IntroductionGrowing evidence supports the therapeutic potential of cannabis and its constituent phytocannabinoids in treating a range of neurological and psychiatric conditions (Friedman and Devinsky, 2015; Montero-Oleas et al., 2020; Sarris et al., 2020; Serpell et al., 2014). Cannabis (Cannabis sativa (C. sativa)) contains around 100 cannabinoids, as well as various non-cannabinoid phytochemicals such as terpenes, flavonoids and phytosteroids (Baron, 2018; Fordjour et al., 2023). Among these compounds, Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD) are the most abundant and extensively studied phytocannabinoids.
THC is increasingly recognised for its therapeutic benefits. It is approved as an anti-emetic in cancer treatment, has shown anti-spastic effects beneficial for managing multiple sclerosis (MS), and provides analgesic relief for chronic pain (Argueta et al., 2020; Collin et al., 2007; Davis, 2016; Marinelli et al., 2017). Pharmacologically, THC acts as a partial agonist at both cannabinoid receptors 1 (CB1R) and 2 (CB2R; Paronis et al., 2012; Pertwee, 2008; Sim et al., 1996). While its euphoric effects are primarily mediated through CB1R activation (D’Souza et al., 2005; Sherif et al., 2016), its therapeutic effects likely involve additional actions, including interactions with CB2R, and modulation of glutamatergic and opioid signalling pathways (Baron, 2018; Leinen et al., 2023).
CBD is another prominent constituent derived from C. sativa (Grotenhermen and Muller-Vahl, 2012), widely recognised for its therapeutic potential. Unlike THC, CBD lacks any euphoric effects (Pertwee, 2008), likely due to its distinct pharmacological profile, showing minimal affinity for CB1R and CB2R. Plant-derived highly purified CBD (>98%) is approved as Epidiolex® for the treatment of seizures associated with Dravet syndrome, Lennox–Gastaut syndrome and tuberous sclerosis complex in the US and other territories, and anticonvulsive efficacy of CBD has also been demonstrated in animal models of seizure and epilepsy (Gray et al., 2020; Klein et al., 2017; Patra et al., 2019). Generally, CBD has been shown to have diverse actions; preclinically, there is potential evidence in treating neurodegenerative diseases such as Alzheimer’s and Parkinson’s disease by reducing neuroinflammation and enhancing neuronal protection. Its anxiolytic properties suggest applications for anxiety-related disorders, while it also possesses anti-psychotic properties, offering benefits without the euphoric effects associated with THC (Bergamaschi et al., 2011; Crippa et al., 2004; McGuire et al., 2018; Moreira and Guimaraes, 2005; Zuardi et al., 2012a).
Current evidence suggests that CBD’s diverse actions result from multiple, only partially understood mechanisms, including its negative allosteric modulation of CB1Rs (Laprairie et al., 2015; Jakowiecki et al., 2021; Ibeas Bih et al., 2015) and modulation of glycine and opioid receptors (Baron, 2018). CBD also reduces neuronal hyperexcitability and inflammation through modulation of transient receptor potential vanilloid (TRPV1) channels, GPR55 antagonism and inhibition of adenosine uptake via the ENT-1 transporter (Gray et al., 2020). Additionally, its interactions with serotonin 5-HT1A receptors may contribute to its anxiolytic-like and neuroprotective effects (Baron, 2018; Leinen et al., 2023; Stella, 2023).
Nabiximols, marketed as Sativex®, is a botanically derived agent containing a wide variety of cannabinoids and non-cannabinoids with the most abundant cannabinoids being THC and CBD in a 1.08:1 ratio, that has been approved in several countries for the treatment of spasticity in patients with MS (Patti et al., 2022). Unlike plant-based cannabis extracts, which can have varying concentrations of THC and other cannabinoids, nabiximols offers a consistent formulation with a higher concentration of CBD. This balanced ratio is thought to enable higher THC doses while minimising undesirable side effects, as CBD may partially counteract THC’s euphoric and sedative properties, as well as some of its pharmacodynamic effects (Britch et al., 2017; Boggs et al., 2018; Zuardi et al., 2012b). The combination of THC and CBD has also shown to be effective in chronic pain management (Henson et al., 2022). It is also thought that nabiximols may provide therapeutic benefits for inflammation, digestive disorders and anxiety (Bilbao and Spanagel, 2022; Bridgeman and Abazia, 2017), although the formulation has not received marketing authorisation to treat these disorders. However, the precise mechanisms governing the interaction between CBD and THC remain unclear and are likely influenced by factors such as the compounds’ ratio and the timing of their administration and individual responses to both THC, CBD, metabolites and other non-phytocannabinoid species. Gaining further insight into the distinct central effects (endophenotypes) induced by CBD, THC or their combination in controlled settings will refine drug development strategies and deepen the understanding of their therapeutic potential (Bilbao and Spanagel, 2022; Turner et al., 2017).
Non-invasive brain imaging is increasingly used to capture drug effects on the brain in both clinical and preclinical settings (Khalili-Mahani et al., 2017; Reneman et al., 2021). Functional magnetic resonance imaging (fMRI) can reveal fluctuations in brain blood flow and oxygenation linked to drug-induced inhibitions and excitations within distinct neural circuits (Bifone and Gozzi, 2012; Borogovac and Asllani, 2012; Grade et al., 2015; Matthews and Jezzard, 2004; Zhang et al., 2018). Here, we used these imaging techniques – specifically, blood oxygenation level dependent (BOLD) contrast fMRI and arterial spin labeling (ASL) – to examine how two cannabinoids, CBD and THC, and their combination (nabiximols, THC:CBD in 1.08:1 ratio), alter brain connectivity and regional blood flow, respectively, in anaesthetised rats. Given the ethical and logistical challenges of drug testing in humans, preclinical pharmacological imaging provides a valuable means to identify functional connectivity (FC) signatures that can be compared to those of novel and putative therapeutic compounds in future studies.
Our approach builds on prior neuroimaging studies indicating that cannabis constituents like THC and CBD modulate brain activity (Grimm et al., 2018; Madularu et al., 2017; Mathew et al., 2002; van Hell et al., 2012; Volkow et al., 1996). We conducted whole-brain analyses along with targeted assessments of preselected regions, informed by clinical literature (Bossong et al., 2019; Volkow et al., 1996; Wall et al., 2019). By integrating multimodal analyses across scales, we aimed to establish comprehensive endophenotypes following acute administration of each intervention and model treatment-induced changes in FC.
DiscussionA combination of fMRI, to assess the brain FC, and ASL, to measure regional CBF, was used in rats to identify endophenotypic signatures of the acute pharmacological effects of THC, CBD, and their combination, THC:CBD. We began by applying global measures of whole-brain connectivity using network-based statistics and graph theory. This was followed by characterisation of connectivity within specific subnetworks and, finally, in several a priori selected regions of interest. CBF data were initially analysed independently to quantify blood flow in brain regions. Subsequently, we explored the covariance between BOLD fMRI and CBF using both linear regression and partial least squares analyses.
The difference in FC from vehicle, depicted as connectograms (Figure 2), illustrates the pronounced distinction between cannabinoid compounds. THC elicited the most changes, with approximately 1% of connections showing statistically significant alterations, involving both increases and decreases relative to the vehicle. Increases were primarily in the cortical areas, while decreases were observed in subcortical and posterior regions, including the cerebellum. The THC:CBD group had fewer connections surpassing the statistical threshold, and there were even fewer in the CBD group. Neither CBD nor THC:CBD showed overall network-level significant differences to vehicle when correcting for multiple comparisons using NBS.
Whole-brain changes were further probed using graph theoretical metrics (Figure 3(b)). To this end, THC significantly increased two out of three global measures: FC and CC, whereas GE remained unchanged. Increased FC suggests that the regions are more synchronised that is, correlated with respect to temporal activity, whereas increased CC reflects a higher level of segregation, which impacts on the efficiency of information processing (Sporns, 2018). Increases in FC and CC – both might in fact reflect a level of disorganisation (Hillary et al., 2015) which, coupled with increased perfusion (see below, Figure 7 and Table 1) could indicate higher overall load on the brain, consistent with the euphoric action of THC (Burggren et al., 2019).
In contrast, THC:CBD increased FC only, with GE and CC unchanged (Figure 3(b)). This finding corroborates the suggested tempering effect of CBD on THC, perhaps lowering its euphoric properties as previously suggested (Boggs et al., 2018; Britch et al., 2017; Gunasekera et al., 2022; Stella, 2023; Zuardi et al., 2012b). Interestingly, CBD alone significantly decreased all three global metrics (FC, GE and CC) – an effect that could be related to its proposed anxiolytic properties (Schouten et al., 2023).
Graph theory metrics were next calculated within smaller networks: the Default Mode (DMN), Salience and BGNs, which were selected a priori based on the proposed effects of cannabinoids on the brain (Batalla et al., 2020; Gunasekera et al., 2022; Figure 4). In agreement with the whole brain results, THC in comparison to vehicle significantly increased most metrics in all three networks, with the exception of CC in the BGN and DMN, and GE in the DMN. Conversely, neither THC:CBD nor CBD showed significant effects compared to the vehicle, although there were some significant differences between the CBD and the THC group, demonstrating the divergent effects of CBD and THC where CBD commonly shows the lowest mean value while THC often has the highest. Moreover, while no significant differences were found between THC and THC:CBD, the study was likely underpowered to show these, as the data demonstrate a trend suggesting that CBD attenuates the effects of THC.
Voxelwise ‘seed-based’ analyses (Figures 5, 6 and Supplemental Figure 2) evaluated correlations between the whole brain all-voxel signal versus time courses of signal from the specific a priori selected ‘seed’ regions. Cannabinoid receptor-rich areas (Batalla et al., 2020; Gunasekera et al., 2022; Stella, 2023) – the dorsal hippocampus, cingulate cortex, striatum and sensorimotor cortices) – showed divergence between the three compounds. THC caused the most pronounced changes throughout the brain, with significant increases in inter-cortical and cortical-hippocampal connectivity, and some notable decreases in seed-to-brain covariance in the posterior regions, notably the brain stem, and the cerebellum. The THC:CBD combination produced a similar, but less pronounced and more diffuse pattern, lacking the posterior decreases seen with THC alone. CBD alone showed no significant changes and only a few small magnitude, but widespread, regional covariance alterations.
Divergence between these compounds is further highlighted in the averaged regional FC differences between the drugs and the vehicle (Figure 6): connectivity was primarily increased with THC, negatively modulated by THC:CBD, and predominantly decreased with only a weak increase in some ROIs in the CBD group. This underscores that THC significantly altered overall brain connectivity and efficiency, and this was modulated by the presence of CBD. The effects of CBD alone were more subtle, pointing toward generally weakening connections across the brain.
Preclinical literature reveals a paucity of fMRI studies with cannabinoids that use clinically relevant methodologies similar to ours. Broadly agreeing with our results are two studies in rats that employed BOLD-sensitive fMRI and showed both BOLD signal increases and decreases with THC (Madularu et al., 2017), whereas CB1R agonist HU210 showed mostly increases (Shah et al., 2004). However, direct comparisons with these studies are limited by methodological considerations regarding the timing of drug administration and the dose(s) used, and importantly not measuring FC or blood flow as we have done. Moreover, their approach is limited to studies where a drug is administered acutely in the scanner, which affects clinical translatability.
The existing clinical (human) literature on imaging and cannabinoids is highly heterogeneous with respect to methods, subjects and paradigms (Cupo et al., 2021; Gunasekera et al., 2022). Most human studies focus on stimulus- or task-related fMRI, unlike our focus on a stimulus-free paradigm. While these studies report mixed effects of THC and CBD on task-related activations, they are not easily generalisable to resting-state fMRI studies performed in anesthetised animals. Moreover, many are conducted in (chronic) cannabis users, who may also use other substances, complicating the interpretation of fMRI results due to observed baseline perfusion alterations as a result of chronic substance use (Filbey et al., 2018), and therefore potential receptor changes as well as test substance heterogeneity.
There are variable results amongst clinical studies that used acute THC or cannabis in the naïve or washed out subjects to measure resting-state BOLD signal fluctuations. Van Hell et al. (2012) and Klumpers et al. (2012) reported resting signal increases, aligning with our results of increased FC. Conversely, Ramaekers et al. (2016) and Mason et al. (2019) both found decreased connectivity between nucleus accumbens and the cortical regions (but did not examine any other areas), Bossong et al. (2019) detected decreases in one out of four regions examined, whereas Grimm et al. (2018) reported no acute effect of THC, possibly due to too low plasma exposure (ca. 0.4 ng/mL compared to typical values of 15–30 ng/mL (Lawn et al., 2023)). Wall et al. (2019, 2022), like us, demonstrated that the effects of THC-rich cannabis are attenuated in a CBD-rich THC + CBD combination; however, they observed THC-induced decreases in several networks, and not increases. Comparisons with and between such studies are difficult as the majority are conducted with highly selective a priori hypotheses and restricted to very few regions or networks. Where whole-brain dynamic FC was assessed (Zaytseva et al., 2019), both increases and decreases were seen in the brain connectome after cannabis in occasional users, with a particular brain state of hyperconnectivity being associated with an intoxication element of the drug, echoing our results of increased connectivity. Although we did not measure chronic effects of the drugs, it is interesting to note that in chronic cannabis users, similar hyperconnectivity of the brain was reported (Ramaekers et al., 2022; Vergara et al., 2018).
In addition to rs-fMRI, we measured cerebral perfusion using ASL, potentially enhancing the translational relevance of our findings (Jonckers et al., 2015; Khalili-Mahani et al., 2017). ASL provides a non-invasive, quantitative assessment of brain perfusion, specifically regional CBF (Wang et al., 2011). Given the direct relationship between CBF and neural firing, CBF is often used as a proxy for changes in neural activity (Attwell and Iadecola, 2002; Hosford and Gourine, 2019; Paulson et al., 2010). Our results show that both THC and the THC:CBD combination increased CBF in cortical and subcortical areas across most regions analysed, while CBD alone did not affect CBF (Figure 7). In contrast to rs-fMRI findings, where CBD modulated THC’s effects, the addition of CBD did not alter THC’s impact on CBF. Additionally, since the dose of CBD in the ‘CBD alone’ condition was higher than in the THC:CBD mixture (150 vs 10 mg/kg, respectively), it remains unclear what effect, if any, CBD at 10 mg/kg would have on CBF. These CBF data support that THC markedly influences brain perfusion, while CBD alone does not impact brain perfusion, and it does not appear to modify THC’s perfusion effects when combined.
The clinical literature generally aligns with findings on cannabinoid effects on cerebral perfusion. THC is notably associated with increased CBF, especially in the frontal and cortical regions (Batalla et al., 2014; O’Leary et al., 2002; Ogunbiyi et al., 2020; Richter et al., 2018). Similarly, animal studies have observed increases in cerebral blood volume, a metric related to CBF, following the administration of CB1R agonists (Chin et al., 2008; Yao et al., 2009). However, no quantitative ASL studies have measured CBF after cannabinoid administration in rodents to date. Interestingly, several studies using an older technique, in vivo 14C-iodoantipyrine autoradiography, reported decreases in CBF following THC (Bloom et al., 1997; Goldman et al., 1975), or anandamide, a CB1/2R agonist (Stein et al., 1998). The discrepancies between MR-based findings and these early autoradiography results remain unclear, likely due to methodological differences.
The effects of CBD on CBF in humans appear smaller and more varied, likely due to its proposed polypharmacological properties. Some studies report CBD-induced increases or decreases in CBF within hippocampal regions, while others show a normalisation of elevated hippocampal CBF in patients at risk of psychosis (Bloomfield et al., 2020; Crippa et al., 2004; Davies et al., 2023), underscoring the subtler impact of CBD compared to THC. CBD’s effects are often most pronounced when a pre-existing deficit is present, as seen in patient populations or animal disease models (Batalla et al., 2020; Davies et al., 2023; Gunasekera et al., 2022). This suggests that CBD’s effects may be challenging to detect in healthy individuals, whether human or animal.
When considering the effect of cannabinoids on brain perfusion, it is important to note that THC can directly affect the vasculature through vasodilation via activation of CB1 receptors on endothelium (Ho and Kelly, 2017; Morse et al., 2023; Richter et al., 2018). CBD does not cause vasodilation or increased brain perfusion (Crippa et al., 2004). Therefore, it is possible that at least some of the effects of THC and THC:CBD on cerebral perfusion may be driven by a direct effect of THC on the vasculature. Disentangling the relative contributions of cannabinoid-induced vascular versus neural-driven CBF changes is challenging (Batalla et al., 2014). While THC and THC:CBD similarly impact CBF, their global and regional effects on fMRI patterns differ. CBD administration decreases regional brain connectivity without significantly affecting CBF. Thus, it is unlikely that the observed global connectivity changes are primarily dictated by CBF impacts, though interactions between drug-induced CBF changes and brain connectivity cannot be entirely ruled out.
To further characterise brain activity patterns and endophenotypes associated with each compound, we leveraged the within-subject design to probe FC and CBF data acquired from the same subject within the same session. Although FC and CBF are known to be associated and moderately correlated to each other, they reflect distinct physiological processes (Attwell and Iadecola, 2002; Liang et al., 2013a). CBF primarily reflects local neural activity, with highly active ‘hub’ regions – such as those in the DMN or auditory structures like the inferior colliculus – displaying greater perfusion. In contrast, FC captures temporal synchrony between regions, which may not correspond to the magnitude of activity within individual regions; a region can be highly active but weakly connected to others, or vice versa. Furthermore, physiological changes induced by factors such as illness, ageing or drugs can modulate the coupling between FC and CBF (Galiano et al., 2020; Qiu et al., 2017; Zhu et al., 2017).
We first confirmed a linear relationship between nodal FC strength and mean CBF in rats, consistent with findings from human studies (Liang et al., 2013a; Supplemental Figure 3). While a slight reduction in correlation coefficient was qualitatively observed in the THC group, differences in the relationship were not formally tested across groups due to a predicted lack of power to detect small differences. Overall, the correlation between FC and CBF across regions appeared preserved in all groups despite anaesthesia.
Instead, we utilised the within-subject design to perform a multivariate PLS analysis (McIntosh and Misic, 2013), incorporating FC and CBF data from all brain regions and subjects (Figure 8) to explore how these measures jointly vary with drug treatment. This revealed two LV1 and LV2 with distinct feature loadings: LV1 was primarily driven by CBF, and LV2 by FC. These latent patterns more effectively distinguished the treatment groups. THC showed strong positive LV1 scores, followed by THC:CBD, whereas LV2 best distinguished CBD from both THC and vehicle. Each drug group displayed a unique LV1/LV2 profile, suggesting a unique neural signature that could inform the classification of novel compounds. Notably, the strong contribution of CBF to THC’s LV1 profile is consistent with its robust regional perfusion effects (Figure 7), while CBD’s elevated LV2 scores reflect its overall decrease of FC (Figures 3 and 6).
Considering the tight coupling between cerebral metabolism and blood flow, it is also relevant to evaluate studies that have measured glucose consumption in the brain following cannabinoid administration. An animal study using 18F-fluorodeoxyglucose Positron Emission Tomography (FDG PET) with a cannabinoid agonist indicated increased metabolism (Nguyen et al., 2012), which aligns with our findings of elevated CBF. However, several earlier studies using 14C-2-deoxyglucose (2DG) autoradiography in rats showed metabolic decreases at the doses of THC we used (Freedland et al., 2002; Margulies and Hammer, 1991; Pontieri et al., 1999; Whitlow et al., 2002). These findings not only contradict our results of increased CBF but also contradict similar findings of increased brain perfusion in humans. Reasons for these discrepancies are unclear and may include factors such as anaesthesia, animal restraint (used in 2DG method) or analytical methodologies.
We also quantified CBD, THC and their metabolites in the animals in our study, and the specific concentrations are summarised in Table S1. Comparing doses of cannabinoids between animals and humans is complex due to logistical and metabolic reasons: human administrations of cannabinoids are typically via inhalation, although sometimes oral, and less frequently injection, whereas animals are usually injected. For instance, in human applications, typical plasma exposures approximately 20 minutes after vapour-based administration are around 15–30 ng/mL of THC, or 40 ng/mL of CBD (Lawn et al., 2023). In our study, median plasma concentrations in rats, measured approximately 4 h after administration of THC:CBD, ranged from 40 to 175 ng/mL for THC (depending on whether it was administered alone or combined with CBD), and approximately 100–1530 ng/mL for CBD depending on dose (Table S1). Considering the difference in timing (4 h in rats vs 20 min in humans) and an estimated conversion factor of 4–6 between human and rodent doses due to differences in metabolic rate (Nair and Jacob, 2016), these plasma concentrations appear to be comparable between both species although this would have to be verified by the appropriate drug metabolism and pharmacokinetic analysis that was outside the scope of this study.
In line with previous clinical (Lawn et al., 2023) and preclinical (Hlozek et al., 2017) studies, our study found that co-administration of CBD with THC led to a significant (approximately 4×) increase in THC concentration in both brain and plasma, compared to similar doses of THC alone. This effect might be attributed to the known impact of CBD on THC’s P450 enzymatic degradation in the liver (Smith and Gruber, 2022). Contrary to Hlozek et al. (2017), in our study, levels of CBD in the brain and plasma did not appear to be influenced by THC co-administration, when corrected for dosage difference (150 vs 10 mg/kg; Table S1). Additionally, CBD may influence THC metabolism in other ways, such as modulation of the ATP-binding cassette (ABC) transporter p-glycoprotein, which mediates THC efflux across the blood-brain barrier (Stella, 2023). If THC efflux is reduced, its effect could be extended and/or mitigated; or via CB1R – TRPV1 cross-talk (Chen et al., 2016), which could create regional interaction patterns between THC and CBD according to the distribution of both receptors (Britch et al., 2017). This interaction is complex and its unravelling is beyond the scope of this study. Nevertheless, our findings reinforce the hypothesis that the effect of co-administering CBD and THC is more substantial than the sum of their individual impacts, underscoring the need for further exploration into THC:CBD’s effects on the brain. Additionally, we observed a small amount of THC in the brains (but not plasma) of rats administered CBD alone, as also found by (Hlozek et al., 2017), which is highly likely due to the extremely low concentrations of THC being present in the CBD botanical formulation.
Our findings have implications for stratifying therapeutic indications in neurological and neuropsychiatric disorders, even though we used healthy animals rather than a disease model. Cannabinoids, especially CBD, have demonstrated clinical efficiency and received marketing authorisation for the treatment of rare epilepsy disorders including Lennox–Gastaut syndrome, Dravet syndrome, and Tuberous Sclerosis Complex (Devinsky et al., 2017, 2018; Miller et al., 2020; Thiele et al., 2018, 2021). Preclinical studies suggest increased brain connectivity in the models of epilepsy (Gill et al., 2017; Otte et al., 2012), in particular higher clustering and GE (Carboni et al., 2020; Richter et al., 2018), though weaker network connections have also been reported (Christiaen et al., 2019). Our finding that acute CBD reduces these metrics aligns with CBD’s use in epilepsy (Klein et al., 2017).
In clinical applications, THC is suggested to be effective in managing pain, particularly chronic pain (Stella, 2023; van de Donk et al., 2019). Key brain areas involved in modulating such pain include the prefrontal and cingulate cortex (Stella, 2023). In our study, THC significantly modulated FC in the cingulate cortex and, to some extent, in the orbital cortex, a part of the prefrontal cortex. Additionally, individuals with chronic pain exhibit altered activity within the SN and DMN (van Ettinger-Veenstra et al., 2019), both of which were also affected by THC in our study. However, THC also induces acute psychotic symptoms, which might exacerbate symptoms of schizophrenia, and potentially lead to addictions (Stella, 2023), all of which have been linked to aberrant processing within the SN and DMN (Bolton et al., 2020). The cingulate cortex and prefrontal cortex mediate many reward and emotional circuits in the brain – highly relevant to the experience and processing of both pain and psychosis. Therefore, the acute effect of THC we report may reveal alterations in the neural substrates of aberrant salience processing, reward mechanisms and pain pathways.
There are general methodological considerations that may have influenced the results of this study. Specifically, anaesthesia presents a confound for all rodent pharmacological fMRI studies unless conducted in awake animals. The majority of rodent fMRI studies use anaesthesia, which inevitably affects FC – it is used to reduce stress and immobilise the animals while scanning, since motion degrades image quality (Liang et al., 2012; Mandino et al., 2024). To minimise the impact of anaesthetics, we used a protocol that is associated with robust and stable BOLD responses as well as synchronised activity networks (Grandjean et al., 2014, 2020, 2023); You et al., 2021). Moreover, we included an appropriate vehicle group in the study design; therefore, it is less likely that the effects we see on FC are driven by the anaesthesia protocol used. In fact, our results are in line with another preclinical study investigating the effects of THC on BOLD response in awake rats (Madularu et al., 2017).
However, a particular concern regarding anaesthesia is the reported interaction between cannabinoids, particularly via CB1Rs, and noradrenaline signalling (Cathel et al., 2014). This interaction may involve α2-adrenergic receptors, which are also the primary site of action of medetomidine (Sinclair, 2003), the anaesthetic used in this study. We cannot dismiss a potential interaction between medetomidine and the cannabinoids administered, especially in the THC alone group (Tapley and Kellett, 2019). To conclusively address this interaction, future studies could compare different anaesthetic agents or perform the experiment in conscious animals. However, the former is complicated by the broad pharmacological profile of cannabinoids, which can interact with various systems implicated in anaesthesia – such as opioid, glutamatergic, or GABAergic pathways – potentially confounding results with alternative anaesthetics. The latter is made challenging by the well-documented influence of cannabinoids and endocannabinoids on stress-related processes (Navarrete et al., 2020), which are particularly relevant in conscious rodent fMRI, where stress effects are difficult to eliminate despite careful experimental control (Mandino et al., 2024).
Another limitation is that we only used male animals to minimise physiological variability and to keep the groups manageable as the inclusion of both sexes would have required larger groups. As thus far most cannabinoid research has been conducted in male animals, the inclusion of both sexes in future studies is recommended.
In summary, we have demonstrated that acute THC administration resulted in increases in FC and regional CBF, acute CBD administration resulted in an overall reduction in FC with negligible effect on CBF, and the combination drug THC:CBD resulted in effects similar to, but lower than THC alone. Our application of functional neuroimaging has thus identified differential pharmacodynamic signatures for THC and CBD in anaesthetised adult male rats. Further work should encompass an investigation of the effects of sub-chronic administration of phytocannabinoids on brain activity in animal models with relevance to selected disease indications to investigate changes on FC in a perturbed system, more applicable to the disease state. Overall, this neuroimaging work supports the role of preclinical fMRI to generate functional neuroimaging signatures as endophenotypes that could help inform indication selection as well as potential site(s) of action in drug discovery and development (Carmichael et al., 2018).