CBD has a chemical structure similar to Δ⁹-THC; however, both differ on the spatial conformation, fact that helps to explain the differences observed in relation to their physiopharmacological properties. Δ⁹-THC presents a planar structure that allows binding to the rCB₁. In contrast, CBD has a slightly angular structure that produces a steric hindrance that hampers its ability to bind to this receptor [31]. As consequence, CBD displays 100 times less affinity for rCB₁ than Δ⁹-THC [32]. This may justify the absence of reinforcing properties of CBD as opposed to Δ⁹-THC [31]. In fact, data achieve to date show that CBD did not induce euphoria or intoxication in healthy volunteers [33,34,35]. Animal studies suggested that CBD may not present reinforcing properties since it did not exhibit drug abuse potential in the conditioned place preference, spontaneous withdrawal and oral self-administration, common animal models used to evaluate the abuse potential of drugs [36,37,38].

During the last years, many researchers studied the potential therapeutic utility of CBD in different diseases pointing out its possible antimicrobial, immunosuppressive, antiemetic, anti-resorptive, spasmolytic, antitumor, antifibrotic, anti-inflammatory, and anticonvulsant efficacy [39,40,41,42]. Some of these properties were further explored leading to its approval for treating spasticity in multiple sclerosis (Sativex) [43] and more recently for treating seizures associated with Lennox-Gastaut or Dravet syndromes in children [44]. Furthermore, other reports suggest that CBD may be useful for treating neurodegenerative [45,46,47,48] and psychiatric disorders [39,49,50,51]. Both animal and clinical studies pointed out that CBD presents anxiolytic, antidepressant, and antipsychotic properties, this will be further explored in detail in the following sections.

Today, over 260 million people worldwide suffer from anxiety and mood disorders, affecting an estimated 25% of the European population. Apart from its high incidence, these psychiatric disorders present high rates of prevalence, leading to a substantial reduction in the quality of life and disruptions in work/school performance, family/social life and common daily activities. In fact, anxiety and mood disorders are the main mental health causes for years lived with disability (YLD), standing at 302 YLD and 850 YLD per 100,000 inhabitants in Europe, respectively [82]. Consequently, both psychiatric disorders entail high economic costs, of around EUR 170 billion per year in Europe.

According to the DSM-V, anxiety disorders are classified into generalized anxiety disorder, panic disorder, specific or social phobias and social anxiety disorder (SAD) [83]. All types share common symptoms, including feelings of uneasiness, panic and fear; sleep problems; not being able to stay calm; being cold and/or sweaty; shortness of breath; heart palpitations; dry mouth; nausea; and avoidance of situations. Depressive disorders present high complexity and may be classified into disruptive mood dysregulation disorder (MDD), persistent depressive disorder (dysthymia), premenstrual dysphoric disorder, substance/medication-induced depressive disorder, associated with another medical condition, other specified depressive disorder and unspecified depressive disorder. Patients suffering from depressive disorders experience emotional, cognitive, physical and behavioral alterations including sadness, anxiety, guilt, irritability, impaired memory, thoughts of death and suicide, loss of motivation, disturbed sleep or appetite, tiredness, neglect of responsibilities, changes in personal appearance, and withdrawal from others [84]. The severity of depressive disorders, assessed by clinician-administered depression assessment scales, such as the Hamilton Depression Rating Scale, varies from one patient to another, with moderate-severe cases presenting worse prognosis [85]. Anxiety and depressive disorders are strongly associated with high rates of comorbidity (around 50%) [86], which can reach 90% in psychiatric patients [87,88]. Comorbidity worsens clinical management and consequently, prognosis; it increases resistance to treatment and recurrence, and it dramatically heightens the risk of suicide.

From a pharmacological point of view, anxiolytics and antidepressants are used in the clinical management of both mental disorders. For instance, benzodiazepines, the most common anxiolytic drug, is useful at the beginning of pharmacological treatment of depressive disorders [89]. Similarly, buspirone (a serotonin 5-HT_1A receptor agonist) is an anxiolytic for treating depressive disorders [90]. Antidepressants, especially selective serotonin reuptake inhibitors, are the most commonly used first-line treatment for anxiety disorders [91]. Although neurochemical alterations underlying anxiety and depression still remain to be elucidated, the beneficial effects found during the co-administration of both types of drugs suggest the involvement of shared neurobiological pathways.

Despite the available pharmacological treatment options, efficacy is limited, especially for preventing relapse and recurrence [92]. For example, one in three patients diagnosed with MDD develops resistance to antidepressant drugs. More importantly, current pharmacological treatments do not improve the cognitive dysfunctions associated with this mental disorder, even when combined with psychotherapy [93,94]. Conversely, the side effects of these medications, such as weight gain, loss of sexual desire and others, affect the risk-benefit ratio. Therefore, it is necessary to find new pharmacological alternatives to improve treatment outcomes for such psychiatric disorders without the burden of disabling side effects. In this respect, published animal and clinical studies, whose main results are detailed below, provide information supporting the anxiolytic and antidepressant properties of CBD.

The potential anxiolytic and antidepressant properties of CBD have been examined in several animal models since the late 1970s. Although preliminary findings were contradictory [95,96], subsequent dose-response studies showed that CBD induced an anxiolytic-like effect, which followed an inverted U-shaped curve, resulting effective at intermediate doses but not at low or high doses [97,98,99]. CBD also attenuated physiological and behavioral responses to stressful situations, reducing restraint stress along with cardiovascular and anxiogenic-like responses [100] by blocking the activation of the hypothalamus-pituitary-adrenal (HPA) axis [101] and activating the 5-HT_1A receptor [100]. Additional results from studies carried out using the Vogel-conflict and the marble-burying tests showed that CBD reduced anxiety- and compulsive-like behaviors, respectively [49,102,103] (Table 2). Curiously, cannabinoid CB₁ receptor (CB₁r), but not 5-HT_1A receptor, appears to mediate such effects [104]. The administration of CBD also abolished anxiety-like behavior, hyperthermia and hyperlocomotion induced by tetrahydrocannabinol (THC), modifying c-Fos expression in brain regions (medial preoptic nucleus and lateral periaqueductal gray) [105]. In contrast, CBD failed to modify anxiety induced by repeated administration of THC [106].

Complementary results indicated that the strain and pattern of administration (single and repeated) may affect CBD actions. In male C57BL/6 mice and spontaneously hypertensive rats, CBD failed to induce any effect [99,105]. Regarding the pattern of administration, chronic treatment with CBD induced an anxiolytic-like effect, whereas acute administration did not [126]. However, another study reported the opposite results [133].

Other authors have evaluated the influence of age and gender in CBD anxiolytic effects. Chronic CBD administration produced an anxiolytic-like effect when given to mice at 5 months, but not at 3 months of age [130]. However, this group of mice significantly reduced its locomotor activity. This may act as a confounding variable when interpreting the results. In contrast, in another study no effects were found in either adolescent or in adult male mice. Moreover, CBD increased anxiety in adult female mice, suggesting sex-dependent effects [128].

Administering CBD also produced interesting findings in specific brain regions. Microinjections into the bed nucleus of stria terminalis [114] and periaqueductal gray [112,113] showed anxiolytic and panicolytic-like effects, respectively. On the other hand, intracisternal or intra-prelimbic medial prefrontal cortex injection of CBD blocked the autonomic activation and anxiogenic-like responses induced by restraint stress [115,139]. These effects appear to be related with 5-HT_1A receptors [112,113,139].

In the fear-conditioning model, a type of associative learning task, acute administration of CBD reduced contextual fear- and anxiety-related behaviors [127,140], whereas chronic administration induced just the opposite [123]. Similarly, microinjections/infusions on specific brain regions, such as intracerebroventricular [117] and prelimbic medial prefrontal cortex, reduced freezing and anxiety [118]. However, infralimbic prefrontal cortex infusion showed contradictory results, increasing freezing [118] or facilitating fear extinction [119], depending on the total number of microinjections given. Indeed, CBD seems to disrupt aversive memory consolidation [109,120], involving anandamide, CB₁r, CB₂r, and peroxisome proliferator-activated receptor gamma (PPARγ) receptors in a time-dependent manner [70,119,120]. CBD also reduced the influence of PFC on corticolimbic circuits, modulating dopamine and immediate gene expression (c-fos and zif-268 proteins) [118,141], and it functionally modified the mesolimbic circuit through the direct activation of 5-HT_1A receptors [142,143].

The efficacy of CBD for reducing fear conditioning, together with its anxiolytic properties, stimulated the development of studies focused on evaluating its potential efficacy in animal models of PTSD, a psychiatric disorder currently classified as trauma and a stressor-related disorder. Acute or sub-chronic administration of CBD reduced the long-lasting anxiogenic-like effects induced by predator stress exposure, suggesting the intervention of 5-HT_1A receptors in these actions [144,145,146]. Furthermore, a new mice model of PTSD conducted by our research team showed that the administration of CBD alone or in combination with sertraline significantly reduced fear conditioning, anxiety-like behaviors and long-term gene expression alterations in the HPA axis, ECS and serotonin systems. These results support the efficacy of CBD, reducing the intense and long-lasting effects of the PTSD model [129].

Along with its anxiolytic properties, CBD displayed antidepressant efficacy in animal models of depression, inducing an antidepressant-like effect when given alone [137] or in combination with sub-effective doses of the antidepressants fluoxetine or desipramine [136], mainly through the activation of 5HT_1A serotonergic receptors [137]. More importantly, CBD showed a rapid and sustained antidepressant effect. A single dose of CBD induced a dose-dependent antidepressant-like effect in Swiss mice, even 7 days after its administration. Similar results were found in Flinders Sensitive and Flinders Resistant Line (FSL/FRL) rats and in Wistar rats [107]. Neuroplasticity changes were also associated, since synaptophysin, postsynaptic density protein 95 (PSD95) and brain-delivered neurotrophic factor were increased on the pre-frontal cortex (PFC) and hippocampus (HIPP) after CBD administration. This effect involves the activation of tropomyosin receptor kinase B/ mammalian target of rapamycin (TrkB/mTOR) signaling [107]. Moreover, treatment with DNA methylation inhibitors (5-AzaD and RG108) and CBD induced an antidepressant-like effect, preventing the alterations induced by stress exposure on DNA methylation and DNA methyltransferase (DNMT) activity in the HIPP and PFC [135]. This study supports the involvement of epigenetic mechanisms on CBD antidepressant properties.

Additional results indicated that the doses and the strain of rodent used may modulate the effects of CBD. In Swiss Webster mice, only the highest dose induced an antidepressant-like effect, whereas no effect was observed in DBA/2 mice [138]. Furthermore, a recent study suggested that the actions of CBD may be gender-specific, since antidepressant-like effects were found in male but not in female FSL rats [110].

Potential differences due to the pattern of CBD administration (acute vs. chronic) were also explored. In C57BL/6J mice submitted to an olfactory bulbectomy, a rodent model of depression, single and chronic administration of CBD induced anxiolytic and antidepressant-like effects. These behavioral alterations were accompanied by increases of serotonin and glutamate levels in the PFC and 5-HT_1A receptor function on the dorsal raphe, CA1–CA2 fields of the HIPP, amygdala and medial PFC [132]. The study further supported the involvement of the 5-HT_1A receptor rather than CB₁r on CBD effects. Similar antidepressant-like effects were observed in Wistar rats [111] and Wistar–Kyoto rats as well as in a genetic model of depression [124] and in animal models displaying depressive-like symptoms such as the diabetic and normoglycemic rats [125]. Moreover, specific brain site microinjections of CBD, e.g., intra-IL or intra-prelimbic, induced antidepressant-like effects in Wistar rats involving 5-HT_1A and CB₁ receptors [121].

The administration of CBD displayed antidepressant-like effects at different doses in adolescents and adult male Sprague-Dawley rats [122]. The long-lasting effects of CBD were different: 2 days for adolescent and 21 days for adult rats. Therefore, the outcome appears to depend on the age at which the CBD treatment was administered. Taken together, these findings are relevant to further explore efficacy and safety depending on the patient’s age and gender.

The effects of CBD were also evaluated for chronic unpredictable mild stress, an animal model that involved the presentation of repeated mild stressors for several weeks. Following this exposure, rodents exhibited depressive-like behavioral alterations, mainly a persistent reduction of their responsiveness to pleasurable stimuli, such as a palatable sucrose solution [147,148]. In this model, different doses and routes of CBD administration prevented anxiogenic and depressogenic-like behaviors and displayed a neuroprotective effect [116,134,149] through CB₁r and CB₂r [116,131]. Indeed, culture cell studies showed that CBD induced progenitor proliferation and cell cycle progression, depending on CB₁r and CB₂r activation and anandamide increase [149]. Overall, these results support the involvement of the endocannabinoid system in the antidepressant-like effects of CBD.

SCZ is a heterogeneous psychiatric disorder with onset in late adolescence or early adulthood [177]. While heterogeneous, the symptoms are classified into three main categories: positive symptoms (hallucinations, delusions, disorganized thoughts, and senseless speech, bizarre behaviors); negative symptoms (social withdrawal, anhedonia, lack of emotional and facial expression, reduced speech, reduced ability to begin and sustain activities); and cognitive dysfunctions (impaired executive function, working memory and attention) [177,178]. SCZ affects only 1% of the worldwide population; however, it is a subject of intense research due to the limited efficacy of antipsychotic drugs [179]. Current treatments improve only positive symptoms following the first episode of psychosis in just 50% to 70% of patients; they present moderate efficacy for negative symptoms and have no effect on cognitive deficits [180]. At the same time, antipsychotic drugs induce severe side effects, including extrapyramidal symptoms, hyperprolactinemia, and cardiovascular complications or interval QT prolongation (depending on the type of the antipsychotic drug), limiting their chronic use [180]. New antipsychotic drugs have a better risk-benefit balance, but they still show limitations for safety and efficacy. Thus, there is a need to identify new, more effective, safer drugs for the pharmacological management of SCZ [181]. In this respect, CBD has been proposed as a new potential treatment based on findings from several preclinical studies, and more recently in clinical trials, showing its antipsychotic effects [182].

The development of animal models for complex psychiatric disorders, such as SCZ, has been instrumental in increasing our understanding of the neurobiological basis of this disorder and for identifying novel antipsychotic drugs [183]. Different experimental approaches have been used to reproduce the main features of SCZ, mostly in rodents (Table 4). Depending on the type of the manipulation used to induce these alterations, rodent models are classified into developmental models (e.g., maternal immune system’s reactivation); pharmacological models (e.g., amphetamine or ketamine administration); and lesion (e.g., neonatal ventral hippocampal lesion) or genetic (e.g., deficient functioning of the DISC1 gene) manipulation models [183]. Jointly, they enable the reproduction of some behaviors simulating positive and negative symptoms, and to a lesser extent, cognitive impairments.

A large number of these animal models have been used to assess the potential efficacy of CBD for modulating SCZ-related behavioral and neurobiological alterations. One of the most frequent symptoms in people with SCZ is psychomotor agitation, which is pharmacologically reproduced in rodents by administering dopamine receptor agonists such as amphetamine or dexamphetamine. Antipsychotic drugs can modulate this drug-induced motor hyperactivity. In this model, a high dose of CBD reduced amphetamine- and dexamphetamine-induced motor hyperactivity, without causing additional motor effects [126,196]. Similarly, CBD normalized ketamine-motor hyperactivity when given chronically, but not acutely. Interestingly, CBD did not induce catalepsy, showing a similar profile as atypical antipsychotics [196].

Another frequent symptom in schizophrenia is the inability to filter out irrelevant stimuli or make associations for further processing, both effects that are linked to alterations in sensorimotor gating. In animal models, these alterations are measured by the pre-pulse inhibition (PPI) of the startle response, which enables the evaluation of SCZ-like behaviors and the efficacy of new potential antipsychotics. In this model, systemic or intra-nucleus accumbens (NAcc) pre-treatment with CBD attenuated the amphetamine-induced PPI alterations in Swiss mice [197]. In the same study, the authors reported similar results after pre-treatment with the anandamide hydrolysis inhibitor URB597, suggesting that the improvement achieved with CBD may be related with its ability to increase anandamide availability [67,200]. These results are consistent with those found in a clinical study, further explained in the next section of this review, studying the parallels between the improvement of SCZ-related symptoms following administration of CBD and the increase in plasma concentrations of anandamide [201]. Furthermore, similar PPI and locomotor hyperactivity normalization were found in rats pre-treated with CBD. These behavioral alterations may be associated with the regulation of mTOR/p70S6 kinase pathways phosphorylation in the NAcc shell [185].

CBD has displayed interesting effects in other SCZ animal models. Saletti et al. showed that acute CBD administration fully normalized PPI alterations induced by MK-801, a non-competitive antagonist of NMDA receptors, in capuchin monkeys (Sapajus spp.) [202]. Similarly, both acute [198] and chronic [192] administration of CBD modulated the PPI impairment induced by MK-801 in mice, involving, at least in part, TRPV1 receptors [67,198,203]. Chronic administration of CBD also regulated the impairments induced by MK-801 administration in social interaction and novel object recognition tests in mice, behaviors that try to simulate the negative and cognitive symptoms of SCZ. In this study, the high dose of CBD showed the same efficacy as the antipsychotic drug clozapine [193]. Moreover, social interaction and novelty object recognition tests revealed protective effects of CBD when administered after the end of the MK-801 chronic administration, in which 5-HT_1A receptors appear to play a relevant role [194]. Despite these promising results, in other rodent studies, CBD slightly modulated the PPI impairment induced by MK-801, without normalizing locomotor hyperactivity or social interaction [184,186]. However, pre-treatment with CBD avoided both alterations [187].

More recently, genetic animal models of SCZ, such as the mutant mice of neuregulin1 (Nrg1 HET), have been used to evaluate the potential antipsychotic-like effects of CBD. Neuregulin1 is a protein involved in neuronal migration, myelination and the regulation of glutamatergic NMDA and GABAergic GABA_A receptors expression (for a review, see [204]). Chronic administration of high doses of CBD increased social interaction in mutant mice, with a modest recovery of PPI impairment [199]. Authors identified an increase of GABA_A receptor binding in the granular retrosplenial cortex of the mutant mice treated with CBD, suggesting that these GABAergic receptors may be partly responsible for CBD-induced behavioral modulation. In fact, some authors have proposed that CBD may act on GABAergic and glutamatergic systems indirectly, but not exclusively, through its direct action on different targets of the ECS, serotonergic or opioid systems [205]. However, recent in vitro [206] and in vivo [207] studies suggested that CBD modulates the GABAergic system by acting directly on GABA_A receptors. Consequently, the effects of CBD on the GABAergic circuits may be the result of both direct and indirect modulation of this system. More studies are needed to further investigate the role of the GABAergic system in the antipsychotic-like effects of CBD.

Epidemiological human studies revealed that exposure to adverse events during pregnancy increases the risk of developing SCZ later on [208]. For this reason, the number of studies attempting to simulate SCZ-like behaviors by exposing pregnant rodents to different disturbances greatly increased in recent years. One of these models is based on the administration of polyinosinic: polycytidylic acid (poly I:C) or the anti-mitotic agent methylazoxymethanol acetate (MAM) on early gestational days (GD) to induce the activation of the maternal immune system. In mice exposed to poly I:C (GD 9 or 15), CBD normalized the increased motor activity [195] and reduced alterations on recognition, working memory, and anxiety [188]. The authors found a normalization of CB₁r and glutamate decarboxylase in PFC and HIPP, respectively [189]. Similarly, chronic CBD administration at early ages of development modulated long-term behavioral and neurobiological consequences, including CB₁r brain alterations, induced by MAM administration on GD 17 [191]. Previous studies suggested the involvement of CB₁r receptors in the antipsychotic-like effects of CBD—one of the first mechanisms described [26,209]. However, the interaction between CBD and CB₁r is controversial. On the one hand, CBD appears to activate CB₁r by increasing anandamide levels, probably by inhibiting its reuptake and metabolism [201,210]. Conversely, some reports suggested that CBD may act as a negative allosteric modulator of CB₁r [65,66,211,212]. Consequently, more studies are needed to further explore the role of CB₁r on the antipsychotic actions of CBD. Similarly, the implication of CB₂r on CBD antipsychotic-like actions could be evaluated, since this cannabinoid receptor has been related with SCZ in rodents and humans [213,214], and CBD appears to act as an inverse agonist of such receptors [65,67].

Muscarinic M1/M4 receptors and choline acetyltransferase were also associated with the modulating effects of CBD on poly I:C induced-behavioral alterations [190]. In addition, an in vitro study showed that CBD may also act on dopamine D2 receptors, inhibiting dopamine binding in the homogenized striatal tissue of rats [79]. Notably, the modulation of dopaminergic activity by CBD seems to be brain-region specific, since its administration to patients with psychosis and Parkinson’s disease modulated psychotic symptoms without worsening motor activity [215]. Thus, the selective modulation of the dopaminergic system in the striatum enable an antipsychotic effect without extrapyramidal side effects. In addition, both pre-clinical [216] and clinical [201] studies showed that, unlike typical antipsychotics, CBD does not increase plasma prolactin, adding more evidence to support its good safety profile.

Therefore, the existing scientific results suggest that CBD may be useful for modulating SCZ-related features, with a pharmacological profile similar to atypical antipsychotics [182], involving a variety of mechanisms. However, further studies are needed to increase the understanding of CBD efficacy and safety in SCZ.

The promising results found in animal models have encouraged the development of clinical trials to evaluate its therapeutic utility for managing people who have or are at high risk of schizophrenia (Table 5). Two clinical trials evaluated the effects of chronic CBD administration [217,218] in stable antipsychotic-treated patients with SCZ. In the first, CBD did not produce changes in positive or negative symptoms, as assessed on the MATRICS Consensus Cognitive Battery (MCCB) and Positive and Negative Syndrome Scale (PANSS) in comparison with placebo. In addition, CBD failed to produce any improvement in cognitive impairments, evaluated with the MATRICS Consensus Cognitive Battery (MCCB) scale. On the other hand, CBD did not induce movement alterations, clearly a great advantage compared with current antipsychotic drugs [217]. Only sedation was significantly prevalent in the CBD-treated group compared to placebo.

In the second clinical study, a multicenter randomized controlled trial, CBD significantly improved positive psychotic symptoms (PANSS). There was also a tendency to increase cognitive performance (Brief Assessment of Cognition in Schizophrenia, BACS) and overall functioning (Global Assessment of Functioning, GAF). The administration of CBD did not modify prolactin concentrations in plasma, Simpson Angus Scale rating, weight, waist circumference, liver function, inflammatory markers, or HDL cholesterol levels—common harmful effects of current antipsychotic drugs. The prevalence of adverse events was similar in CBD- and placebo-treated patients, though there was a high proportion of mild gastrointestinal events in the CBD-treated group [218].

Similarly, in a double-blind, randomized clinical trial, CBD led to significant improvements on the PANSS scale, comparable to amisulpride, but with fewer extrapyramidal symptoms, less weight gain and a lower prolactin increase. Furthermore, CBD was well tolerated and did not significantly affect hepatic or cardiac functions. Therefore, the safety profile of CBD was better than the atypical antipsychotic amisulpride. There was also an increase in anandamide plasma concentrations in schizophrenic patients treated with CBD, highlighting this as a potential mechanism of action underlying the effects of CBD [201].

Additional clinical studies using functional magnetic resonance imaging (fMRI) showed that a single dose of CBD attenuated the reduced activity found in the mediotemporal, prefrontal and striatal brain regions of schizophrenic patients while performing verbal paired learning tasks. CBD also attenuated hippocampal-striatal functional connectivity in these patients compared to healthy controls [219]. Neuroimaging studies have also used the fMRI technique in antipsychotic-naïve patients at clinical high risk for psychosis during a verbal learning [220] or a monetary incentive delay task [221]. In the verbal learning task, a single dose of CBD improved the activation in the right caudate and in the parahippocampal gyrus and midbrain during encoding and recall, respectively [220]. In addition, CBD attenuated the hyperactivation of the left insula/parietal operculum and normalized the reaction time in the monetary incentive delay task [221]. However, CBD did not improve selective attention in schizophrenic patients, assessed by the Stroop Color Word Test. Despite these results, authors did not discard a possible beneficial effect after the chronic administration of CBD [222].

Currently, three ongoing clinical trials (NCT03883360 [223]; NCT02926859 [224]; NCT04411225 [225]) are assessing the efficacy of CBD versus placebo or olanzapine in psychosis and SCZ. The results of these clinical trials should be available in the next few years, providing evidence about the potential usefulness of CBD in psychotic disorders.

In summary, although the clinical studies are heterogeneous, the results found suggest the potential of CBD as monotherapy or as an adjunctive treatment for SCZ. However, more double-blind, placebo-controlled clinical trials are needed to evaluate effectiveness and clarify its profile of side effects.