Driven by the pressing clinical need for more effective and durable treatments for kOA, this study provides compelling evidence for the involvement of the endocannabinoid system (ECS) in modulating synovial inflammation and nociception within the complex pathogenesis of kOA. While the current data provides fundamental descriptive observations of expression levels and correlations in a well-powered human cohort, it offers valuable foundational insights into patient potential responsiveness to targeted CB2R-specific intervention. This is particularly crucial given the limitations of current non-operative therapies, such as corticosteroids and NSAIDs, which offer only transient pain relief and fail to modify the underlying chronic inflammation and fibrosis that drive disease advancement. Notably, some NSAIDs may even carry risks of chondrotoxicity or accelerating disease progression [6]. The identification of specific CB2R-positive cells and deeper mechanistic insights is the subject of ongoing investigation.

The data reveal a significant inverse relationship between synovial CB2R expression, patient-reported pain (as measured by KOOS), and histological synovitis. This observation suggests that targeting CB2R could potentially limit inflammatory joint damage by modulating the activity of synovial fibroblasts and macrophages while concurrently altering nociceptive signaling [13,36]. It further implies that individuals experiencing lower pain levels may possess a more effectively regulated ECS, characterized by reduced chronic activation and, consequently, less CB2R desensitization and downregulation. This preserved receptor availability could explain the higher CB2R expression observed in the low-pain cohort.

A significant challenge in managing kOA is the frequent decoupling between subjective patient-reported symptoms and objective measures of disease severity, such as radiographic findings or range of motion [8,37]. This discrepancy, where imaging often reveals asymptomatic kOA, suggests that factors beyond gross structural damage, such as the neuroinflammatory state of the synovium, are critical [9]. A subset of patients develops hyperalgesic kOA, marked by spontaneous pain and heightened sensitivity to stimuli, often necessitating opioid use and significantly diminishing quality of life [38]. This hyperalgesia may stem from peripheral and central sensitization mechanisms, further complicating the pain phenotype. Pathological changes within the synovium, particularly synovitis-induced neuroinflammation and hyperalgesia, appear crucial in driving symptom severity and disease progression. The highly innervated synovium undergoes inflammatory and fibrotic remodeling, altering microvascularity and nociceptive signaling, likely contributing substantially to kOA arthralgia [9]. This points toward a neurogenic inflammatory basis for kOA pain, with TRPV1 upregulation identified as a key factor in synovial hyperalgesia [13].

The results demonstrate significantly increased TRPV1 expression in synovial tissues from patients reporting high pain levels. TRPV1 is known to be activated by a range of noxious stimuli, including heat, inflammatory cytokines (such as those found in kOA), and the acidic pH characteristic of inflamed tissues [13,36,39]. The co-localization of TRPV1 with the neuronal marker PGP9.5 suggests that TRPV1 is present on both nociceptive nerve fibers in the subintima and potentially on hyperplastic synoviocytes in the intima, consistent with its expression on non-neuronal synovial cells [20]. This finding suggests a strong association between local synovial pathology and neurogenic pain signaling, with TRPV1 upregulation appearing to be a key molecular event, involving the release of neuropeptides such as CGRP, which promote neurogenic inflammation and hyperexcitability [9]. Our data suggest that, in combination with local CGRP levels of the SF, the distribution of TRPV1 in synovial tissues can help indicate pain severity. Patients exhibiting pain hypersensitization often display hyperexcitable responses to sensory input, complicating pain management. Therefore, quantifying synovial TRPV1 distribution could help identify samples from patients with pain hypersensitization who might benefit most from therapies targeting this pathway.

The chronic pro-inflammatory milieu, rich in cytokines (e.g., IL-1β, transforming growth factor-beta (TGF-β), and TNF-α) released by activated macrophages and infiltrating lymphocytes within an acidic microenvironment, may drive the elevated TRPV1 expression observed in high-pain patients, which can alter receptor turnover and sensitivity [40]. This is supported by findings that link such inflammatory mediators to increased nerve fiber density and hyperalgesia [36]. Chronic TRPV1 activation by these inflammatory and potentially mechanical stimuli likely promotes local receptor upregulation and the development of hypersensitive pain responses. This chronically active TRPV1 pathway will also lead to increased 2-AG levels in the local environment, known to be synthesized in response to TRPV1 activation [24].

The observation that TRPV1 expression was elevated in the high-pain group without a concomitant significant increase in overall synovial nerve fiber density (PGP9.5) is particularly noteworthy. While inflammation can lead to a decrease in nerve fiber density [11], this suggests that the increased TRPV1 signal is not merely a reflection of greater innervation, but rather an increase in the TRPV1 channel on existing nerve fibers and/or non-neuronal synoviocytes. This points to a qualitative change in the nociceptive apparatus within the synovium of high-pain patients.

Conversely, the finding that synovial CB2R distribution is lower in patients presenting with highly severe synovitis and pain aligns with what results from chronic CB2R activation. CB2R agonism on human osteoarthritic fibroblast-like synoviocytes has been shown to suppress M1 macrophage polarization and subsequent inflammation [36]. However, persistent ECS activation in chronic inflammatory states, such as kOA, can lead to a desensitized (unresponsive) and downregulated CB2R [31], potentially diminishing its protective anti-inflammatory capacity. This phenomenon explains the observation of lower CB2R expression in the high-pain/high-synovitis group. It also provides a plausible explanation for the elevated 2-AG levels in the SF of high-pain patients. While 2-AG is typically undetectable in healthy SF, it is markedly elevated in patients with kOA [41]. As shown in Figure 3B, CB2R is typically coupled to an inhibitory G protein, Gαi/o, an "off switch" for inflammation​. When activated at normal physiological levels, this leads to the inhibition of adenylate cyclase, resulting in reduced cyclic adenosine monophosphate (cAMP) and the suppression of pro-inflammatory cytokine expression. This suppression of cytokine expression is achieved by inhibiting transcription factors, such as NFκB, a master regulator of pro-inflammatory cytokine production [42]. This is in contrast to what may take place under states of sustained inflammation, where overactivation of CB2R is observed and a possible paradoxical shift to coupling with a stimulatory G protein, Gαs​, can occur as an "on switch" of inflammation, which has primarily been demonstrated in leukocytes [43]. As shown in Figure 3C, if operative in synoviocytes, progression down this stimulatory pathway could lead to elevated cAMP levels, thereby activating the NFκB pathway and resulting in increased pro-inflammatory cytokine production and inflammation [44].

Furthermore, chronic inflammation, which causes the production of noxious stimuli, can result in the constant activation of TRPV1, driving increased 2-AG production in turn [24]. This increase in 2-AG in the local environment can overwhelm the ECS and CB2R, leading to CB2R downregulation and agonist accumulation, thereby diminishing its therapeutic effects (Figure 3C). Moreover, endocannabinoids like 2-AG can be metabolized via Cox-dependent pathways to pro-inflammatory prostaglandins, which can drive downstream mediators such as IL-6, TNF-α, and IL-1β linked to synovitis and hyperalgesia [45]. Elevated 2-AG in cerebrospinal fluid (CSF) and SF has also been associated with increased acute postoperative pain following TKA, further suggesting a pro-inflammatory role for accumulated 2-AG in settings of chronic inflammation and CB2R downregulation [46]. These findings strongly support the involvement of CB2R and the ECS in regulating TRPV1-mediated neuroinflammatory pain.

Preclinical studies demonstrate that endocannabinoids (like AEA, which also binds TRPV1 with affinity similar to capsaicin) and selective exogenous CB2R agonists (like JWH133) act as partial agonists at TRPV1 [47]. These interactions highlight TRPV1 as an "ionotropic cannabinoid receptor" [47]. They may induce a small inward calcium current, insufficient to trigger nociceptor firing but adequate to desensitize TRPV1 via calcium-dependent mechanisms such as calcineurin activation or by affecting TRPV1 trafficking [48]. The observation that CB2R-mediated analgesia is attenuated in TRPV1 knockout mice further substantiates that functional crosstalk between CB2R and TRPV1 contributes significantly to the analgesic mechanism via interplay between the ECS and EVS [48]. This implies that while targeting CB2R alone may offer baseline analgesia, therapeutic strategies leveraging the CB2R-TRPV1 crosstalk, potentially through CB2R-triggered endocannabinoid release that then modulates TRPV1 or direct receptor interactions, could achieve a greater analgesic effect.

The results advocate for investigating synthetic CB2R agonists, capable of engaging this crosstalk and desensitizing TRPV1, as potential disease-modifying analgesics for kOA. This aligns with the understanding of TRPV1's "dual role" as a "pain switch," capable of both sensitization and desensitization [49]. The concept of "functional selectivity", where a single therapeutic agent can be engineered to activate only a specific, desired signaling pathway, in CB2R signaling emphasizes that different agonists can produce distinct cellular responses by activating differing repertoires of signaling pathways [50]. Administering such agonists earlier in the disease course might help maintain ECS homeostasis, preventing the cycle of overactivation, desensitization, and downregulation observed in chronic inflammation. Preclinical evidence also supports the ability of peripherally administered cannabinoid agonists to attenuate hyperalgesia [51]. The synthetic CB2R agonist JWH133, for instance, reduces inflammatory mediators and synovitis in animal models [36]. JWH133 exhibits antioxidant, anti-fibrotic [52], and anti-inflammatory properties [36] without the hepatotoxicity reported for CBD [26]. This suggests that selective CB2R agonists could offer superior pain relief and potentially slow disease progression through multifaceted mechanisms, including inhibiting fibroblast production of IL-6/IL-8 and modulation of TGF-β1 [53]. Future research should also evaluate the ability of select allosteric agonists, alone or in combination with orthosteric activators, to restore adequate CB2R distribution or functionality, including the use of novel bitopic ligands, advanced therapeutic molecules designed to bind to a receptor in two different places at once for a more specific and potent effect [54]. If the CB2R can be favorably modulated using synthetic agonists, this may decrease the accumulation of endogenous agonists like 2-AG, which will limit ECS dysregulation and further inflammation.

This study has several limitations. First, our cross-sectional design, while providing valuable correlational evidence, cannot establish a direct causal relationship between the observed receptor expression levels and pain severity. The findings demonstrate a strong association between a dysregulated CB2R/TRPV1 axis and pain, but they do not prove that altering receptor expression will directly cause changes in pain perception. Second, the lack of a true non-OA control group limits our ability to make conclusions about baseline CB2R and TRPV1 expression levels in healthy synovium. However, our robust stratification of a clinically homogeneous population into high- and low-pain groups allowed us to identify critical molecular differences that exist even among patients with end-stage disease, which may be more relevant for predicting therapeutic responsiveness. Finally, the logistical challenges of intraoperative sample collection led to natural variability in sample sizes across different assays. Although we maximized our use of all available samples, as shown in Table 2, this variability suggests that some conclusions, especially regarding CGRP and 2-AG levels, are based on a smaller subset of the overall patient group and need to be confirmed with additional studies.

Future studies are warranted to validate these findings and to incorporate functional assays that can directly measure receptor activity.