A systematic review of minor phytocannabinoids with promising neuroprotective potential

Embase and PubMed were systematically searched for articles addressing the neuroprotective properties of phytocannabinoids, apart from cannabidiol and Δ9‐tetrahydrocannabinol, including Δ9‐tetrahydrocannabinolic acid, Δ9‐tetrahydrocannabivarin, cannabidiolic acid, cannabidivarin, cannabichromene, cannabichromenic acid, cannabichromevarin, cannabigerol, cannabigerolic acid, cannabigerivarin, cannabigerovarinic acid, cannabichromevarinic acid, cannabidivarinic acid, and cannabinol. Out of 2,341 studies, 31 articles met inclusion criteria. Cannabigerol (range 5 to 20 mg·kg−1) and cannabidivarin (range 0.2 to 400 mg·kg−1) displayed efficacy in models of Huntington's disease and epilepsy. Cannabichromene (10–75 mg·kg−1), Δ9‐tetrahydrocannabinolic acid (20 mg·kg−1), and tetrahydrocannabivarin (range 0.025–2.5 mg·kg−1) showed promise in models of seizure and hypomobility, Huntington's and Parkinson's disease. Limited mechanistic data showed cannabigerol, its derivatives VCE.003 and VCE.003.2, and Δ9‐tetrahydrocannabinolic acid mediated some of their effects through PPAR‐γ, but no other receptors were probed. Further studies with these phytocannabinoids, and their combinations, are warranted across a range of neurodegenerative disorders.

. Interaction with these targets has given CBD status as a neuroprotectant, anti-inflammatory agent and antioxidant (Fernandez-Ruiz et al., 2013;Maroon & Bost, 2018). These features, along with its favourable safety profile in humans (Millar et al., 2019;World Health Organization, 2017) has made CBD, in many respects, a more desirable drug candidate than Δ 9 -THC. CBD has shown promise in several animal models of neurodegeneration as well as clinical trials for Parkinson's, Alzheimer's and amyotrophic lateral sclerosis (Iuvone, Esposito, de Filippis, Scuderi, & Steardo, 2009). Furthermore, a fixed combination of CBD and Δ 9 -THC (1:1) is currently licenced by GW Pharmaceuticals under the brand name Sativex ® to treat pain and spasticity associated with multiple sclerosis (MS), and Epidiolex ® (pure CBD) is licensed to treat Lennox-Gastaut syndrome and Dravet syndrome, which are severe forms of childhood epilepsy. Other cannabis-based medicines (CBMs) are also under development. GW Pharmaceuticals has four compounds (structures are not disclosed) in the pipeline for neurological conditions including glioblastoma, schizophrenia and neonatal hypoxic-ischaemic encephalopathy (GW Pharmaceuticals, 2019).
Phytocannabinoids are highly unique compounds, they are promiscuous in action, modulating a range of pharmacological targets as well as exhibiting high antioxidant capability due to their phenolic structures and the presence of hydroxyl groups (Borges et al., 2013;Hampson, Grimaldi, Axelrod, & Wink, 1998;Yamaori, Ebisawa, Okushima, Yamamoto, & Watanabe, 2011). These features, along with their lipophilicity and ability to act as anti-inflammatory agents, makes them desirable therapeutic candidates for the treatment of CNS disorders, as they can effectively cross the blood-brain barrier (BBB), modulate the immune response, and target the many aspects of neurodegeneration (Deiana et al., 2012). These characteristics have been well established for Δ 9 -THC and CBD but are less well known for some of the minor constituents of the plant. Thus, in order to understand the full therapeutic potential of Cannabis sativa, the pharmacology of the lesser-known components of the plant should be elucidated (Turner, Williams, Iversen, & Whalley, 2017). Given the wideranging neuroprotective effects of Δ 9 -THC and CBD already established, it is not unreasonable to suggest other phytocannabinoids may exhibit similar or more potent neuroprotective properties. Therefore, the aim of this systematic review was to collate all available data on the neuroprotective effects of Δ 9 -tetrahydrocannabinolic acid (Δ 9 -THCA), Δ 9 -tetrahydrocannabivarin (Δ 9 -THCV), cannabidiolic acid These phytocannabinoids were selected based on their abundance in the plant, ease of synthesis, efficacy in other fields (e.g., as anticancer agents or treatments for inflammatory bowel disease), and similarities in their structure to CBD and Δ 9 -THC (which have already shown promise as a neuroprotectants and displayed safety in humans) and are therefore more likely to have neuroprotective potential and exhibit human translatability.

| Eligibility and exclusion criteria
Conference abstracts and review articles were excluded. No restrictions were applied to type of study, publication year, or language. Inclusion criteria were as follows: an original, peer reviewed article that involved the application of emerging phytocannabinoids in an in vivo or in vitro model of neurodegeneration or neuronal damage. Studies that looked at two derivatives of CBG, known as VCE-003 or VCE-002.3 were also included because current research is focused on these compounds, based on their increased affinity for PPARγ. Studies that assessed CBD, Δ 9 -THC, Δ 9 -THC:CBD 1:1 (Sativex ® ), or similar combinations of phytocannabinoids (i.e., different ratios of phytocannabinoids) were excluded from this review. After duplicates and irrelevant articles were removed, the full text was obtained for the remaining articles, and studies were examined for data regarding Δ 9 -THCA, reported mechanistic data, this was also described in Section 3.

| RESULTS
The preliminary search retrieved 2,341 studies, which after duplicates were removed left 1,851. A total of 107 cannabinoid studies were retrieved; once exclusion criteria were applied, 26 articles were considered to be potentially relevant and their full texts obtained.
After additional screening (including reviewing reference lists for any potential studies), 28 studies were included in this review; see Figure 1. Table 1 summarizes the in vitro data included in this review, and Table 2 summarizes the in vivo data.
In HEK293 cells transfected with TRPV1, 2, and 3 channels, CBDV caused a concentration-dependent bidirectional current at TRPV1 channels similar to capsaicin, and capsazepine (TRPV1 channel antagonist) blocked this effect. Furthermore, 5'-iodoresiniferatoxin (5'-IRTX), a selective antagonist of TRPV1 channels counteracted the effect of CBDV in the duration but not amplitude of neuronal burst.
These data suggest that CBDV acts as an agonist at these channels, however, no mechanistic data were reported to determine how these effects were mediated (Schubert et al., 2019).

| Cannabichromene (CBC)
In a model of electroshock seizure, CBC (10-75 mgÁkg −1 i.p. per day) significantly depressed motor activity during the first 10-min interval, but subsequently only the highest dose was effective (Davis & Hatoum, 1983). In vitro, Shinjyo and Di Marzo (2013) found that 1μM CBC increased viability of adult nestin-positive neuronal stem cells when applied in medium without growth factors (B27 medium), by inducing ERK phosphorylation. No antagonist data were presented in these studies.

| Cannabinol (CBN)
Only one in vivo study assessed CBN (5 mgÁkg −1 per day) in an were used in this study (Nadal et al., 2017).

| DISCUSSION
To our knowledge, this is the first systematic review on the neuroprotective effects of lesser-known, minor phytocannabinoids in various models of neurological disease. Data obtained from our search revealed that CBG, VCE.003, VCE.003.2, and CBDV were the most promising candidates as neuroprotectants, while Δ 9 -THCV, Δ 9 -THCA, CBC, and CBN have limited but encouraging data as neuroprotectants. CBG, VCE.003, VCE.003.2, and Δ 9 -THCA mediated their neuroprotective effects at least in part by the nuclear receptor PPARγ. CBDV was found to mediate some of its antiepileptic effects via TRPV1 channels, and Δ 9 -THCV was found to be a CB 1 receptor ligand and a possible CB 2 receptor agonist, but no experiments were conducted to establish whether its neuroprotective action was mediated by CB 1 or CB 2 receptors. No other receptors were investigated, and no studies assessed the neuroprotective potential of CBDA, CBGA, CBGV, CBCV, CBGVA, or CBDVA.
CBG was first isolated in 1964 by the same group that reported the structure of Δ 9 -THC (Gaoni & Mechoulam, 1964). It exhibited antioxidant and anti-inflammatory properties, while displaying no psychotropic effects, as it is a poor CB 1 receptor agonist (Gauson et al., 2007;Navarro et al., 2018;Rosenthaler et al., 2014). CBG is a partial agonist at CB 2 receptors, a potent α 2 -adrenoceptor agonist (EC 50 0.2 nM) and a moderate 5-HT 1A receptor antagonist, as well as interacting with various TRP isoforms including TRPV1 and 2 channels (Cascio, Gauson, Stevenson, Ross, & Pertwee, 2010;De Petrocellis et al., 2012). Studies included here show that these compounds have imaging has revealed marked microglial activation, which was correlated with impairments of neuronal activity (Tai et al., 2007). Microglial activation along with increases in pro-inflammatory mediators has also been detected in post-mortem Huntington's disease brains (Palpagama, Waldvogel, Faull, & Kwakowsky, 2019). Interestingly, microglial mediated neuroinflammation was suppressed with the activation of CB 2 receptors (Ehrhart et al., 2005). However, given VCE-003 and VCE.003.2's protective effects were likely to be CB 1 and CB 2 receptor-independent, their effects on microglial activation are likely to be via a different mechanism, possibly through the activation of PPARγ, which has an important role in regulating the inflammatory response, especially in the CNS (see Bright, Kanakasabai, Chearwae, & Chakraborty, 2008;Villapol, 2018). It is also worth noting that microglial activation can be protective, preserving neurons by secreting anti-inflammatory cytokines such as IL-4 and IL-10 as well as various trophic factors (see Le, Wu, &Tang, 2016, andPöyhönen, Er, Domanskyi, &Airavaara, 2019). In line with these observations, there effectively needs to be a balancing act between enabling some degree of microglial activation to protect neurons, while limiting their overactivation that would ultimately lead to damage. Given that the symptoms of Huntington's disease are currently managed using VMAT inhibitors (such as TBZ) to decrease levels of monoamines, it would be useful to assess whether CBG and its derivates have any efficacy as VMAT inhibitors, or whether their protective effects in models of Huntington's disease are independent of this mechanism. If the latter is the case, future studies should investigate low-dose VMATs (to minimize neuropsychiatric side effects) together with CBG or its derivatives as an adjuvant therapy to assess if there is an additive, or even synergistic, protective effect of these compounds.
Long-term dose tolerability and a lack of accumulation in tissue are both essential features of neuroprotective agents as these drugs are typically taken for life after disease onset. In a study conducted by Deiana et al. (2012), CBG was found to have similar PK profiles in rats and mice but exhibited slower brain penetration in mice. Both animals also had higher concentrations of CBG following i.p. injection compared to oral administration, but interestingly in rats, this did not equate to higher concentrations in brain tissue (Deiana et al., 2012).
From the results in our review, treatment with CBG, VCE-003, and VCE.003.2 was well tolerated and ranged from just 3 days to 10 weeks with two studies dosing CBG orally and seven studies dosing intraperitoneally. Deiana et al. (2012) reported that animals tolerated CBG better after i.p. administration, compared with the oral route. In humans, i.p. dosing is not a viable means of regular administration, and all drugs given orally have a larger side effect profile. Moreover, patients receiving certain oral therapies for neurological conditions, such as levodopa for Parkinson's disease, must also take medications to minimize peripheral effects (Fahn, 2008). Therefore, dose formulation and route of administration for these compounds should be carefully assessed, based on thorough ADME profiling and feasibility of long-term dosing.
CBG exhibited positive effects in two Huntington's disease models, despite one study using oral and the other i.p., administration.
Of note, CBD has already been trialled in Huntington's disease patients; CBD (10 mgÁkg −1 ; 700 mg average daily dose) was given for 6 weeks and resulted in a consistent plasma level of 5.9-11 ngÁmL -1 .
Once treatment had stopped, elimination was between 2 and 5 days, suggesting CBD did not accumulate and remain in plasma longer than 5 days in these Huntington's disease patients (Consroe et al., 1991). Like CBD, CBDV is a agonist at TRPV1/2 and TRPA1 channels, and an antagonist at TRPM8 channels, which may explain similarities in their neuroprotective properties, particularly the action of CBDV as an agonist at TRPV1 channels (De Petrocellis et al., 2011;Iannotti et al., 2014;Scutt & Williamson, 2007). In our review, studies showed that CBDV did not affect neurotrophic levels or epilepsy-related gene expression. Thus, it can be assumed that CBDV mediates its protective effects independent of these pathways (Amada et al., 2013;Vigli et al., 2018). Deiana et al. (2012) reported that CBDV was rapidly absorbed in mice and rats, but there was a higher drug concentration in plasma and brain following oral treatment in rats compared to mice.
Furthermore, while i.p. injection resulted in similar PK profiles in the two species, brain concentrations in rats were higher. This brings into question the differences in the amount of CBDV delivered to the brain in the studies conducted in mice compared with rats presented in this review and whether this influenced study outcomes. Only two studies reported chronic CBDV dosing both in models of Rett syndrome, highlighting the need for future studies to assess the longterm tolerability of CBDV as an anti-epileptic agent and how different species exhibit different bioavailability of this compound, as these will both affect the translatability of CBDV to humans.
Although out of the scope of this review, it is worth noting that CBDV has already been trialled as an anti-convulsant by GW Pharmaceuticals in a phase IIa, placebo-controlled study of 162 adult patients (clinical trial number: NCT02369471/NCT02365610). The drug GWP42006 (which contains CBDV as its main ingredient) was dose titrated (over 2 weeks) up to a 800 mg twice daily dose for a 6-week stable treatment period. However, focal seizures were inadequately controlled with this dose and GWP42006 displayed no difference in efficacy to the placebo control group (Schultz, 2018). While this may cast doubt on the translatability of the evidence presented in this review, it is worth highlighting that the maximum dose in humans from the GW study would be considerably less than if the same dose regimens as the in vivo studies were followed for a 60-kg human. Furthermore, Morano et al. (2020) have suggested that the inability of CBDV to control seizures was in part due to an extremely high response from the placebo group and that the use of purified CBDV may have also influenced the study outcome. Therefore, it is important to exercise caution when extrapolating the findings from the in vitro and in vivo data presented here and what doses may be effective in clinical trials.

Cannabichromene (CBC) was first isolated in 1966 by Gaoni and
Mechoulam and is a non-psychotropic cannabinoid that does not interact with CB 1 receptors (Gaoni & Mechoulam, 1966). CBC is an agonist at CB 2 receptors and TRP channels, acting potently at TRPA1 as well as displaying some activity at TRPV3 and TRPV4 channels De Petrocellis et al., 2008, 2011de Petrocellis et al., 2012;Udoh, Santiago, Devenish, McGregor, & Connor, 2019). CBC (0.001-1 μM) exhibited promising antiinflammatory effects in an in vitro model of colitis, decreasing LPS increased nitrite levels and attenuating IFN-γ and IL-10 secretion in peritoneal macrophages . More recently CBC acted as a CB 2 receptor agonist in AtT20 cells transfected with these receptors and was confirmed by application of the CB 2 receptor antagonist AM630, which blocked the effects of CBC (Udoh et al., 2019). We found only two papers related to neuroprotective effects of CBC; in vivo CBC suppressed motor activity while in vitro CBC improved viability of neural stem cells (Davis & Hatoum, 1983;Shinjyo & Di Marzo, 2013). The anti-inflammatory effects of CBC may play a pivotal role in its ability to act as a neuroprotectant, as inflammation and overactivation of the immune response are important features of neurodegenerative conditions.Thus, further research should assess this compound in neuro-inflammatory conditions, where it may have potential.
Cannabinol (CBN) is an oxidation product of Δ 9 -THC and was the first cannabinoid to be discovered and isolated (Wood, Spivey, & Easterfield, 1899). Like Δ 9 -THC, it has been shown to activate CB 1 receptors (K i 211.2 nM) but with lower potency, as well as acting as an agonist at TRPV2 channels (Rhee et al., 1997;Russo & Marcu, 2017). CBN (1 mgÁmL −1 ) was recently shown to reduce mechanical sensitization and sensitivity of afferent muscle fibres in an in vivo model of myofascial pain, but no mechanism of action was investigated (Wong & Cairns, 2019). From our search, limited data showed that CBN decreased cell damage and acted as a potent antioxidant in a cell-based Huntington's disease model (Aiken et al., 2004). The antioxidant activity of CBN is a characteristic feature of cannabinoids, which as previously mentioned, is thought to be due to the presence of the phenolic ring and carboxyl moieties, as well as the ability to increase antioxidant defences. CBD has already shown extensive antioxidant properties, including increasing the levels and activity of antioxidants, capturing ROS, and transforming them into less active forms, as well as activating nuclear erythroid 2-related factor (NrF2) that governs the transcription of many antioxidant genes (see Atalay, Jarocka-karpowicz, & Skrzydlewskas, 2020). Oxidative stress is a key feature of neurodegenerative disorders including Parkinson's and Alzheimer's disease.
In the latter condition, Aβ deposits contain a significant number of binding sites for biometals (zinc, copper, and iron) that contribute to oxidative stress in patients (Huang, Zhang, & Chen, 2016;Kozlowski et al., 2009). Furthermore, Alzheimer's disease patients have decreased levels of antioxidant enzymes and increased products of oxidative stress, such as peroxidised lipids and oxidized proteins in brain tissue (Kim et al., 2006;Sultana et al., 2011). Also, large amounts of ROS are generated by reactive microglial cells, with studies showing superoxide produced by microglia directly contributing to the death of dopaminergic neurons in Parkinson's disease (Hernandes, Café-Mendes, & Britto, 2013). It is clear that more information is needed on the pharmacology of CBN, especially its antioxidant potential. Moreover, the ability of CBDV, CBG, CBC, and CBN to reduce Aβ deposits in vitro is also noteworthy and it is clearly of interest to examine the antioxidant and anti-inflammatory potential of these compounds in Alzheimer's disease models in vivo and whether these compounds act through mechanisms, similar to those of CBD.
We found two studies where Δ 9 -THCV showed promise as an antiepileptic agent and protected neurons in two models of Parkinson's disease, while García et al. (2011) suggested Δ 9 -THCV mediated some of its protective effects by acting at CB 1 and CB 2 receptors, the possible mechanisms of action of Δ 9 -THCV was largely unexplored (García et al., 2011;Hill et al., 2010). In an earlier study, Δ 9 -THCV displaced [ 3 H]CP55940 from specific sites in mouse brain and CHO-hCB 2 cell membranes (K i values 75.4 nM and 62.8 nM, respectively), and along with data from GTPγS-binding experiments, the authors concluded Δ 9 -THCV acted as a CB 1 and CB 2 receptor antagonist (Thomas et al. 2005). Other groups have shown Δ 9 -THCV can block CB 1 receptor activity in murine cerebellar slices and, at 5.8 μM, increased GABA release from neurons, sharing the same properties as AM251, a CB 1 receptor antagonist (Ma, Weston, Whalley, & Stephens, 2008;Pertwee, 2008). Thus, while there is evidence to suggest Δ 9 -THCV mediates some of its protective effects via CB 1 and CB 2 receptors, the data remain largely unclear, and there is also a lack of investigation into the potential of Δ 9 -THCV to act at other known cannabinoid targets.
Microglial activation and the presence of neuroinflammatory factors are well known characteristics of Parkinson's disease and well documented among patients (Mogi et al., 1994;Qian et al., 2011).
Moreover, studies have demonstrated that microglial overactivation leads to deleterious effects and the exacerbation of the immune response, especially the release of pro-inflammatory mediators. As IL-6, and iNOS (Vallée, Lecarpentier, Guillevin, & Vallée, 2017). Therefore, it would be of interest to determine whether Δ 9 -THCV is able to reduce microglial activation through the same mechanism as CBD, involving the activation of PPARγ.
Limited pharmacokinetic data on Δ 9 -THCV have shown it exhibits rapid absorption in rats and mice when administered either i.p. or orally but is rapidly eliminated when orally administered (<1.5 h) compared to i.p administration where its elimination rate is >5 h (Deiana et al., 2012). Interestingly, Δ 9 -THCV exhibited extensive brain penetration (exceeding plasma levels), regardless of the route of administration, meaning it can effectively cross the BBB. At 24 h, Δ 9 -THCV was no longer detected, suggesting that it exhibits a lack of accumulation in brain tissue (Deiana et al., 2012). Altogether, these features, along with evidence collected in this study, support Δ 9 -THCV as a neuroprotective agent. However, clearly, more data with Δ 9 -THCV are needed, especially to assess safety after chronic dosing and whether this compound exhibits tolerance with long-term use.
Δ 9 -THCA is the acidic precursor of Δ 9 -THC, and competition binding assays revealed that this compound was unable to achieve displacement of [ 3 H]-CP55,940 (CB 1 and CB 2 receptor agonist) up to 10 μM, suggesting Δ 9 -THCA exhibits poor affinity for CB 1 or CB 2 receptors (McPartland et al., 2017). Results from this study also showed that Δ 9 -THCA has little efficacy at these receptors as it exhibited no inhibition of forskolin-mediated cAMP, compared to Δ 9 -THC that acted as an agonist in this assay. Our search revealed that Δ 9 -THCA had anti-inflammatory effects that improved neural viability in a model of Huntington's disease, but interestingly, it did not affect the survival of dopaminergic neurons in a model of Parkinson's disease (Moldzio et al., 2012;Nadal et al., 2017). In a recent study, Anderson, Low, Banister, McGregor, and Arnold (2019) reported that Δ 9 -THCA had extremely poor brain penetration (an optimistic brainplasma ratio of 0.15) in both vehicles tested. Furthermore, studies have shown that Δ 9 -THCA has poor stability and rapidly decarboxylates to Δ 9 -THC, bringing into question whether the ability of Δ 9 -THCA to act as a neuroprotectant in the studies presented here is actually due to nearly unavoidable contamination with Δ 9 -THC (Anderson et al., 2019;McPartland et al., 2017). Overall, these data warrant further investigation into Δ 9 -THCA as a potential neuroprotective and anti-inflammatory agent, but with caution, and such studies should include purity data on Δ 9 -THCA to enhance the robustness of the experimental data.
There were no studies identified in this review that looked at the potential neuroprotective effects of other cannabinoid varins or their acidic forms such as CBGV, CBGVA, CBDVA, CBCV, and CBCVA. This may be due to the lack of commercial availability of these compounds due to their low concentrations in the plant, costly synthetic production or that these compounds are not very stable. CBDA was only used in one study on Huntington's disease, where it had no protective effects.
This compound, however, has shown promise in other conditions including breast cancer migration, inflammatory pain and nausea (Bolognini et al., 2013;Rock, Limebeer, & Parker, 2018;Takeda et al., 2012), with groups suggesting that CBDA is 1,000 times more potent at the 5-HT 1A receptor than CBD (Bolognini et al., 2013). Activation of the 5-HT 1A receptor is protective both in vitro in Parkinsonian models and in vivo in models of hypoxia ischaemia (Miyazaki et al., 2013;Pazos et al., 2013). Although Anderson et al. (2019) concluded that CBDA displayed poor brain penetration in an oil-based formulation, uptake was increased when CBDA was formulated in a Tween-based vehicle. Also, CBDA was anti-convulsant at 10 and 30 mgÁkg −1 displaying greater potency compared to CBD (100 mgÁkg −1 ).
These data support CBDA's efficacy in the brain, as well as highlighting its potential as an anticonvulsant (Anderson et al., 2019). Considering these points, CBDA may be also protective in conditions such as ischaemic stroke and Parkinson's disease and warrants further investigation. Recent studies have also shown that CBDA, CBGV, and CBGA interact with various TRP channel isoforms including TRPV1, TRPV2, TRPA1, and TRPM8 channels. Of note, CBGV and CBGA were also potent desensitizers of TRPV3 and TRPV4 channels, respectively (De Petrocellis et al., 2012). While the extent of the role of TRP channels in neuroprotection has yet to be fully understood, these receptors are involved in a wide range of neurological disorders. For example, TRPA1-deficient mice were more likely to sustain damage post ischaemia and TRPA1 channel activation in Alzheimer's disease may have a crucial role in regulating astrocyte-mediated inflammation (Lee et al., 2016;Pires & Earley, 2018). Conversely, TRPV1 channel activity has been implicated in epilepsy having a role in neuronal excitability and synaptic transmission (Nazıroglu, 2015). Therefore, CBDA, CBGV, and CBGA interactions at TRP channels may be beneficial in conditions that involve these channels in their pathophysiology.
Translatability of these data and the viability of minor phytocannabinoids as neuroprotectants will also rely on understanding and perhaps manipulating their bioavailability and pharmacokinetic properties. In a recent systematic review conducted by our group, Millar, Stone, Yates, and O'Sullivan (2018) highlighted discrepancies regarding CBD bioavailability, C max , T max , and half-life (t 1/2 ) in humans depending on the route of administration and formulation and whether CBD was dosed in a fed or fasted state. That being said, studies conducted in piglets (Garberg et al., 2017) and rodents (Hammell et al., 2016;Long et al., 2012) have shown a dosedependent relationship between CBD administration and brain and plasma concentrations. Limited data extracted by Millar et al. (2018) showed that administration of CBD in humans also led to dosedependent increases in plasma concentrations, suggesting the same may apply to brain concentrations in man.
Information on the human metabolites of CBD, Δ 9 -THC, and other phytocannabinoids is scarce, with the majority of research focusing on the extensive first pass metabolism of CBD and the identification of its urinary metabolites. Of interest, a patent filed by Mechoulam et al. (2010) described that two major metabolites of CBD, 7-hydroxy (7-OH) CBD and 7-carboxy (7-COOH), are both antiinflammatory and dose dependently inhibit TNF-а, NO, and ROS.
However, these data have yet to be confirmed in academic studies or found to be true of other phytocannabinoids. In addition, the cytochrome P450 (CYP) superfamily is responsible for metabolizing 60%-80% of CNS-acting drugs, 23% by CYP3A4 and 38% CYP2C19, both of which accept CBD as a substrate (Cacabelos, 2010;Iffland & Grotenhermen, 2017). Altogether, these findings highlight that there are major gaps in the ADME of phytocannabinoids, as well as a lack of identification of metabolites and whether they have biological effects.
In phase II trials, the minor phytocannabinoids presented in this review will, in all likelihood, be used alongside current therapies to see if they can augment survival of neurons and/or symptom burden, rather than being used as a single agent. In light of the above, it will be essential to consider the interactions that these compounds may have when administered in conjunction with conventional drug therapies (where they exist) and to establish potential synergistic or deleterious effects. Looking forward, initial ADME data will be essential to determine whether these compounds have true clinical potential and for their subsequent formulation and administration.

| CONCLUSIONS
This review aimed to collate and summarise all current data on the neuroprotective potential of phytocannabinoids other than Δ 9 -THC and CBD. Despite the lack of studies available in this area, we found that all phytocannabinoids tested displayed neuroprotective properties in a range of disorders. CBG and its derivatives displayed significant anti-inflammatory effects and were particularly effective in Huntington's disease models. CBDV, Δ 9 -THCV, and CBC were effective as anti-seizure agents, while CBN displayed antioxidant activity and Δ 9 -THCA had anti-inflammatory effects. CBG and Δ 9 -THCA, like CBD, mediate their anti-inflammatory effects through PPARγ. Many of the studies were screening studies that conducted no mechanistic probing, suggesting that research into these compounds is still in its early stages. Extensive pharmacokinetic and pharmacodynamic data in larger mammals are also necessary on these compounds, given that all in vivo studies in this review were conducted in mice and rats. This would provide more evidence for the facilitation of these compounds as therapies in humans. Further studies are required to investigate the full neuroprotective potential of these compounds particularly the mechanisms underlying their protective effects, as well as exploring whether their combinations may enhance their capabilities as neuroprotectants. While we have focused on a select number of minor phytocannabinoids, based predominantly on their shared physical and biological similarities to CBD, there are over 100 phytocannabinoids and terpenes present in the Cannabis plant that could potentially display neuroprotective potential.

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMA-COLOGY (http://www.guidetopharmacology.org), and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander, Christopoulos et al., 2019;Alexander, Cidlowski et al.,