In our study, CIA was reliably induced in rat model, as evidenced by clinical arthritic scores and paw thickness. The disease activity was also indirectly reflected by weight loss in the animals (Alabarse et al., 2018). The oral dose of CBG uses 30 mg per rat to ensure sufficient systemic exposure despite low bioavailability, staying within safe and effective ranges reported in literature (Sztolsztener et al., 2024; Weerts et al., 2024; Borrelli et al., 2013). The treatment effect of CBG was demonstrated in the clinical score at the end of the experiment, which was significantly lower than the placebo score. There was a trend towards delayed onset of clinical score increase, i.e., appearance of the disease symptoms in the CBG group. While the extent of CBG efficacy did not match that of GC according to the clinical score and paw volume, the CBG group experienced the least weight loss of all the groups in which CIA was induced. This may reflect the effect of CBG and its potential in stimulating appetite (Brierley et al., 2016).

The thickness of the paws in the CBG group is very similar to that in the PCB group, which suggests that edema and joint hyperplasia were not affected by CBG. Although paw thickness was similar, differences in arthritis scores arise because the scoring system evaluates all four limbs and individual joints, assigning points for localized swelling or redness that may not substantially alter overall paw thickness. These results are consistent with previous findings that CBG does not prevent acute inflammation (paw edema) or reduce pain hypersensitivity (hyperalgesia and allodynia) in carrageenan-induced hyperalgesia and mechanical pain sensitivity model in rats (Weerts et al., 2024).

Conversely, GC, which have well-demonstrated efficacy in RA, have previously been found to reduce bone destruction in the murine CIA model on mice based on X-ray analysis, which corresponds to our observations (Toonen et al., 2014).

In our study, CBG treatment significantly suppressed the expression of TLR-4, TLR-5, and TLR-7 in the synovial tissue, as well as TLR-2 in blood on day 29. In contrast, GC treatment had no effect on these receptors but was found to markedly decrease the expression of TLR-9 in synovium. TLRs signaling leads to activation of NF-kB, AP-1, MAPK pathway and interferon regulatory factors (IRFs) to induce pro-inflammatory cytokines as TNF, IL-1β, IL-6 and INF and it has been shown that altered TLR expression is closely associated with RA, as summarized recently (Botos and Davies, 2011; Kawasaki and Kawai, 2014; Cui Sun et al., 2024; Unterberger et al., 2021). Therefore, the observation of TLR expression suppression by CBG supports the mechanistic plausibility of the treatment effects of CBG as reflected in the clinical scoring.

Our findings support the existing hypotheses that cannabinoids act as key regulators of TLR signaling, consequently decreasing the expression of pro-inflammatory cytokines and the activity of transcription factor NF-κB (Cui Sun et al., 2024; McCoy, 2016). However, there has been no direct experimental evidence specifically evaluating the impact of CBG on TLR expression until now. In a study by Delgado-Arevalo et al., the authors reported that CBG significantly reduced TLR-2-mediated IL-1β release in human keratinocytes stimulated with Cutibacterium acnes, however, they did not directly measure TLR-2 expression at the gene or protein levels (Perez et al., 2022). Their proposed mechanism was inferred from published literature. In another study, the authors observed that CBG (alone or in combination with CBD) reduced TNFα expression (Mammana et al., 2019). The combination of CBG and CBD also prevented IkBα phosphorylation and nuclear translocation of NF-kB in a motoneuron cell line that had been exposed to medium from lipopolysaccharide (LPS)-stimulated macrophages. This effect could be associated with modulation of TLR-4, a key pathway activated by LPS. However, this conclusion was also based on prior literature and was not supported by direct measurement of TLR-4 expression or activity (Cui Sun et al., 2024; Mammana et al., 2019). Therefore, our findings provide novel and direct evidence of CBG-mediated modulation of TLR expression in an in vivo model of inflammation.

Regarding the effect of GC on TLRs, our findings are consistent with the established hypothesis that GCs primarily suppress inflammation by inhibiting the downstream TLR signaling pathways, rather than directly reducing TLR gene expression (Cain and Cidlowski, 2017).

The TLR-dependent activation of the NF-κB and MAPK pathways occurs via MyD88 (Myeloid Differentiation Primary Response 88) or TRIF (TIR-domain-containing adapter-inducing interferon-β) proteins (Cui Sun et al., 2024). Within the NF-κB pathway, we assessed the relative expression of the NF-κB p65 subunit, which forms the most abundant form of NF-κB heterodimer (p50 + p65) activated by the canonical pathway. Our findings suggest that CBG exerts a time-dependent systemic inhibitory effect on NF-κB p65 expression, which is evident in the early phase (blood, day 16) and may contribute to the observed reduction of expression of systemic cytokines such as IL-1β and TNFα (Figures 3, 4). These results are consistent with findings from a previous study investigating CBG in a mouse model of atopic dermatitis (Jeong et al., 2025). In that study, CBG led to downregulation of IκBα, NF-κB, and p-NF-κB protein levels, along with a reduction in IL-1β mRNA, supporting its anti-inflammatory potential through modulation of the NF-κB pathway.

Surprisingly, we observed a lack of significant modulation of NF-κB p65 expression in the synovium by CBG, which may be a result of limited tissue penetration, tissue-specific regulatory mechanism or represent time dependency of this effect. Interestingly, GC treatment led to increased NF-κB p65 expression in the synovium, despite its known role as a transcriptional inhibitor of NF-κB. This paradox could be due to compensatory feedback mechanisms, where the cell upregulates NF-κB components in response to inhibited activity.

Further, our focus was on inflammasomes, which are key components of innate immunity. The activation of these cytoplasmic multiprotein complexes has been shown to contribute to the pathogenesis of RA (Jiang et al., 2021). Although the anti-inflammatory properties of cannabinoids are the subject of an increasing amount of research, no published data currently exist that directly examine the effects of cannabinoids, particularly CBG, on inflammasome signaling in the context of RA. Existing literature includes data few other cannabinoids, but only in models of other inflammatory disease (Suryavanshi et al., 2021).

In our study, CBG significantly reduced the expression of NLRP1A, caspase-1, and gasdermin D in synovial tissue, suggesting a targeted inhibitory effect on canonical inflammasome activation and pyroptosis (Figure 5). In addition, CBG reduced the expression of NLRP3 and AIM2 in blood, which may be mediated by the modulation of TLRs, NF-KB and CB2 receptor. In a previous study, selective CB2 receptor agonist HU308 was found to inhibit the NLRP3 inflammasome in mouse peritoneal macrophages and an experimental colitis mouse model (Ke et al., 2016). Another study showed that CB2 receptor activation attenuated neuroinflammation through autophagy-mediated degradation of the NLRP3 inflammasome in an LPS-induced microglia inflammation model and a mouse model of spinal cord injury (Jiang et al., 2022a). The CB2 receptor antagonist AM630 produced opposite results.

The expression of inflammasome-dependent cytokine IL-1β was significantly decreased by both CBG and GC in blood samples collected in the early phase of the experiment (day 16), with a non-significant downward trend still observed at day 29 (Figure 4). In synovial tissue, IL-1β suppression was observed only with GC treatment. This observation could be caused by tissue-specific mechanisms or limited CBG distribution into the synovial tissue. The reduction of IL-1β and AIM2 expression by CBG in blood may be associated with its CB2 receptor agonism, potentially modulating the canonical way of NLRP3 inflammasome activation through the downregulation of TLR2, NF-κB and NLRP3 (Figures 2, 3, 5). The key role of AIM2 in RA have been described previously, and silencing of the mRNA AIM2 expression has been associated with diminished IL-1β production (Zhao et al., 2022; Chen et al., 2020). No data exists on the effect of CBG or other cannabinoids on AIM2 expression or activity, highlighting the originality of our results.

Cannabinoids have also been shown to modulate immune responses via the JAK/STAT pathway in T-lymphocytes through activation of the CB2 receptor (Xiong et al., 2022; Jiang et al., 2022b). Abnormal activation of the JAK/STAT pathway in RA has been associated with pathological processes of synovial inflammation and bone destruction (Qin et al., 2022; Simon et al., 2021; Hu et al., 2022). The Janus family of kinases (JAKs) comprises four non-receptor protein tyrosine kinases: JAK-1, JAK-2, JAK-3, and TyK-2 (tyrosine Kinase 2). These kinases act as dimers and signal transduction peptides upon phosphorylation, triggered by the binding of specific cytokines to their membrane-bound receptors (Babon et al., 2014a; Kameda, 2023). Once phosphorylated, JAK dimers recruit proteins from the signal transducer and activator of transcription (STAT) family, including STAT1–4, STAT-5A, STAT-5B, and STAT-6. After dimerization, STAT proteins translocate to the nucleus, where they regulate transcription of target gene, often by recruiting coactivators (Ivashkiv and Hu, 2004). This STAT-mediated transcription enhances cytokine production and other immune components, creating a positive feedback loop that perpetuates inflammation (Ivashkiv and Hu, 2004). Key cytokines regulated by the JAK/STAT pathway include INF-γ (via JAK-1, JAK-2, STAT-1, 3, 5), IL-6 (via gp130 receptor–JAK-1, JAK-2, TYK-2, STAT-1, 3, 5), IL-12 and IL-23 (via JAK-2, TYK-2 and STAT-3,4), GM-CSF (via JAK-2, STAT-5), and Receptor Activator of Nuclear Factor Kappa-Β Ligand (RANKL) (via IL6-gp 130 receptor–JAK-2/STAT-3 → IL-6, RANKL) (Hu et al., 2022; Kameda, 2023).

Although modulation of the JAK/STAT pathway has predominantly been studied in the context of cancer, one study has demonstrated that CBG modulates this pathway in a mouse model of atopic dermatitis (Jeong et al., 2025). Topical administration of CBG (0.1 mg/kg or 1 mg/kg) significantly reduced the protein levels of JAK-1, JAK-2, TYK-2, STAT-1, STAT-2, STA-3, p-STAT-3, STAT-6, p-STAT-6 (p < 0.05) and markedly decreased mRNA expression of pro-inflammatory cytokines including Tslp, IL-1β, IL-4, IL-6, IL-13, IL-17, IL-18, IL-22, and IL-33 (p < 0.001). Our results support the immunomodulatory and anti-inflammatory potential of CBG, although the extent to which these markers were modulated was less pronounced in our CIA model. This discrepancy could be attributed to differences in the molecular mechanisms of inflammation in different animal models, as well as variations in CBG dosage and posology. In our study, CBG after oral administration had a more pronounced effect on the systemic circulation (Figure 8), although mRNA levels of STAT-3 and STAT-5a were also affected in the synovial tissue. We have seen no marked changes in total STAT3-or p-STAT-3 protein levels following CBG treatment in the synovium in our WB analysis (Figure 9), which may reflect the protein stability and their longer half-life (reduced mRNA levels may not immediately translate to decreased protein abundance). Additionally, post-transcriptional regulatory mechanisms such as mRNA stability, translation efficiency, and microRNA-mediated repression can affect protein synthesis independently of mRNA expression, contributing to the lack of corresponding changes in STAT-3 protein levels (Vogel and Marcotte, 2012).

Our findings also suggest that GC exerts a broader suppressive effect on JAK-STAT signaling, both locally in synovial tissue and systemically (Figures 7, 8). This is consistent with established mechanisms of GC action. GC influence the JAK/STAT pathway in part through GR-mediated interference with downstream components of TLR signaling pathways, including inhibition of NF-κB, AP-1, and induction of MKP-1, as well as direct physical interaction of GR with STAT-3, STAT-5a, STAT-5b, and STAT-6 (Bhattacharyya et al., 2011).

GC also modulates JAK/STAT pathway by regulation of suppressors of cytokine synthesis (SOSCs), which are known to negatively affect this pathway (Chinenov and Rogatsky, 2007). It has been reported that GCs induced SOCS-1 expression with subsequent interfering with TLR-3 and TLR-4-mediated STAT-1 activation (Bhattacharyya et al., 2011). SOCS3, on the other hand, plays a critical role in regulating immune responses via its interaction with gp130, which is essential for the effective inhibition of IL-6/gp130-dependent STAT-3 activation–a major reason why we included it in our analysis (Babon et al., 2014b). In our study, GC treatment significantly reduced SOCS-3 mRNA expression, whereas CBG did not significantly alter SOCS-3 expression levels compared to the PCB group. This suggests that the modulation of STAT-3 is not mediated via the SOCS-3 regulatory pathway. Results of studies investigating the effects of cannabinoids on SOCS-3 expression vary across different cell types and conditions (for example, the selective CB2 receptor agonist, HU-308, facilitated SOCS-3 protein expression in a model of systemic sclerosis, whereas CBD reduced SOCS-3 protein expression in an intestinal epithelium model) (Tian et al., 2024; Koay et al., 2014).

IL-6 is a key pro-inflammatory cytokine closely associated with the activation of the JAK/STAT signaling pathway, particularly through STAT-3 and STAT-1 (Hu et al., 2021). Importantly, IL-6 also serves as a crucial factor that connects NF-κB and STAT-3 signaling pathways. While IL-6 and its receptor complex can effectively activate STAT-3, IL-6 itself is a known transcriptional target of NF-κB. Furthermore, STAT-3 plays a vital role in the activation of the NF-κB pathway, forming a positive feedback loop that amplifies inflammation (Hu et al., 2021). In our study both treatments significantly reduced STAT-3 mRNA expression, a statistically significant reduction of IL-6 mRNA expression was detected only in the GC-treated group in synovial tissue (Figure 4). Neither CBG nor GC showed a significant effect on IL-6 mRNA expression in blood samples (Figure 4), and no changes in IL-6 protein levels were observed by WB (Figure 9). We assume that this has been caused by high severity of the inflammation induced in our CIA model–evident from high clinical score and fact that also GC treatment induced only partial efficacy. Few studies support the conclusion that the effect of CBG leading to a measurable decrease of IL-6 gene or protein expression is achieved only in inflammation of mild to moderate intensity (Jeong et al., 2025).

In our in vivo model of CIA, CBG reduced TNF expression in the blood, achieving significance in half of the experiment, however no effect was seen in the synovial tissue. This fact is confirmed by the results of immunoblotting. Although previous research suggested that COX-2 could be another target for CBG another study in murine models of colitis as well as CIA did not find any influence of CBG on COX-2 expression (Sztolsztener et al., 2024; Borrelli et al., 2013). Our study supports the later findings.

Our findings suggest that CBG may also specifically target pyroptosis-related pathways, as evidenced by the significant reduction of gasdermin D mRNA levels in synovial tissue. These findings support limited evidence available for CBD, e.g., in a recent study that CBD induced downregulation of gasdermin D expression in a mouse model of oral ulcers (Qi et al., 2022).

In RA, synovial fibroblasts and immune cells (B-cells) exhibit elevated BCL-2 expression, contributing to apoptosis resistance and the persistent inflammation (Lee et al., 2013; Yang et al., 2016). Interestingly, our results indicate that neither CBG nor GC treatment reduces this anti-apoptotic marker; in fact, they led to increased BCL-2 mRNA expression. Although GC are known to induce apoptosis in cancer cells, RA-derived T cells are resistant to GC-induced apoptosis, associated with upregulation of BCL-2 (Makrygiannakis et al., 2008). Consistently, we also observed reduced expression of the pro-apoptotic BAX in the GC group, whereas no significant change in BAX mRNA expression was found in the CBG group compared to the PCB. Similarly, in an in vitro model of neuroinflammation, CBG was shown to inhibit apoptosis by preventing the increase in BAX level and reducing cleaved caspase-3 and increasing of BCL-2 expression (Gugliandolo et al., 2018). In the context of RA, this anti-apoptotic effect may be potentially detrimental, as resistance to apoptosis contributes to synovial hyperplasia and sustained inflammation (Baier et al., 2003). These findings suggest that while CBG may promote cell survival under inflammatory stress, its pro-survival properties could inadvertently exacerbate pathogenic synovial cell accumulation and chronic inflammation, highlighting a need for caution in therapeutic applications for RA.

The MAPK/ERK pathway is a signaling cascade that is essential for cell growth, differentiation, and survival (Thalhamer et al., 2008). It initiates a kinase cascade that includes Raf (MAPKKK), MEK1/2 (MAPKK), and ERK1/2 (MAPK) (Zhang and Liu, 2002). This pathway plays a central role in CIA model as well as RA in driving synovial inflammation and joint destruction (Thiel et al., 2007). Both CBG and GCs modulate the MEK/ERK signaling pathway, a driver of inflammation in RA, but they do so via distinct mechanisms. As a partial CB₂ receptor agonist, CBG is believed to indirectly suppress ERK activation via CB₂-mediated anti-inflammatory pathways, although direct data on ERK activation after CBG treatment in the CIA model are currently lacking. We acknowledge that this aspect could be more clearly addressed by including a CB₂ receptor antagonist or conducting complementary in vitro experiments to confirm receptor involvement. (Navarro et al., 2018). We observed that the level of p-ERK/ERK was slightly influenced in the CBG group, though to a lesser extent than in the GC group. GCs exert a dual mechanism to inhibit ERK1/2 activation: they induce the expression of MAP kinase phosphatase-1 (MKP-1), which dephosphorylates ERK1/2, and simultaneously prevent MKP-1 degradation in immune cells like mast cells. However, in certain contexts, GCs fail to inhibit ERK1/2 due to continued proteasomal degradation of MKP-1 despite its upregulation (Kassel et al., 2001). Thus, although both CBG and GC may attenuate ERK signaling, CBG may provide a more durable, albeit less potent anti-inflammatory effect through stable CB2 engagement. The efficacy of GCs can fluctuate depending on cell-specific regulation of MKP-1 stability. This makes CBG a possible candidate for conditions where long-term modulation of ERK signaling could be beneficial.

RA associated joint destruction is accompanied by overexpression of several including matrix metalloproteinases (MMPs), particularly MMP-3 and MMP-9, which are considered as valid biomarkers of disease progression and joint destruction cytokines IL-23, IL-17, and chemokine RANKL (Bian et al., 2023; Ju et al., 2008; Adamopoulos et al., 2011). In our study, MMP-3 levels measured by qPCR, WB, and ELISA were consistent across methods. In the GC treatment group, serum levels of MMP-3, were significantly lower compared to the PCB group. Similarly, the relative mRNA expression of MMP-3, MMP-9, IL-17, IL-23 and RANKL in the synovial membrane was significantly reduced in the GC group. Although protein levels of MMP-3 showed only a decreasing trend, the overall data suggest that GC treatment effectively suppresses MMP expression, potentially limiting bone erosion and synovial inflammation. In contrast, there were no significant changes in MMP-3 levels in the CBG group, with only a trend towards decreased IL-17 expression in synovium and blood compared to the PCB group. Although a reduction in arthritis score might be expected to coincide with decreased MMP levels, our findings suggest that CBG’s beneficial effects may instead occur via mechanisms independent of MMP modulation. Also, the CBG effect was insufficient to diminish the enhancement of erosion and joint destruction related to the disease with highly induced inflammation and severity.

A limitation of the present study is that we investigated only the downstream inflammatory signaling pathways characteristic of the CIA model, without addressing the specific receptor-mediated mechanisms of CBG action. Moreover, all experiments were conducted in vivo using a rat model, and no complementary in vitro cellular assays were performed. Although the CIA model represents a well-established and compact system for studying rheumatoid arthritis–related inflammation, additional studies, including receptor-specific analyses and mechanistic in vitro and in vivo experiments, are required to fully elucidate the molecular targets and translational relevance of CBG.