for its …

  • Cancer-fighting activities
  • Ability to block pain and inflammation
  • Positive effect on brain cells

CBG is also studied for its medicinal use. It can decrease inflammation in the digestive tract and it can protect nerve cells and the eyes. It also supports healthy bladder function and fights cancer cells.

A full-spectrum CBD oil will also contain terpenes such as limonene, alpha-pinene, and beta-pinene. These are also naturally occurring medicinal substances found in all hemp.

Together, cannabinoids and terpenes create the entourage effect. This happens when compounds in hemp oil work synergistically to boost the medicinal properties of hemp oil.

CBD extracted with CO2 (I’ll talk about this in a moment) pulverizes the terpenes. This will make them hard to detect in testing and they won’t show up on the Certificate Of Analysis …

… but they’ll still be there and will contribute to the CBD oil’s medicinal effects.

CBD extracted with solvents will better preserve the terpenes. So you will find them noted on the Certificate Of Analysis.

But I don’t recommend solvent extracted products, which leads me to my next point …

2. Make Sure Your Dog’s CBD Uses CO2 Extraction

There are two common ways to extract the CBD oil from the hemp plant:

CO2 Extraction

As you’ve probably guessed, CO2 extraction uses carbon dioxide to extract oil from the plant. Using a high-pressure chamber, CO2 puts pressure on the hemp. This breaks down the hemp and releases the oil.

This method of extraction creates oils with a higher concentration of CBD. That means your dog will get more from his supplement. Of course, that also makes the product more expensive … but it’s better than the alternative.

Solvent Extraction

The cheapest way to extract oil from the hemp plant is with solvents, such as …

  • Propane
  • Butane
  • Petroleum products

But residue from these solvents will be in the product and they can be toxic to your dog.

Some CBD extraction uses natural solvents, such as ethanol or olive oil. This is much safer for your dog but these oils can destroy the hemp plant’s waxes and the resulting oil isn’t as beneficial.

3. Look For A Certificate Of Analysis

If your dog’s CBD oil doesn’t have a certificate of analysis (COA), run away!

A certificate of analysis is a document that shows the amount and type of cannabinoids in the CBD product. And it usually comes from a third-party laboratory,

COAs protect your dog from poor quality products and the manufacturer should have one for each batch of hemp. If there isn’t a COA on the company’s website, you’ll want to ask for one before you buy any CBD oil.

When looking at the COA, there are 5 important things to look for.

CBD Is The Same As Advertised

This is more common than you would think … in fact, we were once tricked by this!

What you might see is something like “500 mg CBD” on the product label. But don’t take the label at face value! Make sure the COA says the same amount as the label does.

Some lab tests express the CBD content in mg/g. So to calculate the amount of CBD, you need to know how many grams are in the bottle of CBD.

For example, let’s say the COA shows 16.9 mg/g CBD. To calculate how much CBD is in the product, multiply the number of mg/g by the number of grams the bottle weighs. (A typical 1-ounce dropper bottle of CBD will weigh 30 grams.) This will give you the total mg of CBD in the bottle. In this example, it’s 507 mg (16.9 mg/g x 30 gram bottle).

CBD Is Really Full Spectrum

Again, never take the label at face value! Some CBD is from isolate, which means it won’t have other important cannabinoids and terpenes.

Remember the entourage effect? You won’t get this extra boost with CBD isolate. So how do you find out if your dog’s CBD is from isolate? The COA will show that the product only contains CBD and no other cannabinoids. Stay away from these products.

There’s Not Too Much (Or Too Little) THC

If your dog’s CBD contains more than 0.03% THC, it’s probably marijuana and not hemp. It’s not legal and your dog won’t enjoy the psychoactive effects.

You also want to avoid products with zero THC. If there’s none, then your dog’s CBD is from isolate … and the health benefits will be fewer.

A Third-Party Did The Tests

Once again, never take the manufacturer’s word that the product is high quality. Make sure the product was properly tested by a third party lab. Unfortunately, the CBD industry isn’t regulated, which leaves you vulnerable to poor products.

There’s No Contaminants

You need to know where and how the hemp that’s used to make the CBD oil is grown. This plays a huge role in those test results you see in the COA.

Always look for an organic product to reduce any environmental toxin risks. You want to know that the soil and water it’s grown in is as clean as possible. That’s because hemp plants are really good sponges and can absorb contaminants as they grow. And it’s why heavy metal toxicity can be a concern when looking at CBD oils.

So be sure that you check the COA for any contaminants such as pesticides, heavy metals and solvent residues.

Cost Shouldn’t Be A Priority

It can be hard to compare products and some people give up and look at costs only …

… but this is not the best approach!

You want a high-quality and safe product for your dog. Extracting CBD from hemp requires a lot of plant material as well as careful monitoring.

If the product you’re considering has a price that’s significantly lower than the competition, there’s probably a reason for that …

But the most expensive doesn’t mean it’s the best CBD oil for dogs …

Instead, consider what we’ve reviewed …

  • How was the CBD oil extracted? (CO2 is best.)
  • Is the CBD concentration different than advertised? (CBD on COA should match the bottle.)
  • Is it full-spectrum? (The product should have other cannabinoids, not just CBD.)
  • Is the THC content worrisome? (THC should be less than 0.3% but higher than 0%.)
  • Is it organic? (Hemp is a sponge for contaminants.)
  • Was it third party tested? (If you can’t find a COA online, ask the manufacturer for one.)

These variables are what you need to look for when determining the quality of a product. The cost is never a sure sign of a product’s quality.

Side Effects Of CBD Oil For Dogs

The American Holistic Veterinary Medical Association surveyed dog owners. They wanted to see what, if any, side effects they noticed. And the great news is that there weren’t any major effects reported.

The most consistent side effects noticed were:

  • Sedation 19%
  • Overactive appetite 5%
  • Lack of energy 4%
  • Panic reactions 2.7%
  • Dry mouth/excessive drinking 2.3%
  • Nausea 1.7%
  • Vomiting 1.7%
  • Increased seizures 0.69%
  • Impaired mental functioning 0.68%

This means the most likely side effect you may see is that your dog gets sleepy. And that isn’t a bad thing. Especially if your dog suffers from seizures, anxiety, or has any pain, and you’d like to give CBD oil a try …

… but some CBD oils will have other additives and may not be safe.

Caution With CBD Oil Additives

You want to be sure there are no chemical additives or preservatives in the product you buy. These will cancel out the health benefits, even if the hemp is grown organically.

Also be aware of companies who have added essential oils (EOs) to their CBD oil. Even though they’re “natural,” EOs can affect animals profoundly.

If your holistic vet has recommended using a CBD oil with an EO, then follow her dosing recommendations. She’ll know what’s best for your dog’s unique health needs.

Some will recommend using CBD with frankincense as it’s good for tumor reduction in cancer patients. But always check with your holistic vet or herbalist first.

Dogs Taking Other Medications Or Supplements

If your dog is taking any other medications or supplements you will want to check with your holistic vet as well. CBD oil has many health benefits but it can change how your dog metabolizes some medications or supplements.

Researchers have looked at how CBD oil changes metabolism in humans. It can be similar to grapefruit, which causes significant reactions. So if your dog is taking any of the following medications you’ll need to ask your vet about dose changes:

  • Steroids
  • Allergy medications
  • Liver or kidney medications
  • NSAIDs
  • Heart medications
  • Anxiety medications

Hopefully, your holistic vet has helped you find alternatives to the medications above. But even then … CBD can affect herbs and natural supplements.

This doesn’t mean you can’t give your CBD oil if he uses other supplements or medications. You may just need to make adjustments. CBD changes the metabolism of other things but sometimes for the better! Meaning you can use less of another product or skip on the medications altogether.

And less is often more.

CBD Oil Dosage For Dogs

Each bottle of CBD has a specific concentration expressed in milligrams (mg). Most dogs are okay with the taste, so you can just put it on your dog’s food.

Dr Robert Silver recommends giving your dog 0.05 to 0.25 mg/pound of body weight, twice daily. He also suggests starting with a lower dose and working your way up. If 0.05 mg/pound is enough, stay at that dose. There’s no need to increase unless the lower dose stops working. If that happens, increase the dose to 0.125 mg/pound, twice daily and only continue to increase if your dog needs it.

For anxiety or health prevention, you’ll usually find that the lower doses work well. But if your dog is dealing with pain or immune issues, you’ll probably need a larger amount.

DNM RECOMMENDS: Four Leaf Rover’s Full Spectrum CBD Oil is 100% USDA organic, with a wide range of healthy cannabinoids. Buy CBD oil for dogs now >>

CBD oil for dogs is a natural, safe remedy that can help your dogs with pain, anxiety, caner, seizures and more.

McAllister SD, Christian RT, Horowitz MP, Garcia A, Desprez PY. Cannabidiol as a novel inhibitor of Id-1 gene expression in aggressive breast cancer cells. Mol Cancer Ther. 2007 Nov;6(11):2921-7.

Corroon J, Phillips JA. A cross-sectional study of cannabidiol users. Cannabis and Cannabinoid Research. 2018;3(1).

Aviello G, Romano B, Borrelli F, Capasso R, Gallo L, Piscitelli F, Di Marzo V, Izzo AA. Chemopreventive effect of the non-psychotropic phytocannabinoid cannabidiol on experimental colon cancer. J Mol Med (Berl). 2012 Aug;90(8):925-34.

The Impact of Feeding Cannabidiol (CBD) Containing Treats on Canine Response to a Noise-Induced Fear Response Test

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Associated Data

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.


Interest is increasing regarding use of Cannabidiol (CBD) in companion animals due to anecdotal evidence of beneficial behavioral and health effects. The purpose of this investigation was to evaluate the influence of CBD on behavioral responses to fear-inducing stimuli in dogs. Sixteen dogs (18.1 ± 0.2 kg) were utilized in a replicated 4 × 4 Latin square design experiment with treatments arranged in a 2 × 2 factorial, consisting of control, 25 mg CBD, trazodone (100 mg for 10–20 kg BW, 200 mg for 20.1–40 kg BW), and the combination of CBD and trazodone. A fireworks model of noise-induced fear was used to assess CBD effectiveness after 7 d of supplementation. Each test lasted a total of 6 min and consisted of a 3 min environmental habituation phase with no noise and a 3 min noise phase with a fireworks track. Plasma was collected 1 h before, immediately after, and 1 h following testing for cortisol analysis. Behaviors in each 3 min block were video recorded, and heart rate (HR) sensors were fitted for collection of HR and HR variability parameters. Research personnel administering treats and analyzing behavioral data were blinded as to the treatments administered. Data were tested for normality using the UNIVARIATE procedure in SAS, then differences examined using the MIXED procedure with fixed effects of treatment, period, time, and treatment x time interaction. Inactivity duration and HR increased during the first minute of the fireworks track compared with 1 min prior (P < 0.001 and P = 0.011, respectively), indicating the fireworks model successfully generated a fear response. Trazodone lowered plasma cortisol (P < 0.001), which was unaffected by CBD (P = 0.104) or the combination with CBD (P = 0.238). Neither CBD nor trazodone affected the duration of inactivity (P = 0.918 and 0.329, respectively). Trazodone increased time spent with tail relaxed (P = 0.001). CBD tended to increase HR (P = 0.093) and decreased the peak of low- and high-frequency bands (LF and HF, P = 0.011 and 0.022, respectively). These results do not support an anxiolytic effect of CBD in dogs given 1.4 mg CBD/kg BW/d.


Noise aversion or reactivity is one of the most common fearful behaviors in dogs, with 40 to 50% of dogs demonstrating at least one fearful behavior in response to noise exposure (1, 2). There is, however, considerable variation in the behavioral responses to noise. Some dogs will reduce activity while others become hyperactive. Some behavioral changes, such as panting and hiding, are mild, while others, like destructiveness and self-trauma, are more extreme and potentially hazardous to the health and well-being of both dog and owner (3). Such extreme and detrimental; stress associated with fear reduces overall health and lifespan (4, 5).

Despite the prevalence of noise aversion behaviors in dogs, they frequently go untreated with less than one-third of dog owners reporting that they would seek advice for the treatment of noise aversion (1). Potential treatment regimens for various noise aversion behaviors include systematic desensitization with a CD-based training system and administration of medications or natural products (3). There are several commonly prescribed drugs for the treatment of canine behavior disorders associated with fear and anxiety, including benzodiazepines, selective serotonin reuptake inhibitors, and tricyclic anti-depressants (6, 7). However, some owners may be hesitant to administer such medications, whether due to the possibility of undesirable side effects, personal bias against drug use, or cost. This has led to increased interest in the use of natural extract products to alter fearful behaviors, like dog-appeasing pheromones or oral supplementations such as L-theanine, a tryptic hydrolysate of milk protein and fish hydrolysate (8–12). Additionally, there has been renewed interest in the use of cannabinoids, cannabidiol (CBD) in particular, to regulate anxiety disorders in both humans and companion animals (13).

Cannabidiol is one of over 90 cannabinoids produced by Cannabis sativa and has been proposed to exert several beneficial effects, including acting as an anti-inflammatory, immunomodulatory, and anxiolytic agent (14–16). But unlike Δ 9 -tetrahydrocannabinol (THC), the other major cannabinoid produced by C. sativa that is toxic to dogs, CBD does not produce psychoactive effects due to its low affinity for the CB1 receptor (17). The potential anxiolytic effects of CBD have been attributed to several mechanisms, including its activation of 5-HT1A receptors and its ability to indirectly activate cannabinoid receptors by inhibiting the metabolism of the endocannabinoid anandamide (18, 19). This has produced great interest in using CBD as a potential alternative to conventional therapies to reduce anxiety. While there is considerable work examining its use as an anxiolytic in human and rodent models [reviewed extensively in (19)], this effect has yet to be examined in a canine model. But despite the lack of evidence, canine anxiety, and noise aversion are some of the most common reasons that pet owners seek information on and administer CBD to their pets (20).

As interest in, and use of, CBD in companion animals continues to increase, there is a critical need for research evaluating both the safety and effectiveness of CBD use for canine anxiety. Therefore, the objective of the current study was to evaluate the influence of CBD on behavioral responses to fear-inducing stimuli in dogs, with the underlying hypothesis was that CBD would reduce fearful and anxious responses. This hypothesis was tested using a fireworks model of noise-induced fear and anxiety in which the effectiveness of CBD was assessed by comparing CBD to both a positive and negative control and to the combination of CBD with the positive control. All treatments were expected to reduce fearful and anxious responses compared to the negative control.

Materials and Methods

This study was reviewed and approved by the Lincoln Memorial University (LMU) institutional animal care and use committee (protocol number: 1811-RES) prior to the start of the study. All housing and husbandry received were in accordance with the Animal Welfare Act, the Guide for the Care and Use of Laboratory Animals (8th ed.), and all applicable LMU SOPs.

Subjects and Housing

Twenty-four intact, adult dogs (12 male, 12 female; 1 to 5 years old; 17.7 ± 3.9 kg) of various mixed breeds, including Cur, Lab, Hound, Boxer, Shepherd, Dane, Schipperkee, Springer Spaniel, and Pit mixes were received from a local shelter for inclusion in this study. The shelter was asked to provide dogs weighing 16 ± 4 kg. Additionally, the shelter was informed and gave consent for the use of the dogs for research purposes prior to their arrival. Prior to beginning the experiment, each dog had a complete blood count (CBC), and serum chemistry analysis (IDEXX Laboratories, Inc., Westbrook, ME) performed, along with physical evaluation by a veterinarian and a fecal examination to rule out any underlying disease that might preclude enrollment. Dogs were excluded if they demonstrated serious behavioral issues, such as human aggression that would endanger research personnel, were severely emaciated, classified as a body condition score < 2 on a 5-point scale (where one is emaciated and five is obese), or if their initial evaluations revealed an underlying disease that required more than routine treatments (such as heartworm positive dogs). Two dogs were excluded from the experiment due to positive heartworm tests and 4 additional dogs were excluded due to other health or behavioral concerns. Dogs were individually housed in 1.2 × 1.8 m cages within one of two dog kennels at the LMU DeBusk Veterinary Teaching Center. Dogs were stratified by sex and evenly distributed between the two kennels.

Diets and Treatments

Dogs were fed Purina Pro Plan EN Gastroenteric Dry Dog Food (Nestle Purina Inc., St. Louis, MO) to meet the daily metabolizable energy requirements of intact adult dogs at maintenance, calculated as (70 * BW 0.75 ) * 1.8 and split into two meals fed at ~0,730 and 1,830 h each day. Dogs were weighed and body condition scored (5-point scale) weekly and diets adjusted accordingly. Treatments were arranged in a 2 × 2 factorial and consisted of (1) control (placebo treats), (2) 1.4 mg CBD/kg BW/d, (3) Trazodone + 0 mg CBD, and (4) 1.4 mg CBD/kg BW/d + Trazodone. Trazodone was dosed at 100 mg for dogs weighing 10.0–20.0 kg and at 200 mg for dogs weighing 20.1–40 kg as recommended by the veterinarian and based on previous work (21). Because trazodone does not require an extensive adaptation period, trazodone tablets were dosed via a Pill Pocket (Mars Petcare US, Franklin, TN) the evening prior and morning of the behavioral assessment.

The CBD was a constituent of a proprietary industrial hemp extract (AgTech Scientific, Paris, KY) that was incorporated into treats and administered in the form of two treats daily, with each treat containing half the daily dose. Groups not receiving CBD treatment received control treats (0 mg CBD). Both control and CBD treats were composed of the following ingredients: chicken, chicken liver, Asian carp, catfish, and in the case of the CBD treats, industrial hemp extract. Dosage of CBD was selected based on a preliminary palatability study that assessed increasing levels of CBD inclusion on food and treat consumption (unpublished). Treats were formulated to include CBD at a dose of 2 mg/kg BW/d based on an estimation that dogs would weigh an average of 16 kg. However, based on the mean weight of dogs included on the study, actual dosage of CBD was 1.4 mg/kg BW/d.

Treats were offered solely as a reward upon kennel re-entry following twice daily exercise at ~0,700 and 1,800 h each day. Trazodone tablets hidden in Pill Pockets were administered at ~1,830 h the evening before and 1,000 h the morning of each noise-induced fear response test. Empty Pill Pockets were administered to the control and CBD treatment groups on those days to ensure that research personnel administering the treats were blinded as to the treatments administered.

Testing Room and Equipment

The testing room was an ~2.72 × 3.38 m isolation room located on the opposite side of the building relative to where dogs were housed. The room contained a wall-mounted table and cabinets, a set of closed metal kennels, and a cloth dog bed. The dogs could interact with these objects, but none obstructed the dogs from view of the cameras. Two cameras (Model BRC-Z700, Sony Co., New York, NY and Model B07DQPS3KY, QallExpress International, China) were secured on opposite sides of the room near the ceiling–~2 m from the floor—to ensure the dogs would be within sight at all times. Dogs were isolated in the testing room; handlers monitored the dogs from the adjacent room via the cameras and could not be seen by dogs. Two Bluetooth speakers (Bose Co., Framingham, MA) were placed on opposite sides of the room near the cameras to create a surround-sound effect during the noise tests. Between each dog’s test, the room was cleaned with Rescue™ Concentrate (Virox Animal Health, Oakville, ON, Canada), an accelerated hydrogen peroxide-based disinfectant.


After intake and entrance into the study, all dogs were adapted to their environment, diet, daily routine, and the testing room for 3 d ( Table 1 ), in which the dogs spent 6 min in the testing room where behavior was monitored, but not scored. A baseline open field test followed the 3-d adaptation, where dogs were placed in the testing room, behavior was scored, but no noise track was played (described below). The next day, a 6-min baseline fireworks test was conducted (described below). Both the open field test and baseline fireworks test were used solely to select dogs for inclusion in the study. Dogs not exhibiting at least one behavioral change between the open field test and the fireworks test; behaviors such as cowering, shaking, vocalization, destructiveness, or tail tucking, were excluded from the study. Dogs included in the study spent 6 min in the test room every day throughout the experiment to eliminate the possibility of behavioral changes due to the novel environment of the test room. In order to acclimate dogs to the testing procedure, heart rate monitor bands were placed on the dogs for each days adaptation to the test room. Additionally, blood draws were simulated on non-testing days by restraining dogs and holding off cephalic and jugular veins prior to placing them in the testing room. The fireworks test was conducted on the last day of each 7-d period ( Table 1 ).

Table 1

Schedule of events.

Study day Key event
−7 to −6 Animal intake, physical exam, and bloodwork (CBC/serum chemistry)
−5 to −3 Acclimation to diet, daily routine, and testing room
−2 Open field test
−1 Baseline fireworks test
1 to 4 Start of treatment 1 (Squares 1–4 started on consecutive days)
7 to 10 Period 1 Fireworks Test, start of treatment 2 evening after test
14 to 17 Period 2 Fireworks Test, start of treatment 3 evening after test
21 to 24 Period 3 Fireworks Test, start of treatment 4 evening after test
28 to 31 Period 4 Fireworks Test

Open Field and Fireworks Tests

A fireworks model of noise-induced fear and anxiety was utilized to assess the effectiveness of the treatments. All dogs received from the shelter (n = 24) received 1 open field test and 1 baseline fireworks test. All dogs included on the study (n = 16) also received 1 fireworks test per 7-d period (5 total fireworks tests), each lasting 6 min. During the open field test, the dogs were placed in the testing room and their behavior was recorded in two 3-min blocks where no fireworks track was played in order to assess baseline behavior of dogs in the testing room. During the fireworks tests, the first 3-min block was the same as the open field test where no noise was played (Pre-Noise), and the fireworks track was played over a stereo speaker system (mean) during the second 3-min block (Noise).

In previous work using this model, a thunderstorm track was utilized to test the noise-induced fear response in dogs (9, 22). However, a fireworks video ( was used because according to Blackwell et al. (1) a larger percentage of dogs respond to fireworks than to thunderstorms. This noise-induced fear response test used in this study was a modified version of the one developed and validated by Araujo et al. (22). They utilized a 9-min test that included “before,” “during,” and “after” thunderstorm time points. Because they saw no behavioral differences (i.e., near door duration, inactivity duration) between the “during” and “after” thunder time points, the test for this study was shortened to 6 min, ending immediately after the fireworks track (Noise time point) ended. This allowed for the immediate post-test blood sample collection to be obtained more quickly after the fireworks test. The mean of 90 dB was selected based on previous work (9, 22) that were both successful in generating a response using equal or lesser decibel thunderstorm track. Behaviors in each 3-min time block were recorded and analyzed as separate time points (Pre-Noise and Noise).

Experimental Design

Sixteen dogs were included in this study (7 male, 9 female; 1 to 4 years old, mean BW 18.1 ± 0.2 kg). Dogs were selected based on their behavioral response to the baseline noise-induced fear test (described above), in which behaviors such as cowering, shaking, vocalization, destructiveness, and tucking tail upon the start of the fireworks track indicated the dog was reactive to noise. These behaviors were selected as they have been previously used to assess noise reactivity (9, 12, 23). Included dogs were then arranged in a replicated 4 × 4 Latin Square design experiment in which dogs within each square (4 dogs per square) were randomly assigned to receive one of the four treatments each week (Periods 1–4). Each square was tested on successive days for scheduling purposes. Dogs received each treatment for a 7-d period prior to each of the noise-induced fear response tests ( Table 1 ).

On testing days, all experimental procedures started at 1,200 h. CBD treats were administered ~4 to 6 hours prior to the test, and the morning dose of trazodone was administered ~2 to 4 h prior to the test. Dogs received the test at the same time each week. No washout period was included between treatment periods. At the time of the completion of this study (July 2018), there was little literature available on the pharmacokinetics of oral CBD administration. Samara et al. (24) reported that the half-life of IV CBD administration was 6 to 9 h but had no estimate for an oral dose. For a similar dose of trazodone, Jay et al. (21) reported a mean half-life of elimination of 166 min. From these half-lives, it was decided that the 7-d treatment period would be sufficient to allow for elimination of previous treatments prior to the next test while also allowing for acclimation to the next treatment. Additionally, time constraints on the availability of the kennels in which the dogs were housed prevented the inclusion of washout periods.

Because of scheduling constraints, the test started as soon as the dogs entered the testing room on testing days. This did not allow for either HRV or behavior to return to normal after movement from kennel to testing room. To account for this, only data from the last minute of the Pre-Noise time point was utilized to represent the behavior and HRV of dogs during that time point, which served as a reference of their normal behavior prior to the fireworks track starting. Additionally, only the first minute of the Noise time point was utilized to represent the dogs’ behavior and HRV during that time point in order to assess the dogs’ initial reaction to the fireworks tract.

Data Collection

Consumption of food and treats, consistency of stool, frequency of elimination, activity during exercise, mucus membrane color, and other indicators of general health status were monitored twice daily by research personnel. Evidence of any adverse event—defined as any symptom occurrence that would not be expected in normal dogs—was also monitored. However, no adverse events were observed in any dogs following the administration of CBD treats during this study.

On the day of each fireworks test, blood samples (5 mL) were collected via jugular or cephalic venipuncture 1 h prior to testing, immediately after testing (5–10 min after cessation of noise exposure), and again 1 h post testing for cortisol and CBD analysis. Blood samples were collected into EDTA plasma tubes, centrifuged at 1,645 × g, and stored at −80°C for later analysis. Plasma samples were analyzed in duplicate for cortisol using a commercial radioimmunoassay kit (MP Biomedicals, LLC, Solon, OH). The sensitivity reported for the radioimmunoassay was 1.7 ng/mL, and the intra- and inter-assay coefficients of variation were 5.3–8.9% and 7.5–9.3%, respectively.

Polar H10 (Polar Electro Inc., Bethpage, NY) heart rate sensors were used for the collection of heart rate (HR) and heart rate variability (HRV) parameters via Bluetooth connection to an iPhone app (Heart Rate Variability Logger, Marco Altini). Parameters measured are defined in Table 2 . In general, HR will increase and HRV will decrease in response to stressful stimuli as a result of an increase in sympathetic nervous system activity (25). Thus, an effective treatment would be expected to decrease HR and increase HRV, indicating higher parasympathetic activity. Just prior to the open field and fireworks tests, the heart rate monitor bands were placed around the chest of the dogs immediately behind the front legs, with the rubberized surface placed ventrally immediately behind the left front leg. Electrode gel was applied liberally to the rubberized surface of the transmitter band to promote conductivity. Due to all dogs having short hair and the use of electrode gel, dogs did not have to be shaved to promote conductivity.

Table 2

Definition of heart rate (HR) and heart rate variability (HRV) parameters.

Parameter Definition
HR Heart rate, bpm
AVNN Mean beat-to-beat intervals, ms
SDNN Standard deviation of beat-to-beat intervals, ms
RMSSD Square root of the mean squared difference of successive RRs or inter-beat intervals, ms
pNN50 Percentage of successive RR intervals that differ by more than 50 ms, %
LF Peak frequency of the low-frequency band (0.04–0.15 Hz)
HF Peak frequency of the high-frequency band (0.15–0.40 Hz)
LF/HF Ratio of LF-to-HF

Two cameras mounted ~2 m from the floor on opposite corners of the testing room continuously recorded all video and audio data for each test. The duration of behaviors given in Table 3 were logged by a single trained observer who was blinded to treatments using The Observer XT software (Noldus Information Technology Inc., Leesburg, VA). Three of the dogs included on the study had docked tails, and as such had no data on tail posture. The behaviors assessed were selected based on behavioral measures used in previous work evaluating canine anxiety and fear (5, 23, 26). Based on these previous studies, duration of fearful behaviors such as panting, cowering, and tail tuck were expected to increase during the fireworks test. Thus, an effective treatment was expected to decrease the duration of such fearful behaviors. Behaviors in different behavioral categories (i.e., Movement vs. Tail Posture) were not mutually exclusive, whereas behaviors within a behavioral category were mutually exclusive.

Table 3

Ethogram of behaviors tracked by a single trained observer blinded to treatments using The Observer XT (Noldus Information Technology Inc., Leesburg, VA).

Behavioral category Behavior Definition used
Movement Inactive Standing still, sitting, or laying down
Cowering Sudden cessation of movement in response to a stimulus
Pacing Frantically moving back and forth, restlessness
Destruction Scratching or chewing at room furnishings
Eyes Facing door Eyes are focused on the door of the room
Glancing around Eyes are shifting back and forth, possibly looking for the source of a sound
Other Eyes are focused on something else in the room
Ears Ears relaxed Ears are held in natural position
Ears erect Ears raised in response to stimulus
Ears moving Ears moving back and forth
Tail posture Tail relaxed Tail is not rigid and is lower than the top of the body
Tail stiff Tail is rigid and horizontal
Tail wagging Tail is wagging back and forth
Tail tucked Tail is tucked between hind legs
Muzzle Barking Emitting a short, loud sound
Whining Emitting a long, high pitch sound, often repeated
Panting Mouth open wide with tongue protruding while breathing heavily
Licking Using the tongue on own body or another object
Yawning Opening the mouth wide and inhaling
Biting Using teeth on the door or object

Statistical Analysis

The normality of data distribution was tested using the UNIVARIATE procedure in SAS (SAS Institute, Cary, NC) on the residual of the data. In instances where data did not meet normality assumptions, statistical analysis was performed on the natural logarithm transformation of the data. However, data were then back transformed for reporting purposes. The standard error of the back transformed data was calculated from the confidence limits of the transformed data as follows: SEM = (back-transformed upper limit–back-transformed lower limit)/3.92. The denominator relates to the Z-value of a 95% confidence interval (± 1.96). Cowering, pacing, destruction, tail wagging, tail tucked, and all muzzle behaviors could not be analyzed due to insufficient occurrences that prevented data from meeting normality assumptions. With the exception of HR, pNN50, and HF, parameters were not normally distributed and were analyzed using the natural logarithm of the data.

Blood cortisol was then analyzed using the MIXED procedure in SAS including the fixed effects of CBD, trazodone, period (Weeks 1–4), time (−60, 0, and 60 min), the interaction of CBD and trazodone, and the interaction of CBD by trazodone by time. Random effects included square and dog nested within square and repeated effect of time. All behavioral and HRV data from the 1-min immediately prior to (Pre-Noise) and the first min of the noise-induced fear response test (Noise) were also analyzed using the MIXED procedure in SAS including the fixed effects of CBD, trazodone, period (Weeks 1–4), time (Pre-Noise and Noise), and all accompanying interactions. Random effects included square and dog nested within square and repeated effect of time. Effects were considered significant when P ≤ 0.05 and considered a tendency when P ≤ 0.10.


Blood Cortisol

There was an overall effect of period on blood cortisol (P = 0.024). Blood cortisol was reduced in period 1 compared to both periods 3 and 4 (P = 0.003 and 0.003, respectively), but was similar across all other periods (P > 0.05). Blood cortisol was unaffected by time of collection and CBD (P = 0.189, 0.104, respectively). Similarly, neither the CBD x trazodone, time × CBD, time × trazodone, nor the time × CBD × trazodone interactions affected blood cortisol (P = 0.238 0.772, 0.667, and 0.812, respectively). However, trazodone lowered blood cortisol concentrations ( Figure 1 ; P < 0.0001).

Cortisol concentration (ng/mL) for each treatment (n = 16), back transformed after analysis. Error bars represent the standard error of the treatment mean (SEM), which was calculated from the back-transformed confidence interval for each treatment: SEM = (upper limit—lower limit)/3.92. Due to lack of effect of time (P = 0.189) and any interactions with time (P > 0.05), all time points (Pre-Noise and Noise) have been combined. Trazodone treatment reduced cortisol concentration (P < 0.001), whereas there was no effect of CBD (P = 0.104) nor the CBD by trazodone interaction (P = 0.238).

Heart Rate and Heart Rate Variability

There was a period effect on both HR and AVNN ( Table 4 ; P = 0.005 and 0.046, respectively). Heart rate in period 4 tended to be lower than in period 1 (P = 0.075) and was lower than in periods 2 and 3 (P = 0.004 and 0.001, respectively). Heart rate was similar between periods 1, 2, and 3 (P > 0.05). The mean beat-to-beat intervals (AVNN) was increased in period 4 compared to all other periods (P = 0.021, 0.018, and 0.030, respectively), but was similar between all other periods (P > 0.05). All other HR and HRV variables were unaffected by period (P > 0.05).

Table 4

Effect of trazodone (T), CBD (C), CBD by trazodone (C * T) interaction, time (Pre-Noise and Noise), CBD by trazodone by time (C * T * Time) interaction, and period on mean heart rate (HR) and heart rate variability (HRV) parameters for 1-min immediately prior to (Pre-Noise) and the first minute (Noise) of the noise-induced fear response tests administered after each 7-d treatment period.

Parameter Treatment SE 1 P-value
Control Trazodone (T) CBD (C) T+C Trazodone CBD C*T Time C*T*Time Period
HR, bpm 118.03 118.02 124.20 124.07 10.968 0.985 0.093 0.987 0.637 0.005
AVNN, ms 555.96 539.85 539.42 517.50 25.988 0.276 0.266 0.850 0.040 0.807 0.046
SDNN, ms 108.16 102.96 106.54 88.18 7.129 0.200 0.359 0.450 0.977 0.419 0.695
RMSSD, ms 100.35 89.28 91.70 69.71 12.649 0.130 0.189 0.538 0.366 0.654 0.538
pNN50, % 41.19 36.99 36.88 33.73 5.774 0.180 0.168 0.847 0.032 0.773 0.306
LF, Hz 0.090 0.062 0.050 0.048 0.0071 0.188 0.011 0.315 0.010 0.273 0.533
HF, Hz 0.142 a 0.076 b 0.059 b 0.068 b 0.0221 0.205 0.022 0.071 0.036 0.853 0.481
LF/HF Ratio 0.729 0.803 0.711 0.540 0.1052 0.595 0.126 0.183 0.053 0.039 0.908
Pre-Noise 0.651 a , b 0.992 a * 0.907 a , b 0.607 b 0.1402
Noise 0.808 a 0.615 a 0.515 a 0.474 a * 0.1461

1 The standard error (SE) of the back transformed data was calculated from the confidence limits of the transformed data as follows: SE = (back-transformed upper limit—back-transformed lower limit)/3.92.

ab* Within rows, values with different letters differ at P ≤ 0.05 and asterisks indicate a trend at P < 0.10.

Treatment T+C indicates the combination treatment of CBD and trazodone. With the exception of HR, pNN50, and HF, parameters were not normally distributed and were analyzed using the natural logarithm. Data were back transformed for reporting purposes. In the event of a treatment by time interaction, parameters are given as their treatment mean within each time point (Pre-Noise and Noise).

CBD tended to increase overall HR ( Table 4 ; P = 0.093), and decreased LF regardless of time point (P = 0.011). All treatments reduced HF compared to control (P < 0.05). AVNN, SDNN, RMSSD, and pNN50 were unaltered by CBD and trazodone (P < 0.05). No HRV variables were affected by the CBD by time nor the trazodone by time interaction (P > 0.05). The CBD by trazodone by time interaction influenced the LF/HF ratio (P = 0.039). During the Pre-Noise time point, trazodone tended (P = 0.061) to increase the LF/HF ratio compared to control and increased the LF/HF ratio compared to the combination of CBD and trazodone (P = 0.038). During the Noise time point, the combination of CBD and trazodone tended (P = 0.083) to reduce the LF/HF ratio compared to control.


There were no period effects on any behavioral variables ( Table 5 ; P > 0.05). With the exception of Facing Door and Tail Relaxed, all other behaviors were affected by time point (Pre-Noise vs. Noise; P < 0.05). During the Noise time point, duration of inactivity (P = 0.011), Glancing Around (P < 0.001), and Ears Moving (P < 0.001) were increased compared to their duration during the Pre-Noise time point. Conversely, the duration of Other Eyes, Ears Relaxed, Ears Erect, and Tail Stiff were reduced during the Noise time point compared to the Pre-Noise time point (P < 0.05). Across both time points, dogs fed CBD tended (P = 0.072) to spend less time focused on something in the room (Other Eyes). Conversely, trazodone increased overall duration of Other Eyes (P = 0.044) and time spent with Tail Relaxed (P = 0.001), but CBD did not alter tail posture (P = 0.753). No behavioral variables were affected by the CBD by time nor the trazodone by time interaction (P > 0.05).

Table 5

Effect of trazodone (T), CBD (C), CBD by trazodone interaction (C*T), time (Pre-Noise and Noise), CBD by trazodone by time interaction (C*T*Time), and period on the duration of behavioral parameters (s) for 1-min immediately prior to (Pre-Noise) and the first minute (Noise) of the noise-induced fear response tests administered after each 7-d treatment period.

Parameter, s Treatment SE 1 P-value
Control Trazodone (T) CBD (C) T+C Trazodone CBD C*T Time C*T*Time Period
Inactive 55.35 56.33 55.21 56.26 1.214 0.329 0.918 0.971 0.011 0.092 0.993
Facing door 37.45 33.90 34.96 37.70 4.198 0.872 0.796 0.217 0.561 0.556 0.786
Glancing around 16.90 15.65 17.93 15.91 3.460 0.396 0.736 0.841 0.142 0.819
Other eyes 5.10 13.33 4.10 5.48 1.885 0.044 0.072 0.182 0.469 0.792
Ears relaxed 11.37 7.76 12.35 11.43 4.913 0.179 0.168 0.422 0.868 0.567
Ears erect 29.33 34.29 29.93 29.80 5.614 0.304 0.408 0.279 0.747 0.982
Ears moving 19.25 17.79 17.20 18.78 2.076 0.970 0.742 0.351 0.457 0.493
Tail relaxed 37.90 49.86 38.93 50.96 4.857 0.001 0.753 0.992 0.611 0.898 0.990
Tail stiff 18.45 5.55 16.39 6.65 4.582 0.002 0.887 0.644 0.010 0.757 0.896

Treatment T+C indicates the combination treatment of CBD and trazodone.

1 The standard error (SE) of the back transformed data was calculated from the confidence limits of the transformed data as follows: SE = (back-transformed upper limit—back-transformed lower limit)/3.92.

Parameters were not normally distributed and were analyzed using the natural logarithm. Data were back transformed for reporting purposes.

These changes between the Pre-Noise and Noise time points may indicate that the fireworks test generated the desired fearful behavioral response. However, the behaviors Glancing Around and Ears Moving could be considered a normal response to hearing a loud noise and may not necessarily indicate a fearful response to the noise. However, since the common fearful behaviors measured—cowering, pacing, vocalizations, etc.—could not be analyzed due to insufficient occurrences, it is difficult to determine if the fireworks test was severe enough to generate a fearful response.


Since the passage of the Agriculture Improvement Act of 2018, which removed industrial hemp from the Controlled Substances Act and removed CBD from the Schedule I drug list, the market for industrial hemp-derived CBD has been able to expand considerably (27). Just 1 year after the act passed, the market was estimated to be $1.2 billion and is expected to grow to over $10 billion by 2024 (28). Much of this growth can be attributed to public perception of the supposed health benefits of CBD, including analgesic, antioxidant, anti-inflammatory, and anti-anxiety effects. However, despite general public opinion that CBD is a safe and effective treatment for these conditions, the lack of scientific clarity on the safety, dosage, and efficacy of CBD makes it critical for continued research in both humans and companion animals.

The present study is one of the first to describe the effect of CBD on the fear and anxiety response of dogs. The fear-response test was developed and validated by Araujo et al. (22), in which dogs were placed in the test room for 9 min and a thunderstorm track was played from 3 to 6 min. A modified version of this test was used in the current study, with a fireworks track being used instead of a thunderstorm track as previous literature has shown a greater percentage of dogs to be fearful of fireworks than of thunderstorms (1). Additionally, because Araujo et al. (22) saw no behavioral differences during the “after thunder” time period, the test for this study was shortened to 6 min, ending immediately after the fireworks track ended. This allowed for the immediate post-test blood sample collection to be obtained within 10 min of the end of the fireworks test.

If cortisol concentrations had decreased with each subsequent period of the experiment, it would have been an indication that the dogs were adapting to the sound stimulus. While there was a period effect on cortisol, it was due to cortisol in periods 3 and 4 being increased compared to period 1. This may indicate a heightened response to the sound stimulus upon repeated exposure, which suggests that the dogs were being conditioned to be stressed in the testing room despite being placed in the room on non-testing days to avoid such conditioning. The potential for conditioned place aversion is a limitation of the crossover design used in this study. It may be beneficial in future work to either include washout periods or utilize a different design to reduce the number of tests administered to each dog to prevent this conditioning; however, the latter would require a larger sample size than is needed when using a Latin Square design.

The lack of a time effect on blood cortisol concentration was also unexpected. It is possible that cortisol concentrations did not change because the fireworks test may not have produced a sufficient change in fear or stress in these dogs. However, Landsberg et al. (9) demonstrated that the use of a thunderstorm noise-induced fear response test—also averaging 90 dB—resulted in a time-dependent change in blood cortisol, with higher concentrations 5 min post-test compared to 1 h pre- and post-test samples. This time effect was not replicated in this study. Instead, cortisol concentrations decreased at each subsequent timepoint, though not enough to produce an overall effect of time. Other studies have also demonstrated that blood and saliva cortisol concentrations peak between 5 and 20 min after noise exposure and begin to decline as early as 30 min post-exposure (23, 29, 30). For this study, while the blood sample taken immediately after the test was taken within this window, it is possible that cortisol levels had not yet peaked after noise exposure. It would be beneficial in future work to take additional blood samples throughout the first hour after noise exposure to better show cortisol changes after noise exposure. Alternatively, it is also possible that the lack of time effect on cortisol may have been due to elevated initial stress due to the use of shelter animals. Franzini de Souza et al. (23) demonstrated differences in endocrine and behavioral responses between laboratory and companion dogs in response to sound stimuli. While shelter animals were not represented in that study, it is possible that increased stress from the shelter environment, transport, and new environment could impact cortisol concentrations and warrants further investigation.

It is also possible that the time of testing influenced cortisol concentrations. Kolevská et al. (31) showed that dogs not undergoing an exercise regimen had the highest blood cortisol concentrations between 1,000 and 1,300 h and the lowest concentrations between 1,600 and 1,900 h. A similar pattern was seen in this experiment, with the highest cortisol concentrations at the 60-min pre-test sample, which would have been taken between 1,200 and 1,400 h, and the lowest concentrations at the 60-min post-test sample period, which would have been taken between 1,400 and 1,600 h. While blood cortisol concentrations in samples taken from 1,300 to 1,600 h were lower than those taken between 1,000 and 1,300 h (31), that effect was not seen in this study. This could indicate that the noise-induced fear response test did in fact affect blood cortisol concentrations, maintaining the elevated levels through the afternoon rather than the normal drop expected from the circadian rhythm of the hormone. These results warrant further investigation, and future work should consider administering the noise test earlier in the day to account for possible influence of the circadian rhythm of cortisol.

In humans, trazodone has been shown to decrease plasma cortisol concentrations compared with placebo and is commonly prescribed for the treatment of anxiety, depression, and to facilitate sleep (7, 32). While trazodone is not currently labeled for use in dogs, off-label use of trazodone is common for the treatment of anxiety disorders as well as to reduce the agitation and distress associated with post-surgery confinement and reduced exercise (33, 34). In this experiment, treatment with trazodone lowered blood cortisol concentrations compared to all other treatments. On the other hand, CBD did not alter plasma cortisol concentrations compared to control in this experiment. In humans, CBD administration has been shown to attenuate the cortisol decrease associated with the circadian rhythm of the hormone (35, 36). While other anxiolytic supplements seem to reduce anxiety in dogs at least in part by reducing the cortisol response to stressors (9), the results of this study may suggest that CBD does not exert an anti-anxiety effect by lowering blood cortisol concentrations. However, Hurd et al. (37) demonstrated a decrease in salivary cortisol when CBD was dosed to humans at ~5 and 10 mg/kg BW, which may indicate that the CBD dosage selected for this study (1.4 mg/kg BW) was too low to exert an effect on cortisol.

Another possibility is that CBD was administered too early in the day of the fireworks test. Recent work with other oral CBD products with similar dosages to this study demonstrated the time of maximum CBD concentration to be around 1.5 h after administration and the half-life of elimination to be between 1 and 4 h (16, 38, 39). However, at the time this study was completed (July 2018), these works on CBD pharmacokinetics had not yet been published, and earlier literature (24) reported much longer half-life for IV administration of CBD. This resulted in CBD treats being administered between 4 and 6 h prior to the test in this study. In the future, it may be necessary to administer treatments within 2 h of the noise test in order for CBD to have the greatest effect. This was accounted for in the administration of trazodone, as Jay et al. (21) reported that the same dose of oral trazodone had a mean half-life of 166 min in dogs.

Even if CBD was administered too early to exert an anxiolytic effect, CBD did appear to inhibit the ability of trazodone to lower blood cortisol in the combination treatment compared with trazodone alone. This observation may support previous work that shows CBD to be a potent inhibitor of the cytochrome P450 family of enzymes, which is responsible for the metabolism of trazodone to its active metabolite, m-chlorophenylpiperazine, in the liver (40, 41). Several studies have highlighted these potential CBD-drug interactions as well as the lack of information regarding CBD doses that can be deemed safe for use—whether administered alone or in combination with other medications (42–45). The potential interaction between CBD and trazodone demonstrated in this study lends support to these concerns. While there has been some work investigating specific CBD-drug interactions (46), it may be inadvisable to administer CBD concomitantly with other products or medications until these interactions are more fully elucidated.

In agreement with previous work in both dogs and other species, oral CBD administration in this experiment was well-tolerated. No gastrointestinal or constitutional adverse events were observed in dogs receiving CBD during this study. Additionally, food consumption and body weight remained consistent throughout the experiment. However, other studies evaluating the safety of oral CBD administration in dogs have reported the potential for adverse events, including lethargy, gastrointestinal issues such as vomiting or diarrhea, and hematological changes such as increases in liver enzymes (16, 39, 47, 48). However, aside from initial bloodwork evaluated upon animal intake from the shelter, hematological changes were not evaluated during this experiment. As increases in liver enzymes may be indicative of altered liver function, the potential effects of oral CBD administration on clinical chemistry parameters should be monitored in future work.

Heart rate variability has been used as a measure of stress and anxiety in a number of animal species, including dogs. In particular, considerable work has been done using HRV as an indicator of canine fear and anxiety in response to stressful stimuli, in which HRV generally decreases and HR increases when animals are under stress, indicating impaired parasympathetic function and autonomic nervous system dysregulation (49–51). The results of this study concur, showing increased HR and decreased HRV—AVNN, pNN50, LF, and HF—during the fireworks stimulus compared to the Pre-Noise time point when no sound was played. As AVNN represents the interval between heart beats, the decrease in AVNN was expected alongside the increase in HR during the fireworks stimulus. The pNN50 is thought to relate to parasympathetic activity and was also expected to decrease with increased stress from the fireworks stimulus (25, 52). The low frequency band (LF) mainly reflects baroreceptor activity in the heart while at rest, but can be generated by parasympathetic, sympathetic, or baroreceptor activity depending on the situation. Unlike other HRV parameters, the LF band is expected to increase with stress as an increase in baroreceptor activity would be expected to accompany a rise in blood pressure (25, 53). This was not replicated in this study, where LF actually decreased during the fireworks stimuli. The high frequency band (HF), or respiratory band, corresponds to heart rate variations related to the respiratory cycle. Unlike LF, HF only reflects parasympathetic activity, and lower HF is correlated with stress and anxiety (54, 55). Because LF can be influenced by both sympathetic and parasympathetic activity while HF is only produced by parasympathetic activity, the LF/HF ratio has been used as a way to estimate sympathetic vs. parasympathetic activity (56). An increased LF/HF ratio is thought to indicate higher sympathetic drive, which would be expected when exposed to stressful stimuli and has been demonstrated in dogs exposed to sound stimuli (23, 29, 57). In this study, however, the LF/HF ratio tended to be reduced during the fireworks track compared to the Pre-Noise time point. This, combined with the reduction in LF, may indicate that the fireworks track was not sufficient to cause a fearful or stress response.

Additionally, the fireworks tract did not alter SDNN nor RMSSD in this study. The standard deviation of interbeat-intervals (SDNN) measures how interbeat-intervals change over time and has been shown to be reduced by stress (25, 52). As such, SDNN is generally measured over a 24 h collection period, though short-term periods have also been used to evaluate short-term variability (58, 59). The RMSSD reflects beat-to-beat variance and is used to estimate vagal mediated changes in HRV, which reflects self-regulatory capacity (56). Reduced RMSSD has been associated with smoking, high LDL cholesterol, and work stress in humans and has been shown to be reduced in sound-sensitive dogs in response to sound exposure (29, 55). As some of the findings of this study concur with previous work and other results conflict with what was expected upon exposure to fireworks, it is possible that the fireworks test was not successful in generating the desired fearful response. However, some of this conflicting evidence may be a result of the ultra-short time frame used for recording HRV, particularly for some variables that are more commonly measured over longer time periods. Future work should consider recording HRV over longer time frames in order to better assess changes. Only HR and AVNN were affected by the period of the experiment, where HR was reduced in period 4 and AVNN was increased in period 4 compared to all other periods. This may suggest that the dogs were acclimating to the fireworks stimulus, a limitation to this study design. Future work should consider either washout periods or a study design that does not require multiple noise-induced fear response tests in order to avoid this problem.

To our knowledge, no work has been done to evaluate the effect of CBD or trazodone administration on HRV in dogs, though there is some evidence that CBD may improve HRV in healthy humans (60). Since an increase in stress and anxiety due to sound stimuli has been shown to increase HR, LF, and the LF/HF ratio while decreasing RMSSD, and HF (23, 29, 57), it was expected that both CBD and trazodone would attenuate these changes. In contrast to these expectations, both LF and HF were decreased by CBD in this study compared to control. Conversely, CBD tended to increase HR, while SDNN, RMSSD, and pNN50 were unaffected by treatment. While the reduction in LF would indicate that CBD attenuated the increase in cardiac sympathetic modulation, the increase in HR and decrease in HF suggest the opposite. Trazodone, again in contradiction to expectations, reduced overall HF in this study, tended to increase the LF/HF ratio during the Pre-Noise time point, and did not affect any other HRV parameters. The combination of CBD and trazodone also reduced HF compared to control and tended to reduce the LF/HF ratio compared to all other treatments during the Noise time point when the fireworks track was playing. The lack of effect on other HRV parameters such as SDNN and RMSSD may be due to the fireworks track not producing a change in these variables rather than a lack of treatment effect. These conflicting results warrant further investigation, particularly considering the lack of information available regarding the effects of both CBD and trazodone on HRV in dogs.

When the fireworks track started, there was a visible change in the demeanor of the dogs compared to both the open field test and the first 3-min block of the noise-induced fear response tests (Pre-Noise). While this may indicate that the fireworks track was able to generate the desired behavioral response, it is also possible that the change in behavior was a result of the dogs’ interest in the noise rather than a fearful response. However, the considerable variability in the type of observed responses makes it difficult to elucidate whether the change was due to fear or if it was just a reflexive response. The predominant response was a decrease in activity, which may or may not have been accompanied by a variety of other behaviors, such as a tucked tail, shaking, or nervous vocalizations like whining. These fearful behaviors would have been a better representation of the behavior changes due to the fireworks test as they have been used to evaluate such changes in other work (9, 23, 29, 57). However, these behaviors occurred too infrequently in this study to allow for statistical analysis. This may be indicative of a lack of behavioral response to the fireworks test. However, as all dogs were selected for this experiment based on the presentation of one or more fearful behaviors during baseline testing, this may simply highlight the variation in behavioral responses to sound exposure. Other anxiolytic supplements and medications have been shown to increase activity or distance traveled using this model (9, 22); however, neither CBD nor trazodone treatment changed activity compared to control. This is particularly surprising for the treatment groups receiving trazodone, which has previously been shown to visibly reduce behaviors associated with a number of stressful situations (34, 61, 62). However, several of these studies relied on owner-completed surveys rather than objective data to assess effectiveness.

In contrast, CBD has been shown to reduce anxious behaviors in mouse, rat, and human models, but at this time there is little to no literature regarding its effect on canine behavior. In mouse and rat models, responses to threatening or unpleasant stimuli were assessed by several methods, including the elevated plus-maze, Vogel-conflict test, contextual fear conditioning, and elevated T maze (63–65). The use of these models has shown that intraperitoneal administration of CBD in doses ranging from 1 to 20 mg/kg produced anxiolytic effects with some responses being dose-dependent (18, 19). Though different models of anxiety were used in rodents, this may indicate that a higher dose is necessary to produce the desired behavioral changes associated with reduced stress and anxiety, particularly if dosed orally due to the considerable first-pass effect on CBD in the liver (24, 66). Future research should investigate the effect of higher dosage of CBD for dogs above the dose tested in this study. Another important consideration is the time of CBD administration prior to noise exposure. As previously mentioned, oral CBD has been shown to have a half-life of

While there was no period effect on any behavioral variables, the lack of behavioral response to treatment could also have been due to acclimation of some of the animals to the firework track. While dogs were selected for inclusion into the study based on their reaction to the baseline noise-induced fear response test, it is possible that the weekly exposure to the stimulus diminished the reaction of some of the dogs during the later tests. This hypothesis is supported by the effect of period on other variables measured in this study, including plasma cortisol, HR, and AVNN. This highlights an important limitation of this study design, where time constraints prevented washout periods. To avoid this issue in future work, dogs could be blocked by their reaction to the baseline test and assigned to just one treatment for the duration of the study. This would eliminate the need for multiple firework tests and would allow baseline and treatment tests to be spaced out over time but would also require a much larger sample size. However, considering the high level of variability in behavioral responses to the fireworks test, it would be difficult to ensure even distribution of dogs even with blocking. If feasible, it would be ideal to utilize the crossover design with longer washout periods to minimize the potential for acclimation to the stressful stimulus. The variability in behavioral responses also makes it difficult to quantify different fear responses. Several of the most common fearful behaviors (shaking, cowering, panting, etc.) were measured, but could not be analyzed due to insufficient occurrences, which may be accounted for in future work by aggregating such behaviors together into one behavioral category. The inclusion of a non-fearful control group should also be considered for future work as it would allow for better evaluation of changes in fearful behaviors in reactive dogs.


The results of the current study do not provide strong support of an anxiolytic effect of CBD in dogs when supplemented at 1.4 mg CBD/kg BW/d. Trazodone, but not CBD, decreased plasma cortisol concentration. When combined with trazodone, CBD appeared to attenuate the effects of trazodone on plasma cortisol. Cannabidiol decreased LF and HF, tended to increase HR, and tended to decrease duration of Other Eyes. Conversely, trazodone increased duration of Other Eyes, increased time spent with tail relaxed, reduced HF, increased the LF/HF ratio.

It would be beneficial in future studies to use increasing doses of CBD to clarify any potential anxiolytic effect, if present, and the dose necessary to elicit that effect. This study demonstrates the considerable variation in canine anxiety behaviors, which makes it difficult to accurately measure the response to treatments. It may be inadvisable to administer CBD concomitantly with other products or medications as the results from this study highlight potential drug interactions associated with CBD use. Considering the increased interest of CBD use in companion animals, continued research is essential to understanding the mechanisms by which CBD may exert anxiolytic effects as well as possible risks, like drug interactions, associated with CBD administration.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics Statement

The animal study was reviewed and approved by Lincoln Memorial University IACUC.