The initial literature search across three databases identified 2,261 research reports (PubMed n = 89, ScienceDirect n = 1172, and Google Scholar n = 1000). Three additional articles were found after screening of reference lists. Of these 2264 publications, 2168 were excluded (clinical trials, reviews, conference abstracts, editorials, opinion articles, case reports, non-English articles, and duplicates). The remaining 96 publications were reviewed in full text, and only 38 studies met our inclusion criteria (Fig. 1). Table 1 summarizes the reports on therapeutic interventions, experimental models of ischemic stroke, tissue and behavioral outcomes observed, and the primary inflammatory mediators involved. In the following sections, the studies were categorized based on their mechanism of action into three broad categories: 1) CB2R selective agonists, 2) non-selective CB derivatives, and 3) indirect-acting compounds.

Β-caryophyllene (BCP) is a phytocannabinoid naturally occurring in cannabis sativa and other CB plants (Meeran et al. 2019). BCP is a nontoxic compound approved by the US FDA as a food additive. It poses promising beneficial effects in various pathologies, for instance, as an antibacterial, antifungal, anticancer, antioxidant, and anti-inflammatory agent (Goncalves et al. 2020). In experimental stroke research, BCP produced favorable outcomes in-vitro and in-vivo (Table 1) (Yang et al. 2017;  Yokubaitis et al. 2021). For instance, in a mouse model of permanent photothrombotic stroke, BCP intervention (1 h pre-injury and 24 h post-injury) significantly reduced infarct size and improved grip strength 3 days after stroke (Yokubaitis et al. 2021). Similarly, using the middle cerebral artery occlusion followed by reperfusion (MCAO/R) mouse model of stroke, treatment with BCP (at days 3, 2, 1 pre-injury, and 2 h post-injury) significantly decreased infarct size (48 h post-injury) and improved the neurological score (24 h post-injury) in a dose-dependent manner (Table1) (Yang et al. 2017). Moreover, another study by Choi IY et al. demonstrated that administration of BCP (3 h post-injury) decreased infarct size and edema volume 24 h after injury in the MCAO/R rat stroke model. (Choi et al. 2013) These studies show that BCP, even when administered late, improves stroke outcomes by reducing infarct size and neurological functional deficits and ameliorating edema, suggesting neuroprotective and anti-inflammatory effects of BCP.

The effect of BCP on inflammatory markers and necrotic pathways was also assessed in-vitro and in-vivo. Primary neurons with or without BCP treatment were exposed to oxygen–glucose deprivation and re-oxygenation (OGD/R). Pre-treatment with BCP 1 h before OGD significantly reduced cell death and the expression of mixed lineage kinase domain-like (MLKL) protein, suggesting protection against necrotic cell death (Yang et al. 2017). These effects were also confirmed in-vivo using mice exposed to ischemia–reperfusion (I/R) injury with or without BCP. It was demonstrated that BCP reduced infarct volume, neuronal necrosis, receptor-interaction protein kinase-1 (RIPK1), receptor-interaction protein kinase-3 (RIPK3), and MLKL phosphorylation after injury. BCP also decreased HMGB1, toll-like receptor 4 (TLR4), IL-1β, and TNF-α levels (Yang et al. 2017). It is well accepted that necroptosis and inflammation are interlinked and can cyclically amplify each other through interaction between many proteins such as interferon-gamma (IFN-γ), TNF-α, TLR4, DAMPs, Fas ligand (FASL), TNF-related apoptosis-inducing ligand (TRAIL) and RIPK1 (Yang et al. 2019; , Pasparakis and Vandenabeele 2015). Cumulatively, these studies suggest a potential role of BCP in improving stroke outcomes by reducing mechanisms attributed to neuroinflammation and necroptosis.

To explore the mechanisms by which BCP exerts its beneficial effects, in the same study by Choi et al., adding AM630 (a CB2R antagonist) masked BCP-induced beneficial effects on infarct and edema volume in OGD/R-exposed rats, suggesting the involvement of CB2R receptor in ameliorating these pathological changes (Choi et al. 2013). To further confirm this finding, a different study using small interfering RNA (siRNA) to target and inhibit the function of CB2R revealed diminished beneficial effects of BCP in-vitro (Guo et al. 2014). Furthermore, it is clear that the beneficial effects of BCP on stroke will persist even if administered after injury, opening the speculation that there are other mechanisms aside from their anti-apoptotic and anti-inflammatory actions. To further explore this assumption, it was shown that BCP administration increases the co-expression of phosphorylated cyclic adenosine monophosphate (cAMP), response element binding protein (CREB), and brain-derived neurotrophic factor (BDNF) in cortical brain areas. Notably, the latter two molecules are known to be involved in neuroprotection and neural repair mechanisms (Choi et al. 2013). Mounting evidence revealed negative correlations between BDNF and neuroinflammation, particularly in psychiatric patients (Porter and O Connor 2022). In addition, exogenous BDNF dampens microglial activation, reduces TNF-α, and induces interleukin-10 (IL-10) expression in a rodent stroke model (Porter and O Connor 2022). Similarly, CREB activation directly inhibits NF-κB, induces IL-10, and promotes neuroprotection and survival (Wen et al. 2010). In parallel with these studies, it was shown that the increased expression of CREB and BDNF proteins induced by BCP produces neuroprotective effects (Choi et al. 2013). This suggests that targeting CB2R using BCP can be helpful in the early hours following an injury where excitotoxicity is dominant and can also be beneficial later when neural repair mechanisms are needed (Alamri et al. 2021). These results were also confirmed in-vitro using neuronal/glial mixed culture exposed to OGD/R in the presence of BCP, in which an increase in BDNF and CREB expression was concomitant with an increase in phosphorylated adenosine monophosphate kinase (p-AMPK)/AMPK (Yang et al. 2019). Evidence suggests that AMPK activation in mice can be involved in both deteriorating stroke injury and preventing neuronal death (Zhao et al. 2017). Given that CREB is regulated by the AMPK signaling family and that phosphorylated-CREB induces BDNF expression, these findings suggest that CB2R activation might act through the AMPK-CREB-BDNF signaling pathway (Porter and O Connor 2022). Taken together, this suggests the potential usefulness of BCP before and after the stroke pathophysiological inflection point. Its ability to promote neuroprotection and support recovery highlights its dual benefits, which may depend on the timing of its administration.

Additionally, the impact of BCP on microglia was also investigated. BCP was shown to suppress the levels of ionized calcium-binding adapter molecule 1 (Iba1; a marker for microglial cells) 3 days following injury in mice (Yokubaitis et al. 2021). Similarly, BCP was found to maintain microglial cells viability by inhibiting phosphorylation of inhibitor of nuclear factor kappa B (IκBα) and the release of pro-inflammatory cytokines such as NF-κB, TNF-α, IL-6, and IL-1β, in-vitro (Guo et al. 2014). These studies indicate that microglial activation plays a role in stroke pathogenesis and that BCP treatment promotes the polarization of microglial cells toward the anti-inflammatory M2 phenotype.

In an attempt to understand the influence of BCP on astrocytes in experimental stroke, Serra et al. used the bilateral common carotid artery occlusion/reperfusion (BCCAO/R) rat stroke model. They found 1) a decrease in transient receptor potential vanilloid 1 (TRPV1), 2) an increase in tropomyosin receptor kinase B (TrkB) and astrocytes marker: glial fibrillary acidic protein (GFAP), and 3) surprisingly no significant change in BDNF and Iba1 levels 60 min after reperfusion. (Serra et al. 2022). These findings suggest that BCP modulates both astrocytes and TRPV1 receptors, but whether its effects on TRPV1 mediate changes in astrocytes or vice versa remains unclear. Interestingly, Yokubaitis et al. found that co-administration of BCP and cannabidiol (CBD), but not BCP alone, significantly reduced Iba1 total fluorescence, cell body size, and count, 3 days after subjecting mice to photothrombotic injury, indicating their desirable combined effects on reducing microglial function. While BCP alone did not reduce Iba1 total fluorescence and cell count, CBD increased Iba1 significantly 3 days post-injury (Yokubaitis et al. 2021). This indicates the synergistic effects of the BCP-CBD combination on reversing microglial activation compared to a single treatment of either BCP or CBD. Taken together, the evidence indicates that the neuroprotective mechanism of BCP involves suppressing the activation of brain inflammatory cells, specifically microglia and astrocytes.

The variability in BCP's effects on BDNF expression may stem from differences in treatment timing. BCP increased BDNF levels when administered 3 h post-stroke, likely due to neural repair mechanisms, but reduced BDNF when given 6 h before injury (Choi et al. 2013;  Serra et al. 2022). Since BDNF's effects are mediated through the TrkB receptor, and TrkB levels do not directly correlate with BDNF, this discrepancy may involve indirect modulation by inflammatory factors (Serra et al. 2022; Simao et al. 2014). Additionally, differences between in-vitro and in-vivo studies, such as genetic material synthesis and evaluation time (60 min only), likely contribute to these contrasting results (Liu et al. 2008). It is expected that in-vivo BDNF levels may require extended observation periods to align with in-vitro findings.

Both neuroinflammation and oxidative stress can negatively affect the BBB, resulting in increased permeability and, consequently, the infiltration of pro-inflammatory mediators (Serra et al. 2022). Since BCP can modulate both components, a study addressed this gap by simulating the neurovascular unit (NVU) composed of neurons, astrocytes, and brain microvascular endothelial cells to evaluate the effects of BCP treatment (24 h before OGD/R) (Tian et al. 2016). They found the BBB molecules remained intact 24 h following reperfusion only if BCP was used before OGD/R, indicating a neuroprotective mechanism through decreasing BBB permeability and neuronal apoptosis. This study also found that BCP reduces pro-inflammatory cytokines (TNF-α, IL-6, IL-1β, and MMP-9), oxidative stress (increase in malondialdehyde (MDA) and decrease in superoxide dismutase (SOD)), and neuronal apoptosis (decrease in B-cell lymphoma 2 (Bcl-2)) in the NVU OGD/R model. In addition, BCP was found to induce neuronal growth proteins such as growth-associated protein-43 (GAP-43) and nerve growth factor beta (NGF-β), indicating the involvement in neuroprotection and neural repair and confirming its multidirectional impact (Tian et al. 2016).

Overall, BCP was frequently documented to ameliorate neuroinflammation through targeting NF-κB, TNF-α, IL-10, IL-1B, and IL-6 and modulating the functionality of microglia and astrocytes. Other BCP actions include reducing oxidative stress and necroptosis and promoting survival and growth pathways. These pathways are going in parallel and are known to improve stroke outcomes.

JWH-133, a selective CB2R agonist, was first developed by Dr. John Huffman in 1999 (Huffman et al. 1999), and has since been found to be effective experimentally in cancer, myocardial infarction, and various neurodegenerative diseases (Hashiesh et al. 2021). Accumulating evidence demonstrated JWH-133's efficacy in stroke preclinical research by improving histological and functional outcomes with concomitant reversal of disease-induced inflammation (Hashiesh et al. 2021; Zarruk et al. 2012; Pottier et al. 2017; Gupta et al. 2020; Hashiesh et al. 2021). Zarruk et al. found that JWH-133 administration, 10 min and 3 h following MCAO in mice, significantly reduced infarct size and improved neurological outcome 48 h following injury (Zarruk et al. 2012). These effects were abolished upon administration of SR144528 (a CB2R antagonist) or using CB2R-knockout mice (Zarruk et al. 2012). Indeed, this suggests that the neuroprotection of JWH-133 is mediated via CB2R activation.

In addition, the effects of JWH-133 on the expression of M1 and M2 markers following ischemia were also investigated. The findings revealed that JWH-133 significantly reduced the expression of M1 markers induced by MCAO, i.e., Iba1, IL-6, interleukin-12 (IL-12), inducible nitric oxide synthase (iNOS), and chemokine ligands (CCL-2, CCL-3, CCL-5), 15 h post-injury (Table 1) (Zarruk et al. 2012). In contrast to other reports in the literature, in which JWH-133 inhibited the M1 phenotype and promoted the M2 phenotype (Tanaka et al. 2020; Lin et al. 2017; Li et al. 2022), in this study, JWH-133 did not only inhibit the M1 phenotype but also the M2 phenotype, as indicated by a significant decrease in the expression of IL-10 and transforming growth factor β (TGF-β), 24 h post-injury (Table1) (Tanaka et al. 2020). In another study, JWH-133 combined with hypothermia given immediately post hypoxic-ischemic encephalopathy (HIE) in rats showed a significant decrease in TNF-α and chemokines (CCL3, CCL5) induced 24 h post-injury (Table1) (Gupta et al. 2020). Similarly, JWH-133 administered 1 h after injury demonstrated notable anti-inflammatory effects, including a reduction in CD11b, a microglial activation marker, when evaluated 7 days after MCAO/R. These findings highlight the potential of JWH-133 in mitigating neuroinflammation through early and sustained modulation of inflammatory responses.

To assess the effects of JWH-133 on BBB permeability, MCAO/R rats were injected with thrombin into the right basal ganglia to induce BBB damage, followed by JWH-133 administration 1 h later. JWH-133 significantly decreased BBB permeability, prevented tight junction-related protein (ZO-1, i.e., zonula occludens-1) attenuation, and reduced MMPs (MMP-9 and MMP-12) 24 h post-injury (Table 1) (Li et al. 2015). Additionally, JWH-133 significantly suppressed Iba1 and p44/42 MAPK phosphorylation (Table1) (Li et al. 2015). P44/42 Mitogen-activated protein kinase (MAPK) is a known signal transducer activated in the brain (particularly in microglia) in response to various stimuli resulting in neuroinflammatory exacerbation and further BBB damage (Koistinaho and Koistinaho 2002). Moreover, inhibiting p44/42 phosphorylation reduces neuroinflammation and apoptosis associated with cerebral ischemia (Liu and Gonzales 2017). Interestingly, the effects of JWH-133 on p44/42 MAPK were reversed when combined with SR144528 (a CB2R antagonist), suggesting CB2R as a target for the modulation of stroke-induced neuroinflammation and its associated molecular signaling pathways (Li et al. 2015). In a nutshell, JWH-133 improved stroke outcomes and modulated the consequent neuroinflammatory events such as microglia phenotypic class switching and BBB damage through CB2R activation. However, these studies demonstrate JWH-133's efficacy only when administered during a short period, 10 min or 3 h post-injury, which may not be the most clinically feasible window for acute ischemic stroke patients. Hence, future studies should assess the efficacy of JWH-133 at broader durations following stroke.

AM1241, a synthetic CB2R agonist, has been extensively studied for its effects on neuropathic pain (Wilkerson et al. 2012; Hsieh et al. 2011), and hepatic and myocardial fibrosis (Ali et al. 2021;  Li et al. 2016). However, only a single study has explored its impact on ischemic stroke (Yu et al. 2015). In this study, administration of AM1241 2 days following MCAO/R showed no significant improvement in behavioral function, infarct size, and inflammatory markers in the lesioned area (Yu et al. 2015). However, it was effective when administered 5 min before injury, as it ameliorated behavioral deficits and brain infarction in rats when assessed 2 days following injury (Table1). Additionally, AM1241 partially reduced Iba1 when assessed on day 6 post-injury (Yu et al. 2015). AM1214 has also demonstrated beneficial effects in-vitro in primary cortical neuronal culture. Pre-treatment with AM1214, 5 min before glutamate, significantly inhibited glutamate-induced neuronal damage, suggesting that CB2R agonists could ameliorate neuroinflammation by inhibiting excitotoxicity and cell necrosis (Yu et al. 2015). These findings suggest that early administration of AM1241 exhibits neuroprotective effects in experimental stroke models. However, this therapeutic window is not clinically feasible for stroke patients. Given the limited studies of AM1241 in stroke models, further research is warranted to clarify its potential.

COR-167, also called SER601, a highly selective CB2R agonist, was first synthesized in 2008 (Pasquini et al. 2008). Since then, it has been examined in various neurological diseases, including multiple sclerosis (Cioni et al. 2019), brain tumors (Cioni et al. 2019), and neuropathic pain (Borgonetti et al. 2023). In the context of cerebral ischemia, few studies have reported beneficial effects of COR-167. For instance, using OGD-exposed cortical slices extracted from Sprague–Dawley rats, it was shown that adding COR-167 to the reperfusion phase inhibited the OGD-induced release of lactate dehydrogenase (LDH) (Table1) (Contartese et al. 2012). The effect produced by COR-167 on LDH was reversed by administration of AM630 (a CB2R antagonist) but not AM251 (a CB1R antagonist), indicating COR167's effects are mediated through CB2R (Contartese et al. 2012). These results align with the findings of other selective CB2R agonists (JWH-133, AM124) in experimental stroke studies.

The effects of COR167 on the inflammatory cascade were also evaluated using rat brain cortical slices subjected to OGD/R, where it was observed to reduce cellular edema compared to vehicle controls (Contartese et al. 2012). Furthermore, COR167 exhibited significant anti-inflammatory properties by reducing OGD/R-induced IL-6 and TNF-α levels, supporting its role in mitigating neuroinflammation (Contartese et al. 2012). This is consistent with another study in which COR167 effectively suppressed the pro-inflammatory cytokines interleukin-4 (IL-4) and interleukin-5 (IL-5) in a cellular model of multiple sclerosis (Annunziata et al. 2017). Notably, COR167 was also found to significantly reduce MDA levels and glutamate release induced by OGD/R, while also restoring glutathione levels, indicating potential protective effects against oxidative stress (Contartese et al. 2012). While the effect of CB2R selective agonists in ameliorating neuroinflammation triggered by ischemia is well established, the selectivity of COR167 to CB2R is much higher than JWH133 (Pasquini et al. 2008). However, the difference in their effects remains unclear due to a lack of studies investigating COR167 effects in experimental stroke.

In addition to COR167, other available selective CB2R agonists include O-3853 and O-1966. Few previous studies have shown beneficial effects of these two agents in animal stroke models (Zhang et al. 2007; Zhang et al. 2009; Ronca et al. 2015). However, recent studies have yet to confirm and further explore the mechanism of action of these compounds.

CBD is a cannabis-derived CB with no intoxicating effects compared to tetrahydrocannabinol (THC), another plant-derived CB compound (Mlost et al. 2020). Due to its high safety profile relative to other CBs, CBD's therapeutic potential was widely reported in various pathological conditions, such as inflammatory diseases, multiple sclerosis, arthritis, schizophrenia, cancer, epilepsy, and stroke (Goncalves et al. 2020). In neonatal rats subjected to hypoxic-ischemic injury, administration of CBD 10 min post-injury significantly reduced lesion volume and improved motor and memory functions 30 days post-injury (Pazos et al. 2012). Similarly, in another study, administration of CBD 30 min following MCAO/R in rats significantly improved motor and somatosensory functions; however, no reduction in infarct size was observed 7 days post-injury (Ceprian et al. 2017). The exact mechanism by which CBD exerts its beneficial effects is still unclear; however, it could be attributed to the upregulation of BDNF, which is heavily linked to stroke recovery and plays a crucial role in neuronal plasticity and function (Mori et al. 2017; Bejot et al. 2011). Interestingly, in the BCCAO mouse models, CBD induced BDNF upregulation and promoted neuronal viability 21 days post-injury (Mori et al. 2017). This was associated with a decrease in neuronal degeneration, improvement in cognitive function, and a decrease in anxiety- and depression-like behavior (Mori et al. 2017).

In addition, CBD was shown to possess anti-inflammatory effects by decreasing microglial and astrocyte activation 7 and 30 days post-MCAO/R in rats (Ceprian et al. 2017). This observation was further supported in the BCCAO/R model, where CBD administration 30 min before or 3, 24, or 48 h after injury produced similar results (Ceprian et al. 2017; Mori et al. 2017). Another study explored the protective effects of CBD against hippocampal damage and cognitive decline caused by brain ischemia in adult male mice. The mice underwent 17 min of BCCAO and were tested in the Morris water maze seven days later. CBD was administered 30 min before and at 3, 24, and 48 h after BCCAO. Results showed that CBD improved spatial learning and reduced hippocampal damage and GFAP levels in ischemic mice (Schiavon et al. 2014). These findings suggest CBD