Unless otherwise specified, chemicals and cell culture reagents were obtained from Merck Life Science S.r.l. (Milan, Italy).

Turmeric oleoresin (TUR) was purchased from Universal Oleoresins (Kerala, India). TUR contained 29.40% ± 4.20% (w/w) of total curcuminoids, of which 18.20% ± 2.60% (w/w) was curcumin (CUR), 6.01% ± 0.85% (w/w) was demethoxycurcumin (DMC), and 5.22% ± 0.74% (w/w) was bisdemethoxycurcumin (BDMC).

Stock solutions of β-Caryophyllene (BCP) and TUR were prepared in 100% ethanol at concentrations ensuring a final ethanol concentration of ≤0.5% (v/v) in the cell culture medium. Rimonabant, SR144528, 3-isobutyl-1-methylxanthine (IBMX), and forskolin (FSK) stock solutions were prepared in 100% dimethyl sulfoxide (DMSO) at a concentration that guarantees a final DMSO concentration ≤0.1% (v/v) in the cell culture medium. Vitamin C stock solution was prepared in water. All the solutions were stored at −20°C until use.

The human chondrosarcoma cell line (SW-1353 – Cat. HTB-94™) was obtained from ATCC^® (American Type Culture Collection - Manassas, Virginia). SW-1353 cells were maintained at 37°C in an atmosphere containing 5% CO₂ at 95% relative humidity. They were used between passages 20 and 30 to ensure consistent cell behavior. The basal medium was composed of Dulbecco’s Modified Eagle’s Medium high glucose, supplemented with 10% fetal bovine serum, 1% L-glutamine, 100 U/mL penicillin, 0.1 mg/mL streptomycin, and 1% non-essential amino acids. For all analyses, negative and positive control groups were incubated in a medium containing 0.5% (v/v) ethanol and/or 0.1% (v/v) DMSO, to exclude any possible solvent-mediated effect.

Cell viability was assessed using the PrestoBlue™ reagent (Thermo Fisher Scientific, Milan, Italy) according to the manufacturer’s guidelines. SW-1353 cells were seeded in 96-well plates at a density of 1 × 10⁴ cells/well and used 48 h post-seeding. They were then treated with TUR at increasing concentrations (n = 8) for 24 h, after which viability was assessed. Fluorescence values were recorded using a Varioskan™ LUX (Thermo Fisher Scientific, Milan, Italy). Cell viability was determined using the following formula: C e l l   V i a b i l i t y   % = Mean   Treated   Group   Fluorescence Mean   Negative   Control   Fluorescence   x   100

Reactive oxygen species (ROS) levels were assessed using CellROX^® Deep Red Reagent (Thermo Fisher Scientific, Milan, Italy) following the manufacturer’s guidelines. Briefly, CellROX^® Deep Red Reagent is a cell-permeable, non-fluorescent reagent in its reduced state that becomes fluorescent upon oxidation by ROS, with an emission maximum of around 665 nm.

Cells were seeded into 96-well plates at a density of 1 × 10⁴ cells/well and subjected to a challenge 48 h post-seeding. The cells were categorized into four distinct groups (n = 8): (1) negative control, (2) positive control, (3) TUR, and (4) TUR + SR144528. The group exposed to the antagonist (group 4) underwent a pretreatment with SR144528 at 1 μM for 10 min. Then, both the TUR and TUR + SR144528 groups received treatment with TUR at 15 ppm for 2 h. Subsequently, menadione at 100 μM was introduced as a ROS-inducing molecule for 1 h in the presence of the treatment. Following this, the medium was replaced with fresh media containing CellROX^® Deep Red Reagent at 5 μM, and the cells were incubated for 30 min at 37 C. The medium was then removed, and the cells were washed three times with Dulbecco’s phosphate-buffered saline (DPBS). Fluorescence values were recorded using a Varioskan™ LUX (Thermo Fisher Scientific, Milan, Italy). Vitamin C at 150 μM served as a standard antioxidant internal control. Supplementary Figure S1 reports the oxidative stress challenge optimization.

Chondrocytes were challenged as described by Pang et al., 2021, with some modifications. Co-treatment with IL-1β and TNF-α was chosen to mimic the inflammatory environment of OA, as these cytokines are primary mediators of inflammation in OA and they activate a distinct set of OA-related genes (Shi et al., 2004).

Cells were seeded into 24-well plates at a density of 5 × 10⁴ cells/well and subjected to a challenge 48 h post-seeding. The cells were categorized into four distinct groups (n = 4): (1) negative control, (2) positive control, (3) TUR, and (4) TUR + SR144528. The group exposed to the antagonist (group 4) underwent a pretreatment with SR144528 at 1 μM for 10 min. Then, both the TUR and TUR + SR144528 groups received treatment with TUR at 15 ppm for 2 h. Following this, a challenge was induced using a mixture of IL-1β and TNF-α both at 10 ng/mL for 24 h, in the presence of treatments. After 24 h, cells were rinsed with DPBS, harvested for RNA extraction, and subjected to qPCR analysis. Supplementary Figure S2 reports the inflammatory challenge optimization.

Total RNA extraction from challenged cells was performed using the NucleoSpin RNA kit (Macherey-Nagel Inc., Bethlehem, United States) following the manufacturer’s protocol. The RNA yield and quality were assessed spectrophotometrically by measuring absorbance at 260 and 280 nm with a Varioskan™ LUX (Thermo Fisher Scientific, Milan, Italy). Samples with a 260/280 ratio below 2.0 were excluded from subsequent analyses.

Afterward, the RNA underwent reverse transcription using the iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, California, United States) according to the manufacturer’s instructions. For qPCR reactions, duplicate assays were conducted using the CFX Connect Real-Time PCR System and iTaq Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, California, United States). Gene expression data were normalized using two reference genes: ribosomal protein L13 (RPL13) and TATA-binding protein (TBP). The 2^−ΔΔCT method (Livak and Schmittgen, 2001) was employed to calculate the gene expression fold change, which is reported in the results as a fold of change relative to the negative control group.

Details regarding primer sequences, expected product length, and GenBank accession numbers are provided in Table 1. Primers were designed using the Primer-BLAST tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) and subsequently obtained from Merck Life Science S.r.l.

Immunofluorescence (IF) staining was performed for total ERK, and cellular myelocytomatosis oncogene (c-Myc), using the protocol already reported by Ghiselli et al., 2021. Briefly, SW-1353 cells were seeded at a of 2.5 × 10⁴ cells/well onto Nunc Lab-Tek II Chamber Slide System (Thermo Fisher Scientific, Milan, Italy), and exposed to an inflammatory challenge as described in Section 2.6 . Cells were then fixed with 4% paraformaldehyde in DPBS for 20 min, followed by permeabilization using 0.5% Triton X-100 (VWR, Radnor, Pennsylvania, United States) for 15 min. Subsequently, the unspecific bond sites were blocked with 10% goat serum for 1 h. Primary monoclonal antibodies, as specified in Table 2, were diluted in a solution containing 2% bovine serum albumin and 0.05% saponins (Alfa Aesar, Haverhill, Massachusetts, United States) in DPBS. Cells were incubated with primary antibodies for 3 h at room temperature in a humidified chamber. Detection of bound primary antibodies was carried out using secondary antibodies conjugated to fluorescein isothiocyanate or tetramethylrhodamine for 1 h (dilutions reported in Table 2), followed by two washes with 0.2% bovine serum albumin and 0.05% saponins in DPBS. Finally, the slides were mounted with Fluoroshield containing 4′,6-diamidino-2′-phenylindole dihydrochloride (DAPI) to stain the nuclei. Images were captured from three different fields using a Nikon Eclipse Ci fluorescence upright microscope at either 20x or 40x magnification (Nikon Corporation - www.nikon.com) and processed using NIS-Elements software (Nikon Corporation - www.nikon.com). To quantify fluorescence intensity, images were analyzed using ImageJ2 (Rueden et al., 2017). For each group, the intensity of total ERK and c-Myc was measured by selecting four random square regions of interest (ROIs) in two different images. Background intensity was measured by selecting an area with no cells to ensure accurate correction. The integrated density for each ROI was calculated as: I n t e g r a t e d   d e n s i t y = A r e a   o f   t h e   s e l e c t e d   R O I × R O I   f l u o r e s c e n c e

The corrected total cell fluorescence (CTCF) was then determined using the following formula: C T C F = I n t e g r a t e d   d e n s i t y − A r e a   o f   t h e   s e l e c t e d   R O I × B a c k g e o u n d   f l u o r e s c e n c e

The figure legends provide detailed information regarding sample sizes and the statistical methods employed. All experiments were performed in duplicate to ensure reproducibility and accuracy of the results. All data are presented as mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism version 10.2.3. The distribution of data was assessed using the Shapiro-Wilk test, with p-values greater than 0.05 indicating a normal distribution. Outliers were identified and excluded using the ROUT method, with a False Discovery Rate (Q) set above 1%.

Each measurement was taken from distinct samples to ensure independent observations. In qPCR experiments, samples with a 260/280 ratio below 2.0 were excluded from the analysis to ensure RNA quality. The datasets for viability (n = 8), oxidative stress (n = 8), qPCR (n = 4), cAMP quantification (n = 6), and immunofluorescence quantification (n = 8) were normally distributed and analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test. Statistical significance was set at p < 0.05.

To determine the highest safe dose of TUR for the study, a viability test was performed. Figure 1 reports the viability assay results. After 24 h of treatment, TUR concentrations above 15 ppm significantly reduced cell viability (p < 0.0001). Consequently, 15 ppm TUR was selected for subsequent analyses. This dose corresponded to 2.73 ± 0.39 ppm CUR (7.41 ± 1.06 μM), 0.90 ± 0.13 ppm DMC (2.66 ± 0.38 μM), and 0.79 ± 0.11 ppm BDMC (2.56 ± 0.36 μM).

CUR, the main bioactive compound of turmeric, has gained significant scientific consideration for its remarkable anti-inflammatory and antioxidant effects, thus helping to slow down OA progression (Russo, 2016). Turmeric extracts such as TUR contain CUR and two important analogs of CUR: BDMC and DMC, which have exhibited varying interesting activities (Almogi-Hazan and Or, 2020). Some research groups have published systematic reviews and meta-analyses comparing the efficacy of CUR alone and turmeric extracts but they were unable to draw definitive conclusions due to the limited number of valid studies (Daily et al., 2016; Shokri-Mashhadi et al., 2021; Wang et al., 2021; Zeng et al., 2021). In the context of CUR’s action on CBR2, Pawar et al. (2022) reported that CUR could selectively act as an agonist on CBR2, reducing inflammation within the myocardium of diabetic mice with myocardial infarction (Pawar et al., 2022). This anti-inflammatory effect was then inhibited by the CBR2 receptor antagonist AM630 (6-Iodopravadoline) (Pawar et al., 2022).

At the time of this publication, no studies have yet explored the role of CBR2 in the observed beneficial properties of CUR or turmeric extracts against OA. This study aimed to investigate the ability of TUR to activate CBR2 and to understand the role of CBR2 in TUR’s anti-inflammatory and antioxidant properties using a CBR2 antagonist in an OA in vitro model.

For this study, after conducting a viability assay, 15 ppm TUR was selected as treatment. Based on titration results, this dose corresponded to 2.73 ± 0.39 ppm CUR (7.41 ± 1.06 μM), 0.90 ± 0.13 ppm DMC (2.66 ± 0.38 μM), and 0.79 ± 0.11 ppm BDMC (2.56 ± 0.36 μM). According to the review by Dei Cas and Ghidoni. (2019), CUR and curcuminoids can exhibit a wide range of serum concentrations, ranging from 0.001 to 3.2 ppm, depending on the subject’s health status and administered dosage (Dei Cas and Ghidoni, 2019). The selected TUR dosage falls within this range, ensuring that a relevant concentration has been used.

In the present study, a docking analysis was conducted on CUR and its analogs targeting CBR2. CUR and curcuminoids showed low predicted binding energies, ranging from −9.6 to −9.7 kcal/mol, suggesting strong potential interactions with CBR2. Typically, predicted binding energies lower than −8.0 kcal/mol indicate possible robust interactions between a ligand and its receptor (Ti et al., 2024). CUR and its analogs showed strong binding affinities for CBR2, similar to the known agonist BCP (Mannino et al., 2023).

The docking analysis also revealed that SR144528, a well-known CBR2 antagonist (Rinaldi-Carmona et al., 1998), interacts strongly with key residues such as TRP258, PHE183, and PHE281, which were highlighted in crystallographic studies by Hua et al. (2020) as crucial for CBR2 conformational changes (Hua et al., 2020). This interaction stabilizes the receptor in its inactive conformation, as reported by Li et al. (2019) (Li et al., 2019), effectively preventing the activation of intracellular Gα inhibitory proteins, which normally inhibit the production of cAMP (Bouaboula et al., 1999). In contrast, CUR and its derivatives (BDMC and DMC) acted as CBR2 agonists, despite being predicted to interact with similar residues within the receptor’s active site as SR144528. Unlike SR144528, these compounds appear to promote the receptor’s active conformation, thereby probably facilitating Gα inhibitory proteins activation. In fact, they maintained low cAMP levels upon FSK stimulation, consistent with the inhibitory effect of CBR2 on adenylyl cyclase (Mallipeddi et al., 2017). This agonistic activity was comparable to that of BCP, another selective CBR2 agonist, which has been well-documented for its therapeutic effects mediated through CBR2 activation (Mannino et al., 2023). Although the docking analysis and cAMP assay suggest strong interactions between TUR components and CBR2, these results should also be validated through other approaches such as X-ray crystallography or site-directed mutagenesis to confirm the binding mechanisms. However, these predicted molecular interactions translated into significant biological effects, particularly in modulating pathways crucial for inflammation and oxidative stress in OA.

As mentioned before, an important aspect of OA involves the degeneration of cartilage, which can be worsen by oxidative stress. It plays a crucial role in the progression of the disease by promoting chondrocyte apoptosis, as highlighted by Lepetsos and Papavassiliou, 2016. Menadione was used in this study to induce ROS generation, creating an oxidative stress condition (Loor et al., 2010). TUR exhibited a strong antioxidant effect, which was blocked by pretreatment with SR144528. This antioxidant action was likely mediated through HMOX-1 upregulation, as it was shown to be induced by TUR and actively counteracted by the CBR2 antagonist. Since ROS-induced damage, along with pro-inflammatory mechanical stimuli, can trigger pathways like MAPK or NF-κB, it is crucial to explore how TUR modulates these pathways.

The anti-inflammatory effects of TUR, shown by reduced IL-6 and COX-2 mRNA expression during the inflammatory challenge, suggest an interaction with NF-κB signaling. Giacoppo et al. (2017) demonstrated that CBR2 activation in RAW264.7 macrophages reduces oxidative stress by preventing IκB-α phosphorylation, inhibiting the nuclear translocation of NF-κB, and modulating the MAPK pathway (Giacoppo et al., 2017). These effects were reversed by a CBR2 antagonist (Giacoppo et al., 2017). Esposito et al. (2006) previously reported that CBR2 activation reduces p38MAPK phosphorylation, triggering an inhibitory NF-κB signaling mechanism that prevents NF-κB nuclear translocation and subsequently inhibits pro-inflammatory gene transcription (Esposito et al., 2006).

In this study, TUR significantly reduced NFKB1 expression, an effect that was reversed by a CBR2 antagonist. The NFKB1 gene encodes the precursor protein p105, which plays a critical role in NF-κB signaling. Upon activation, p105 undergoes proteolytic processing to generate p50, a subunit that can form various dimers, which are crucial for the transcription of pro-inflammatory genes (Concetti and Wilson, 2018). Additionally, the degradation of p105 not only generates p50 but also releases the associated MAPK kinase TPL-2 (tumor progression locus-2), enabling it to activate the ERK/MAPK cascade, another pathway involved in promoting inflammatory responses (Beinke and Ley, 2004). By reducing the synthesis of p105, TUR may prevent its degradation, thereby blocking TPL-2 and inhibiting the ERK/MAPK cascade.

Mariano et al. (2022) reported that the upregulation and activation of CBR2 are linked to reduced cAMP production, decreased phosphorylated ERK (p-ERK), and lower MMP13 levels (Mariano et al., 2022). In OA, increased total ERK levels are associated with chondrocyte hypertrophy and matrix breakdown through MMP upregulation (Wang et al., 2011; Prasadam et al., 2012). Additionally, activation of ERK signaling inhibits apoptosis in chondrocytes, potentially leading to an accumulation of dysfunctional cells (Djouad et al., 2009; Dong et al., 2022). TUR treatment in this study resulted in significant reductions in cAMP, total ERK, and c-Myc protein levels, along with decreased MMP1 and MMP13 mRNA expression, suggesting a role of TUR in the downregulation of total ERK transcription. Li et al. (2017) further indicated that in chondrocytes, the role of p-ERK is context-dependent, showing that its activation may not always lead to favorable outcomes in inflammatory responses, thus complicating the interpretation of ERK signaling in osteoarthritis (Li et al., 2017).

These effects were reversed by the CBR2 antagonist, which increased cAMP, total ERK, c-Myc, and MMP expression. This suggests that TUR’s effects on inflammatory markers and MMP are likely mediated by CBR2 activation, emphasizing the key role of CBR2 in the MAPK and NF-κB pathways, as noted in previous studies. Moreover, TUR also significantly upregulated CBR2 mRNA expression, an effect then counteracted by the receptor antagonist. In summary, TUR’s ability to modulate the MAPK and NF-κB pathways via CBR2 activation provides an interesting molecular basis for its anti-inflammatory effects, which could be pivotal in managing OA-related inflammation and cartilage degradation.

Furthermore, Louvet et al. (2011) reported that CBR2 activation can counteract inflammation by upregulating HMOX-1 (Louvet et al., 2011). In this study, TUR also strongly upregulated HMOX-1 mRNA levels, which were reduced by the presence of SR144528. HMOX-1 not only codes for the antioxidant enzyme heme oxygenase 1 (HO-1) but also modulates immune responses (Sebastián et al., 2018), suggesting that its induction by CBR2 activation could be a critical component of the effects observed with TUR (Ryter et al., 2006).

Finally, TUR also reduced the mRNA expression of DAGL-α and MAGL. MAGL is an enzyme involved in converting 2-AG into arachidonic acid. This, coupled with COX-2 downregulation, could lead to reduced prostaglandin production and increased 2-AG accumulation, which may act as an endogenous agonist of CBR2 (Gonsiorek et al., 2000), further enhancing anti-inflammatory effects. However, TUR’s effects on DAGL-α and MAGL appear to be unrelated to CBR2 activation and mRNA upregulation, as pre-treatment with SR144528 did not influence their expression levels during the challenge, compared to TUR alone. Future studies could investigate other signaling pathways or regulatory mechanisms through which TUR may modulate DAGL-α and MAGL expression, providing further insight into its effects on these enzymes.

Currently, TUR is a promising complementary treatment for OA, with the potential to reduce dependence on nonsteroidal anti-inflammatory drugs and other medications known for their significant side effects, as several clinical studies have been reported on ClinicalTrials.gov. For instance, Paultre et al. (2021) reviewed ten studies comparing turmeric therapy with nonsteroidal anti-inflammatory drugs or no treatment, finding that turmeric significantly improved OA-related pain and joint function from baseline, all while minimizing adverse events (Paultre et al., 2021).

While this study highlights the promising interaction between TUR compounds and CBR2, it is crucial to acknowledge that the role of CBR2 in inflammation is not yet fully understood. While many studies support the anti-inflammatory potential of CBR2 activation, conflicting results in the literature suggest that CBR2’s role in joint inflammation and OA is highly context-dependent. Schuelert et al. (2010) reported that CBR2 agonists unexpectedly exacerbated pain in an OA model, highlighting the complexity of CBR2’s role in pain and inflammation modulation (Schuelert et al., 2010). Moreover, Turcotte et al. (2016) emphasized that CBR2 activation has strong positive effects but they are significantly influenced by the specific cell types and disease models involved (Turcotte et al., 2016). Rzeczycki et al. (2021) observed the upregulation of CBR2 in synovium following joint injury, leading to anti-inflammatory effects (Rzeczycki et al., 2021). On the other hand, Fechtner et al. (2019) found that CBR2 knockdown reduced inflammation in rheumatoid arthritis fibroblasts, indicating that CBR2’s role might differ across inflammatory joint diseases (Fechtner et al., 2019). Mariano et al. (2022) further demonstrated that CBR2 activation in osteoarthritic synoviocytes reduced inflammation through p-ERK and MMP13 downregulation, suggesting that CBR2’s signaling pathways may be particularly relevant in OA (Mariano et al., 2022). These varying outcomes underscore the need for further research to delineate the specific conditions under which CBR2 exerts protective versus unfavorable effects. Furthermore, the absence of quantitative protein assessments, such as cytokine measurements via enzyme-linked immunosorbent assay (ELISA) and CBR expression analysis through western blotting, poses a limitation of the present study. These methods could provide a more nuanced understanding of TUR’s impact on specific inflammatory proteins and receptor engagement, thus supporting the observed molecular interactions at a protein level. Future studies incorporating these techniques may yield a clearer picture of TUR’s mechanistic pathways of its anti-inflammatory effects.

It is also important to acknowledge that CUR and curcuminoids exhibit broad-spectrum activity, impacting various intracellular pathways (Deng et al., 2023), which cannot be fully included within the scope of a single in vitro study. Deng et al. (2023) demonstrated that CUR influenced more than 1,050 differential mRNAs in transcriptomic analysis during OA in human articular chondrocytes (Deng et al., 2023). Moreover, to enhance the understanding of specific mechanisms, future studies should consider investigating the effects of purified components of TUR on ECS and validating these findings in appropriate in vivo models.

Another critical challenge in translating in vitro findings into clinical practice is the bioavailability of CUR and its analogs. Despite their in vitro effects, the poor systemic bioavailability of these compounds poses a significant challenge to their efficacy in vivo. Various strategies, including the development of nanoparticles, liposomes, and phospholipid complexes, have been explored to enhance the bioavailability of TUR components (Bučević Popović et al., 2024). However, concerns have been raised about formulations that enhance CUR’s bioavailability, potentially leading to an increased risk of adverse effects (Turck et al., 2021).

In conclusion, while this study offers promising insights into TUR’s potential mechanisms, it also underscores the complexity of CBR2’s role in inflammation, highlighting the need for ongoing research to fully understand it.