**1. Introduction**

For centuries, the cannabis plant (*Cannabis sativa* and *Cannabis indica*) has been used as a source of food, fiber, and medicine [1–4]. In recent decades, scientific interest in cannabis has increased considerably, as its bioactive compounds have shown promising potential in the treatment of numerous musculoskeletal and neurological diseases in humans [5,6]. Among young people, cannabis is the illicit substance that is most commonly used for recreational purposes [5,7,8], thanks to its anxiolytic effect and the associated sense of euphoria [9]. It is also widely used among pregnan<sup>t</sup> women, mainly to reduce morning sickness, nausea, and vomiting [3,10]. However, because cannabinoids can readily cross the placenta and reach the fetus, they may impact the development of the embryo, which increases the risk of neurological disorders in newborns [3]. The mechanistic pathways by which cannabis and its metabolites affect neurodevelopment and nervous-system functioning remain poorly understood and need to be further investigated. To date, around 500 compounds have been identified in the cannabis plant; these include more than 150 cannabinoids, which generate more than 2000 compounds when smoked [11]. The plant also contains other bioactive compounds of medical and industrial interest, such as phenolics and flavonoids [12]. Cannabinoids and terpenes are abundant in the viscous resin that is produced by the

**Citation:** Licitra, R.; Marchese, M.; Naef, V.; Ogi, A.; Martinelli, M.; Kiferle, C.; Fronte, B.; Santorelli, F.M. A Review on the Bioactivity of Cannabinoids on Zebrafish Models: Emphasis on Neurodevelopment. *Biomedicines* **2022**, *10*, 1820. https://doi.org/10.3390/ biomedicines10081820

Academic Editor: Wesley M. Raup-Konsavage

Received: 1 July 2022 Accepted: 27 July 2022 Published: 28 July 2022

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

glandular structures of the cannabis flowers, called trichromes [4]. The quantitative and qualitative characteristics of the plant are quite variable, with its composition, concentration, and yield greatly affected by the growing conditions, processing, and storage [13,14]. Biochemically, cannabinoids are highly lipophilic substances that are soluble in alcohols, fats, and other nonpolar organic solvents. They can remain associated with cell membranes long after the actual exposure to the substance [10].

In human medicine, cannabinoids are already considered to be antiemetic, antispastic, analgesic, and appetite-stimulating compounds [2,5]. Their therapeutic effects have also been examined in a series of syndromes, including multiple sclerosis [15], Dravet syndrome [16,17], epilepsy [18], fibromyalgia [19], anxiety [20], schizophrenia [21], chronic pain [22], and cancer [23,24]. Many people look to naturally derived compounds, such as cannabinoids, to treat illness and disease because they wish to avoid the strong side effects of synthetic drugs [25]. However, the literature suggests that cannabinoids may unfortunately have serious and undesirable effects, such as dependency, as well as a possible causative association with psychotic illness and cognitive impairment, including deleterious effects on memory [15]. Cannabinoids, for instance, have pronounced effects on the recognition memory and social behavior in pubertal rats, which suggests that the developing brain is sensitive to cannabinoid exposure [26]. Moreover, some studies link cannabis use to adverse birth outcomes, including low birthweights and preterm births, while other studies do not report any negative effects on children [27]. Cannabinoid use is still subject to uncertainty over aspects such as the dosing and side-effect profiles, and there is an overall lack of knowledge of their underlying mechanism of action; clinicians are therefore often reluctant to prescribe cannabis [25,28]. However, cannabis shows a lower potential to cause dependence (8.9%) than do other common substances of abuse, such as cocaine (20.9%), alcohol (22.7%), and nicotine (67.5%) [29]; moreover, it has recently been suggested that susceptibility to psychosis-like symptoms varies between cannabis consumers, as it involves a complex interplay between environmental factors and genetic predispositions [2]. Cannabinoids aside, it is also worth noting that potentially synergistic effects of phytocannabinoids and terpenoids have been reported in the treatment of pain, inflammation, depression, and anxiety [12].

Although the use of cannabis is still illegal in most countries, the cannabis world market is now approaching USD 30 billion, and the profits from illicit trafficking are certainly higher than that [30]. Currently, medications based on both synthetic cannabinoids (e.g., Nabilone® and Dronabinol®) and cannabis extracts (e.g., Sativex® and Epidiolex®) are approved for human use [14,16,24,31]. It is therefore very important for patients, doctors, and the entire scientific community to better understand the effects of cannabis/cannabinoid exposure on health [32].

The two main cannabinoids found in cannabis are Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD) [33]. THC is considered a psychoactive component, while CBD lacks psychotropic activity [10]. THC and CBD are present in the flowers and leaves of the female plants at concentrations ranging between 0.1 and 25% and 0.1 and 2.89% (*w*/*w*), respectively [30]. In addition to CBD and THC, cannabis contains numerous other cannabinoids with known or potential bioactivity [34]. For instance, cannabinol (CBN), the main metabolite of THC, was considered to be an inactive cannabinoid until studies shed light on its biological activities [30,35]. The typical concentration of CBN in cannabis inflorescences ranges between 0.1 and 1.6% (*w*/*w* of dry weight). It forms primarily through the degradation of THC that occurs as the plant ages and as an effect of storage conditions [36]. Other minor cannabinoids present in cannabis are Δ9-tetrahydrocannabivarin (THCV), cannabichromene (CBC), cannabigerol (CBG), cannabigerovarin (CBGV), cannabidivarin (CBDV), and 11-hydroxy-Δ9-tetrahydrocannabivarin (THCV−OH) [14,37]. A novel Δ9- tetrahydrocannabiphorol (THCP) was isolated and reported to have higher in vivo "cannabimimetic" activity than normal THC [38]. Figure 1 shows the molecular structures of the aforementioned cannabis phytocannabinoids.

**Figure 1.** Molecular structures of phytocannabinoids found in cannabis.

THC and its derivatives are studied mainly for their psychotropic properties and other pharmacological activities, including their possible anticonvulsant, antidepressant, hypotensive, bronchodilator, and analgesic actions, as well as their ability to lower intraocular pressure [15]. However, there is also evidence that THC may increase the resilience to certain stressors, as it has been observed that low doses of THC protect against a wide range of neuronal insults, including 3,4 methylene-dioxymethamphetamine (MDMA) and carbon monoxide exposure [27]. In addition, researchers have shown an increasing interest in determining whether THC or other cannabinoids can positively affect neurological health and neurodegenerative disease development in advanced age. This hypothesis is supported by a study that reports that cannabinoids protect against neurodegenerative diseases in many animal models when they are administered in adulthood or advanced age [39]. The anti-inflammatory properties of THC may help to protect the brain against neurodegenerative diseases [40]. Indeed, while high doses of THC can cause memory deficits [41], low doses of THC have been shown to slow or halt Alzheimer's disease (AD) progression by reducing the amyloid beta, which is the main component of the amyloid plaques found in the brains of people affected by AD [42,43], and to restore cognitive function in old mice [44]. Taken together, these findings reinforce the suggestion that the THC doses and patient age determine the beneficial versus detrimental effects of THC on neuronal health [27]. Moreover, the dose seems to influence the resultant behavioral phenotype, as low doses may induce anxiolytic effects, whereas high THC doses generally cause the opposite responses [45]. Some researchers sugges<sup>t</sup> that the plant produces this

molecule in order to protect it from ultraviolet radiation. Indeed, ultraviolet radiation can stimulate cannabinoid biosynthesis [46], and, apparently, the higher the altitude at which cannabis grows, the more THC it produces. Current evidence indicates that even visible LED light can enhance the THC, CBG, and terpene accumulation in the flowers, but not the accumulation of CBD [47].

CBD has been used to reduce bouts of nausea and vomiting, and as an anxiolytic, antipsychotic, antirheumatic, appetite stimulant, and analgesic, as well as a natural remedy for multiple sclerosis and epilepsy [10,25,48]. Moreover, CBD possesses antioxidative and antiapoptotic properties, and it exerts anti-inflammatory effects [7]. It has well-recognized behavioral effects of clinical interest, especially with regard to its anxiolytic properties, and an inverted U-shaped dose–response curve has been reported in several animal models featuring anxiety [48]. In general, CBD does not appear to affect memory formation and may protect against memory impairment [49], but research is still ongoing. Surprisingly, it has been reported that CBD degrades into THC in gastric fluid [50]. As observed with THC, the therapeutic potential of CBD is related to its interaction with the central nervous system (CNS) via several pathways, including the endocannabinoid-system (ECS) pathway, serotonin receptors, and intracellular mechanisms [25,51]. Finally, the CBD:THC ratio also seems to play an important role in determining the symptomatic effects of cannabis [6].

#### *1.1. The Endocannabinoid System*

In 1988, Devane and colleagues [52] discovered a specific brain receptor for cannabinoids. Subsequently, it was understood that the cells equipped with these receptors constitute part of a network of neurons, analogous to that involving dopamine, serotonin, and endorphins, and capable of triggering cognitive, behavioral, or physiological changes. Cannabinoid receptors were found to be particularly expressed in brain areas involved in the control of learning and memory (cortex and hippocampus), motor behavior (basal ganglia, cerebellum), emotions (amygdala), and autonomic and endocrine functions (hypothalamus, pons, and medulla), and they may therefore be involved in the control of numerous neurobiological processes [53,54]. Four years after the discovery of cannabinoid receptors, the same research group [55] isolated an endocannabinoid produced by the human brain. It was named "anandamide" (AEA), after the Hindu term "Ananda", meaning "happiness or bliss". Specifically, there is evidence of a role for AEA in social facilitation, which is closely related to the action of oxytocin [56]. Indeed, the oxytocinergic system is known to regulate social and maternal behavior in mammals [57], and AEA, mediating the action of oxytocin, has been considered crucial for social behavior, and even a possible therapeutic compound for autism-related social impairment [58]. Although the scientific literature on the relationship between zebrafish social behavior and isotocin (teleost homolog of mammalian oxytocin) is quite limited, the effects of bioactive fatty acid amide derivatives on zebrafish bone metabolism [59], growth, and lipid metabolism [60] have been described.

In vertebrates, the ECS involves the cannabinoid receptors 1 and 2 (CB<sup>1</sup> and CB2, re-spectively), endogenous ligands (such as anandamide and 2-arachidonoylglycerol), and the mechanism responsible for receptor and ligand synthesis and degradation [10]. CB<sup>1</sup> has been found to be the most abundant G-protein-coupled receptor within the CNS [5]. CB<sup>1</sup> and CB<sup>2</sup> are activated both by endogenous ligands and exogenous phytocannabinoids, such as THC and CBD [10]. These receptors are typically located presynaptically, and they work as retrograde messengers to decrease the synaptic output. By activating the Gα subunit, cannabinoids are able to inhibit voltage-gated calcium channels and potentiate inwardly rectifying potassium channels [61]. Even though CBD interacts with both cannabinoid receptors, it shows lower affinity compared with THC [3]. Indeed, CBD seems to be 10 times less active than THC on both CB<sup>1</sup> and CB<sup>2</sup> [48]. CB1 controls the vesicular release of gamma aminobutyric acid (GABA) or glutamate by inhibiting voltage-gated Ca2+ channels [62]. Moreover, CB1 is also present in the external membrane of mitochondria [63], where it regulates memory processes via the modulation of the mitochondrial energy metabolism [64].

In addition, several findings have shown that the ECS, through CB1 receptor activation, is associated with the neuronal differentiation and maturation of adult progenitor stem cells into neurons or astrocytes [65], which is a role that could be relevant in the treatment of neurodegenerative diseases. Conversely, although CB2 expression was initially described only in the immune system, more recently, it was also detected in particular brain regions [66], and previous studies in rodents have already reported schizophrenia-related behaviors [67,68], altered cognitive function [69], modified drug-reward behaviors [70], and increased aggressiveness and anxiety [68] in CB2-knock-out mice. By contrast, CB2 overexpression was associated with reduced anxiety-like behaviors and higher resistance to depression in a murine model [71,72]. In this context, it has been suggested that CB2 can regulate the synaptic transmission in hippocampal pyramidal cells and modulate both the gamma oscillation and activity of the sodium–bicarbonate co-transporter, which leads to a hyperpolarization of the neurons [73]. The ECS has also been shown to modulate the expression of neurotransmitters in the basal ganglia that is involved in coordinated movement [74], and it has the ability to control neuronal migration and differentiation by regulating growth-factor activities [10,75]. Through the activation of their receptors, cannabinoids can regulate synaptic neurotransmission, playing a key role in AD, anxiety, epilepsy, multiple sclerosis, Huntington's, and pain perception [39,53,66,76]. While most actions of cannabinoids are mediated through the activation of CB1 and CB2, cannabinoids can produce effects completely or partially independent of the aforementioned receptors, acting instead through other G-protein-coupled receptors, such as GPR18 and GPR55, serotonin receptors (5HT1Rs), and vanilloid transient receptor potential cation channel receptors [10,24], as well as receptors of the dopaminergic, glutamatergic, cholinergic, and opioidergic systems [2,45].

#### *1.2. Zebrafish as a Model System to Test the Bioactivity of Cannabinoids*

Although the bulk of the literature published to date on cannabis and cannabinoids consists of experiments performed using in vitro or rodent models, the zebrafish (*Danio rerio*) has recently gained attention as a powerful in vivo model, combining the experimental efficiency of cell cultures and organoids with the opportunity to study whole living vertebrate organisms [77]. Over the past three decades, the use of zebrafish has helped to further the knowledge and understanding of the neurobiological basis of vertebrate behavior and the pathogeneses of human neurological diseases [2,25,78–80]. Zebrafish show high genetic homology to mammals; the sequencing of the zebrafish genome revealed that 70% of human genes have at least one zebrafish ortholog, and that 84% of genes known to be associated with human disease have a zebrafish counterpart [17]. Many zebrafish genes are duplicated, making the investigation of their functions particularly challenging [81]. One advantage of zebrafish as a model species is that their embryos develop externally, which facilitates the study of embryo development [15]. Zebrafish development progresses quite quickly, with most organs developed within the first hours postfertilization (hpf); muscle activity starts from 17 hpf [81]. Pharmacological screening is among the most common applications of zebrafish [45,77]. At all stages of development, zebrafish can absorb through the skin's small molecules from the surrounding water, and this makes them ideal for performing studies on drug bioavailability and metabolites in a multiorgan system [24]. Moreover, numerous genetic tools, in vivo imaging techniques, and electrophysiological and neurobehavioral assays can be used to study the consequences of drug administration in zebrafish [82–85]. The ECS is highly conserved between zebrafish and mammals—this is not a characteristic of common high-throughput invertebrate model organisms—and ontogenetic analysis has revealed that ECS gene expression begins early during zebrafish development [79]. Recently, the zebrafish ECS has been well characterized: it comprises the same receptors, ligands, and enzymes as its mammalian equivalent [86,87]. Zebrafish larvae begin to express *CB1* mRNA at the three-somite stage; expression is widespread in the CNS (preoptic area, telencephalon, hypothalamus, tegmentum, and anterior hindbrain) at 48 hpf, with the highest expression occurring in the telencephalon at 96 hpf [31,86,88]. In addition,

CB1 protein has been observed in larval zebrafish brain homogenates from 48 hpf through 15 days postfertilization (dpf) [88]. A high level of sequence conservation of CB1 has been shown between zebrafish and mammals. Indeed, the receptor shows 65–69% similarity at the nucleotide level, and 66–75% at the amino acid level [89]. One study showed that the morpholino knockdown of the *cnr1* gene (encoding CB1) led to aberrant patterns of axonal growth and the fasciculation of reticulospinal neurons [90]. These data support the idea that CB1 is needed for brain and locomotor behavior development, even in fish larvae [88]. Less is known regarding the *CB2* expression patterns throughout zebrafish development, but a comparison of the zebrafish CB2 revealed a 39% amino acid similarity with its human counterpart [17]. Elsewhere, after the generation of a CB2-knock-out zebrafish, the resulting homozygote (*cnr2 upr1/upr1*) larvae were shown to be characterized by lower swimming performances and increased anxiety-like behaviors [66]. These findings sugges<sup>t</sup> that zebrafish could be a suitable model for investigating individual ECS gene functions, and for identifying novel genetic modifiers of cannabinoid signaling. Recently, zebrafish were used to test the effects of cannabinoids, administered alone, in combination, and as part of a complex, and were found to offer certain distinct advantages over mammalian models for drug studies [91,92]. The exposure of zebrafish to cannabinoids has been shown to alter a range of behaviors, physiological processes, and gene-expression pathways that are closely related to the ECS [75]. A broad range of behaviors can be analyzed in zebrafish larvae, including multiple swimming parameters, optokinetic and optomotor responses, prey tracking, phototaxis and thigmotaxis, and even learning and memory [93,94]. Due to the rapid development of larvae, these behaviors can be studied within the first week after fertilization. To evaluate the behavioral effects of cannabinoids on zebrafish larvae, most researchers have used the visual-motor-response (VMR) test, which is a validated behavioral assay that measures larval activity first in a light environment, and then in darkness, to study a single transition or dark–light cycles [15,25,31,32]. Typically, zebrafish larvae make frequent low-amplitude movements when exposed to a stable light condition, but an abrupt transition from light to dark causes an immediate increase in their motor activity for 10–15 min, after which it slowly declines to baseline levels [78,86,95,96]. The VMR test has been used to evaluate the sensory-motor function of zebrafish mutants/transgenic lines, and to assess the neurobehavioral responses to nutraceuticals and drugs [78,91]. This behavioral assay thus makes it possible to assess the effects of each compound both on baseline activity and after a standardized stimulus.

To assess the zebrafish anxiety state, and the related efficacy of anxiolytics, two behavioral assays are commonly used: the thigmotaxis paradigm, which is based on an analysis of the preference to swim in close proximity to the tank walls [32], or the light–dark preference test, which is based on the known marked preference of zebrafish larvae for the dark compartment [97]. In the latter test, an increase in activity and time spent in the white/light compartment is considered to reflect anxiolytic behavior, whereas increased activity in the dark compartment indicates anxiety-promoting behavior. Adult zebrafish, due to their size and low housing costs, also provide a cost-effective model for molecularscreening purposes. The most popular, sensitive, and reliable behavioral test in adult zebrafish is the novel tank paradigm, in which the fish locomotor activity and anxiety can be monitored at the same time [98]. Behavioral phenotypes in adult zebrafish are already well characterized [94] and include social, aggressive, affective, and cognitive behaviors [99–101], which are all highly sensitive to a wide range of CNS drugs [102].

We here review the available literature on the use of cannabis/cannabinoids in zebrafish models in order to establish, through a critical analysis of the articles, whether zebrafish might serve as a powerful experimental tool for testing the bioactivity of cannabinoids, and thus for gaining important insights into the safety and efficacy of different cannabis-extract-based products.

#### **2. Materials and Methods**

*Data Sources and Searches*

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The study was registered in PROSPERO; registration number was 344190. The search was conducted by a medical librarian in MEDLINE (via PubMed up to 20 December 2021) using the keyworks "cannabis" (all fields) AND "cannabinoids" (all fields) AND "zebrafish" (all fields). The search yielded 25 matches, but 2 articles were excluded: one because the authors did not discuss the effects of cannabis on zebrafish, and the other because it did not concern zebrafish. The reference lists of these publications were examined, and a further 11 papers were identified. Overall, 34 articles were included in this review. Figure 2 shows a PRISMA flow diagram summarizing the methodology, which was created following the recent indications of Page et al. [103].

**Figure 2.** PRISMA 2020 flow diagram of the literature-search process.

Table 1 gives details of the experimental protocols of all the studies included in the review. It must be emphasized that the single cannabinoids tested were purified standard chemicals, which were used in all the studies, except for one, where THC was purified by using centrifugal partition chromatography [15]. Whereas in the two studies in which the whole-plant cannabis extract was employed, the analytical determination of the main cannabinoids was performed by gas or liquid chromatography coupled with high-resolution mass spectrometry.
