Research article anti-inflammatory and cartilage-sparing properties of hemp polyphenol extract in an in-vitro model of cartilage inflammation

Cartilage has a limited capacity for self-repair due in part to its avascular nature, making proper mechanical loading and nutritional support essential for maintaining joint health. Nutraceuticals derived from agricultural by-products present an opportunity to enhance animal health while retaining product value within the production system. Hemp polyphenol extract, a product derived from a by-product of hemp plant processing, contains bioactive compounds such as cannabidiol (CBD) and flavonoids that have been associated with anti-inflammatory and chondroprotective effects. The objective of this study was to investigate the potential anti-inflammatory and/or chondroprotective properties of a simulated digest of hemp polyphenol extract (HPE) on porcine cartilage explants challenged with lipopolysaccharide (LPS). Two experiments were conducted to determine non-cytotoxic, effective concentrations of HPE. Cartilage explants were aseptically collected from the intercarpal joints of market-weight pigs (Experiment 1: n = 9; Experiment 2: n = 12) and cultured at 37 °C and 7% CO2 for 120 h. During the final 96 h, explants were conditioned with varying concentrations of HPE (1.22, 2.44, 7.32, 12.2, or 24.4 mg/mL) and stimulated with LPS (0 or 10 μg/mL) during the final 48 h. Samples of culture media were collected prior to LPS stimulation and at 24 and 48 h post-stimulation for analysis of inflammatory biomarkers [prostaglandin E2 (PGE2), nitric oxide (NO), and resolvin D1 (RvD1)]. Cartilage structural integrity was assessed by quantifying glycosaminoglycan release into media (mGAG) and retention within explants (tGAG), and explants were stained to evaluate cell viability. In Trial 1, HPEtreatments (1.22, 2.44, and 24.4 mg/mL) did not markedly alter inflammatory biomarkers but reduced cell viability at the highest dose (24.4 mg/mL). Based on these results, Trial 2 was conducted using intermediate concentrations (2.44, 7.32, and 12.2 mg/mL). Conditioning of explants with 7.32 and 12.2 mg/mL significantly reduced PGE2 and NO concentrations compared to controls, while RvD1 was elevated across all treatments, with the highest concentrations observed at 12.2 mg/mL across all time points. At 12.2 mg/mL mGAG was decreased while tGAG was increased. Cell viability remained above 99% in all treatment groups. These findings indicate that intermediate doses of HPE may provide anti-inflammatory and chondroprotective benefits, supporting the potential use of hemp by-products as functional nutraceutical ingredients for joint health.

Articular cartilage is a specialized connective tissue that plays a critical role in the biomechanics of synovial joints, enabling smooth movement and efficient transfer of mechanical loads while minimizing friction (Karpiński et al., 2025). Despite its essential function, articular cartilage possesses a limited capacity for self-repair, and injury or chronic mechanical stress often leads to progressive tissue degeneration and development of osteoarthritis (OA) (Armiento et al., 2019). Osteoarthritis is characterized by the gradual loss of hyaline cartilage, chronic inflammation within the joint, and remodeling of the underlying subchondral bone. This collectively contributes to impaired joint function and pain (Lee et al., 2026). Current clinical strategies for the management of OA remain limited, as pharmacological treatments primarily provide symptomatic relief, while surgical interventions rarely restore full tissue function (Zheng et al., 2026; Chen et al., 2026). Consequently, attention has been directed toward new treatment methods and investigations into compounds that modulate inflammatory pathways and slow cartilage degeneration, as potential strategies for disease prevention or modification (Wang et al., 2023).

One of these strategies is the use of nutraceuticals and dietary supplements as alternatives or adjuncts to established therapeutic strategies (Colletti and Cicero, 2021). These products are being investigated for their potential to reduce inflammation, oxidative stress, and pain while supporting cartilage metabolism and tissue matrix integrity (Colletti and Cicero, 2021). While many supplements claiming to target inflammatory pathways and cartilage health have entered the market, few have scientific evidence supporting their efficacy (Golla, 2018). For this reason, objective scientific investigations of the anti-inflammatory and chondroprotective properties of plant-derived bioactive compounds remain an important step toward the development of evidence-based nutritional interventions for joint health.

Cannabis sativa L. has a long history of use as a medicinal plant, with other applications including fiber supplementation, and as a food (Pisanti and Bifulco, 2018). Historically, cannabis has been cultivated, processed, and administered in multiple forms as medical therapy to treat disease or alleviate symptoms (Whiting et al., 2015), making it an increasingly relevant subject of study within the context of nutraceutical research. The plant is compositionally complex and contains a wide array of constituents, the most widely recognized being the cannabinoids (ElSohly and Slade, 2005). Industrial hemp varieties of cannabis sativa have gained interest as a source of biologically active constituents. Recently, research has identified numerous non-cannabinoid compounds that may contribute to hemp’s medicinal activity (Radwan et al., 2008), some of which include prenylated flavones known as cannflavins. These compounds have been shown to interact with inflammatory signaling pathways (Allegrone et al., 2017), specifically inhibiting microsomal prostaglandin E2 synthase-1 and 5-lipoxygenase pathways (Werz et al., 2014). They are present in the waste products of hemp cannabidiol (CBD) oil extraction, and thus present an opportunity for a value-added byproduct of the CBD industry. For this reason, a hemp polyphenol extract (HPE) derived from this residual biomass may be a source of compounds capable of influencing inflammatory signaling and cartilage metabolism, with it being hypothesized that these compounds support cartilage homeostasis and modulate inflammatory signaling. To evaluate this, the current study employed a well-established cartilage explant model paired with a simulated digestion protocol of the product to assess cartilage responses to inflammatory stimuli in the presence of HPE (Werz et al., 2014; Pearson and Kott, 2019; Pearson et al., 2007a).

All chemical reagents and materials were purchased from Sigma Aldrich (Mississauga, ON, Canada) unless specified otherwise.

Hemp polyphenol extract

The product used in this study was a polyphenol-enriched extract (HPE) manufactured by FSOIL (Woodburn, Oregon, United States) from the residual biomass generated during CO 2 extraction of hemp, using the Canurta patented extraction process. The resulting extract was resuspended in ethanol with residual water. Chemical characterization of the extract was performed using high-performance liquid chromatography (HPLC), which quantified the concentrations of key bioactive components, including cannflavin A (CFA), cannflavin B (CFB), cannabidiol (CBD), and cannabidiolic acid (CBDA). The measured concentrations were 0.11 mg/mL for CFA, 0.06 mg/mL for CFB, 2.50 mg/mL for CBD and 3.33 mg/mL for CDBA.

Simulated digestion and hepatic metabolism

HPE was processed following previously established simulated digestion protocols (Pearson and Kott, 2019; Pearson et al., 2007a; Pearson et al., 2010). An effective concentration for CFA was provided by the HPE manufacturer and was used to estimate a target experimental concentration. The HPE product was provided suspended in ethanol and residual water. Product density was 0.766 g/mL. Using this reference value, a total volume of 6.88 mL of HPE was added to 100 µL of Tween 20 and was shaken by hand for 5 min. It was then suspended in 17.5 mL of simulated gastric fluid (37 mM NaCl, 0.03N HCl, 3.2 mg/mL pepsin) and was shaken at 37 °C for 2 h (Pearson and Kott, 2019). Following gastric incubation, acidity was neutralized by adding equinormal amounts of 2.2M NaOH bringing solution to a final pH of 6.0. Immediately following pH correction, 18.25 mL simulated intestinal fluid (30 mM K 2 HPO 4, 160 mM NaH 2 PO 4, 362 mg pancreatin) was added to gastric digest and was incubated and shaken at 37 °C for an additional 2 h. After the intestinal incubation period, the digest was centrifuged at 4900 rpm for 20 min at 4 °C, and the supernatant was collected and re-spun under the same conditions. To mimic hepatic metabolism, rat liver microsomes (male) (Sigma Aldrich, Mississauga ON Canada) were added to a final concentration of 0.03 mg/mL followed by the addition of NADPH (100 μg in 0.01 M NaOH). The solutions were then incubated for 30 min at 37 °C under 7% CO 2, then centrifuged at 4900 rpm for 30 min at 4 °C. The supernatants were filtered through 0.45 μm filters and then 0.22 μm. Remaining digest was transferred into 50 kDa ultrafiltration centrifuge units (Amicon Ultra; Millipore, Mississauga ON) and centrifuged at 4900 rpm for 25 min at room temperature. A control (blank) digestion product was prepared using the same method but without the inclusion of any HPE (Pearson et al., 2007a).

Trial structure

Two cartilage explant trials were conducted consecutively, using identical core methodologies to evaluate the biological effectiveness of a novel supplement for which effective concentration parameters have not yet been clearly defined in the literature. The initial trial was designed as an exploratory study to assess the presence of biological effects, and the results from this trial were used to inform the subsequent concentrations investigated in the second trial. In Trial 1, cartilage explants were obtained from nine pigs (n = 9) and HPE digest treatments were administered at concentrations of 1.22, 2.44, and 24.4 mg/mL. In Trial 2, cartilage explants from twelve different pigs (n = 12) were used, and concentrations of 2.44, 7.32, and 12.2 mg/mL were evaluated.

Explant model of cartilage inflammation

Cartilage explants were prepared and maintained according to the method described by Pearson et al. (2010) (Pearson et al., 2010). Porcine cartilage was obtained from a commercial, federally inspected meat processing facility within 1 hour after humane slaughter. Distal forelimbs, inclusive of the intact carpal joint, of 9 and 12 (for trials 1 and 2, respectively) commercial market weight pigs were placed on crushed ice and transported from the facility to the University of Guelph, Guelph, ON. Cartilage explants from the intercarpal joint were aseptically harvested using a 4 mm biopsy tool. Twenty explants were removed from each animal washed three times with dissection media [Dulbecco’s Modified Eagle Medium (DMEM) supplemented with NaHCO 3 ] before they were placed, 2 explants per well in a 24-well tissue culture plate with 1000 µL of tissue culture media (TCM). The TCM was comprised of DMEM supplemented with amino acids, sodium bicarbonate (NaHCO 3), lactalbumin hydrolysate, sodium selenite, magnesium sulfate, amphotericin B and antibiotics (10 mL/L containing 10,000 units penicillin; 10 mg/mL streptomycin). The TCM was pH corrected to 7.3 and filtered through a 0.22 um bottle top filter. Once filtered, sterile fetal bovine serum (FBS) was added to reach a final concentration of 10%) (Chan et al., 2005). Explants were subject to an initial acclimatization period where they were cultured for 24 h at 37 °C with 7% CO 2. After a 24 h acclimatization period, TCM was removed from the wells and replaced with 900 µL of fresh TCM. The remaining 100 µL consisted of a mixture of simulated digested product containing the HPE and a simulated digest blank (no HPE). These components were combined at calculated ratios to achieve the target HPE concentrations in the final well volume. Final treatment concentration labels were as follows:

TCM and HPE treatments were collected and refreshed every 24 h. After the initial 72 h of culture and 48 h of conditioning with the HPE extract, explants were stimulated with or without 10 μg/mL of LPS (Strain O111:B4 Trial 1, O55:B5 Trail 2). During the final 48 h (immediately prior to stimulation with LPS) samples were collected into 1.5 mL centrifuge tubes containing 10 µL of indomethacin (10 mg/ml) and immediately frozen at −20 °C until analysis. At the end of the 120 h culture duration all explants were immediately weighed and one explant per well was stained for viability (see Cell Viability staining below) while the other was frozen at −80 °C for tissue GAG analysis (see below).

Sample analysis

Samples were analyzed for biomarkers associated with cartilage inflammation response, matrix homeostasis and chondrocyte viability. Inflammation mediators quantified in TCM included prostaglandin E 2 (PGE 2), nitric oxide (NO), and resolvin D1 (RvD1). All microplate assays were read using Biotech Cytation 5 Cell Imagine Multimode Reader, Agilent Technologies, Inc. (VT, United States).

Cartilage matrix turnover was evaluated by quantifying net glycosaminoglycan release into the TCM (mGAG) as well as GAG retention within the cartilage explants (tGAG). Chondrocyte viability within explants, and bioactive composition of HPE and HPE simulated digest were also analyzed.

Prostaglandin E 2

All TCM samples were analyzed for PGE 2 using a commercially available ELISA kit (Arbour Assays; cat #K051-H5) according to manufacturer instructions. Plates were read immediately at absorbance of 450 nm. A best-fit third order polynomial standard curve was developed for each plate (R2 ≥ 0.98), and these equations were used to calculate PGE 2 concentrations for samples from each plate.

Media GAG

The concentration of GAGs in TCM (mGAG) was measured using a spectrophotometric assay with (1,9-DMB) as originally described by Chandrasekhar et al. (1987) (Chandrasekhar et al., 1987). Samples were analyzed in triplicate in 96-well plates where10 µL of samples were added directly to wells containing 140 µL dilution buffer, followed by the addition of 25 µL of guanidine hydrochloride. Then 175 µL of DMB reagent was added, and absorbance was measured at 530 nm within 5 min. Concentrations of mGAG were determined by generating a best-fit linear standard curve (R2 ≥ 0.97) for each plate using a bovine chondroitin sulfate standard (Sigma, Oakville ON). This equation was used to calculate total mGAG concentration for samples from each plate.

Tissue GAG

Frozen explants were digested in papain prior to analysis using the 1,9-dimethylmethylene blue (1,9-DMB) assay according to Crosbie et al. (2025) (Crosbie et al., 2025b). Explants were blotted dry, weighed, and cut into 6–8 pieces before being placed into microcentrifuge tubes. Each explant was incubated with 600 µL of papain digestion solution containing ammonium acetate (2.6 mg/mL), Na 2 EDTA·2H 2 O (0.38 mg/mL), DL-dithiothreitol (0.31 mg/mL), and papain (40 μg/mL). Samples were digested in a shaking water bath at 65 °C for 72 h. Following digestion, samples were frozen and subsequently analyzed for tissue glycosaminoglycan (tGAG) content. Then tGAG concentrations were quantified using the same assay protocol mGAG measurements. Values were corrected for a 30-fold dilution factor and normalized to explant dry weight and are reported as µg tGAG per mg of cartilage tissue.

Nitric oxide

Nitrite (NO 2-), a stable oxidation product of NO, was analyzed using the Griess reaction. Undiluted TCM samples were added to 96-well plates followed by sulfanilamide (0.01 g/mL) and N-(1)-Napthylethylene diamine hydrochloride (1 mg/mL) in phosphoric acid (0.085 g/L) to each well. Sample absorbance was read within 5 min at 530 nm and was compared to a sodium nitrite standard curve. A best-fit linear standard curve was developed for each plate (R 2 ≥ 0.99), and these equations were used to calculate nitrite concentrations for samples from each plate.

Resolvin D1

All TCM samples were analyzed for RvD1 using a commercially available ELISA kit (MyBioSource; cat #MBS756429) according to manufacturer instructions. Plates were read immediately at absorbance of 450 nm. The standard curve was determined by linear regression analysis of the log of the standard concentrations against the log of the optical density (OD) absorbance readings. (R2 ≥ 0.95) and was used to estimate concentrations of RvD1 in each sample.

Cell viability

Viability of cells within cartilage explants was determined using a Calcein-AM (C-AM)/Ethidium homodimer-1 (EthD-1) Live/Dead Viability/Cytotoxicity Kit, for mammalian cells (ThermoFisher Scientific; cat #L3224). C-AM and EthD-1 were mixed in sterile distilled water at concentrations of 4 μM and 8 μM, respectively. Explants were placed individually into wells of a 96-well microtiter plate and incubated in 200 μL of the C-AM/EthD-1 solution for 40 min at room temperature. Then C-AM and EthD-1 fluorescence in explants were measured using excitation/emission filters of 485/530 nm and 530/685 nm, respectively. Cell viability was determined using the following equation: C-AM/C-AM + EthD-1.

Analysis of HPE and HPE simulated digest

Bioactive components of HPE were quantified by HPLC. Analyses were conducted on the native HPE, the simulated digestion product prior to 50 kDa filtration (retentate), and the fraction obtained after passage through a 50 kDa ultrafiltration centrifugal unit (filtrate). Filtrate (16 mL) was frozen and lyophilized, then resuspended in 5 mL of water. Retentate (11 mL) was used as is. Samples were extracted twice with 20 mL of ethyl acetate and dried under nitrogen gas then resuspended in 1 mL of methanol and then centrifuged for 20 min at 4,696 g to pellet any debris. Samples were resolved using an Agilent 1260 Infinity I HPLC equipped with an Agilent Poroshell 120 EC-C18 (4.6 mm × 150 mm 2.7 µm; Agilent) column at a flow rate of 1.0 mL/min, heated at 25 °C, with solvent A (water, 0.1% formic acid) and solvent B (methanol, 0.1% formic acid). The gradient began at 74% solvent B and increased to 81% solvent B over 13 min, then increased to 95% B over 5 min. Solvent B was then held at 95% for 6 min. CBD and CBDA were quantified at 230 nm and CFA and CFB were quantified at 340 nm. All samples were quantified relative to authentic standards.

Statistical analysis

Linear mixed models in R Studio (2025.09.2 + 418 by Posit Software, PBC) were used to assess treatment effects on NO, PGE 2, mGAG, tGAG and RvD1. Treatment and time were set as fixed effects in the model, and interaction effects (Treatment and Time) were included. Random effects of animal, experimental units for repeated measures and error were included to improve the fit of the model. A one-way ANOVA was used to test if the Live/Dead ratio differed significantly between treatment group. For all assays, significance was evaluated with an ANOVA test and a post-hoc Tukey’s test. Results are presented as mean ± SEM and significance was accepted at p < 0.05. Throughout the results, ‘+’ indicates explants stimulated with LPS. The following nomenclature refers to tested concentrations: T0.5 = 1.22 mg/mL, T1 = 2.44 mg/mL, T3 = 7.32 mg/mL, T5 = 12.2 mg/mL, T10 = 24.4 mg/mL.

HPLC analysis

A chromatogram of the HPE digested filtrate is provided in Figure 1. Concentrations of native HPE, simulated digestion product (retentate) and filtrate were reported in Table 1.

Trial 1: (n = 9)

In trial 1, HPE concentrations of T0.5 (1.22 mg/mL), T1(2.44 mg/mL), and T10 (24.4 μg/mL) were evaluated.

Nitric oxide

A significant (p < 0.001) effect of treatment, time and a treatment-time interaction were observed for mean NO concentrations (Figure 2). At 0 h, NO concentrations were not different across all treatment groups. Following LPS stimulation (10 μg/mL), NO production increased in the T0.5, T1 and B (control) groups. There was no effect of LPS stimulation in the T10 group.

Prostaglandin E 2

A significant (p < 0.001) effect of treatment, time, and a treatment-time interaction was observed for mean PGE 2 concentrations (Figure 3). At 0 h, treatment T10 resulted in elevated levels of PGE 2 prior to LPS stimulation (10 μg/mL). Following 0 h, PGE 2 concentrations increased in T0.5, T1 and B (control) in response to LPS, whereas there was no increase in the T10 group.

Media glycosaminoglycan

A significant (p < 0.001) effect of treatment, time and a treatment-time interaction were observed for mean GAG concentrations (Figure 4). At 0 h, GAG release was not different across treatment groups. Following LPS stimulation (10 μg/mL), an increase in GAG release was observed only in the T1 treatment group.

Cell viability

A significant (p < 0.001) effect of treatment was observed for cell viability (Figure 5). The Live/Dead assay indicated significantly reduced cell viability in the T10 (24.4 mg/mL) treatment group in both stimulated and unstimulated explants.

Trial 2: (n = 12)

In trial 2, HPE concentrations of T1 (2.44 mg/mL), T3 (7.32 mg/mL), and T5 (12.2 μg/mL) were evaluated. Stimulation with 10 μg/mL of LPS (O55:B5) was insufficient to induce statistically significant effects on measured assay outcomes therefore, subsequent analyses were conducted without inclusion of LPS as a factor.

Nitric oxide

A significant (p < 0.001) effect of treatment and a treatment-time (p = 0.01) interaction were observed on mean NO concentrations (Figure 6). Time did not reach the significance threshold (p = 0.06). At 0 h, a dose-dependent effect was observed with T5 explants producing the lowest concentration of NO relative to the control group. This pattern was consistent across all time points.

Prostaglandin E 2

A significant (p < 0.001) effect of treatment, time and a treatment-time interaction were observed for mean PGE 2 concentrations (Figure 7). At 24 and 48 h, all HME treatment groups exhibited reduced PGE 2 concentrations relative to the blank control. A time-dependent increase in PGE 2 production was observed in control and T1 groups, while none was found in group T3 or T5.

Media glycosaminoglycan

A significant (p < 0.001) main effect of treatment and time (p = 0.01) was observed for mean mGAG concentrations (Figure 8B). Across all time points, the T5 treatment group exhibited the greatest reduction in GAG concentrations relative to the control and other treatment groups.

Tissue glycosaminoglycan

A significant (p < 0.001) main effect of treatment on total tGAG retention in cultured explants was observed. The T5 treatment group exhibited the greatest overall GAG retention, differing from all other treatment groups and the control (Figure 8A).

Cell viability

A significant (p = 0.02) main effect of treatment on cell viability was observed (Figure 9). The Live/Dead assay indicated reduced cell viability in the T5 (12.2 mg/mL) treatment group; however, viability remained above 99.0% across all treatments.

Resolvin D1

A significant (p < 0.001) effect of treatment on total resolvin D1 (RvD1) release was observed, whereas neither time (p = 0.2) nor the treatment-time (p = 0.8) interaction reached statistical significance. Across all time points, RvD1 concentrations increased in a dose-dependent manner, with the T5 treatment group exhibiting significantly greater RvD1 release compared with all other treatments and the control. The T3 group also demonstrated elevated RvD1 concentrations relative to control and T1; however, these increases were consistently lower than those observed for T5 (Figure 10).

The purpose of this study was to evaluate the effects of a novel nutraceutical product (HPE) on cartilage responses to inflammatory challenge with respect to outcomes related to inflammation [prostaglandin E 2 (PGE 2), nitric oxide (NO)], cartilage structure [tissue and media glycosaminoglycan (GAG)], inflammation resolution [resolvin D1 (RvD1)] and cell viability. The results demonstrated that an HPE in-vitro concentration of 12.2 mg/mL (T5) reduced mGAG loss and increased tGAG retention, while reducing PGE 2 and NO. Additionally, supplementation of HPE at this concentration increased production of the pro-resolving mediator RvD1 across all time points. Importantly, 12.2 mg/mL did not elicit cytotoxic effects or compromise chondrocyte viability when compared to higher concentrations (24.4 mg/mL).

Preservation of GAG retention within explants represents one of the key findings of this study. While no clear trends were observed in mGAG when testing low (1.22 mg/mL) and high (24.4 mg/mL) concentrations of HPE, analysis of explants cultured in the presence of 12.2 mg/mL of HPE demonstrated the most pronounced preservation of tGAG and reductions in mGAG. In explant culture, some degree of mGAG is expected to be released into media following the event of cartilage harvest due to mechanical injury of the tissue which has been shown to induce GAG release from the extracellular matrix (D’lima et al., 2001). In this context the results of this study demonstrated that mGAG content was reduced by HPE throughout the duration of culture. Mechanistically, this is likely due to the presence of cannabinoids within the HPE, as previous research has shown synthetic cannabinoid agonists’ effects on reducing GAG loss in cultured cartilage (Shang et al., 2023b; Mbvundula et al., 2005; Mbvundula et al., 2006).

Additionally, the noted reduction in NO may represent a key mechanism underlying the observed decreased GAG release and enhanced GAG retention. Production of NO decreased in a consistent, treatment-dependent manner across all sampling time points, with the greatest suppression observed in the T5 (12.2 mg/mL) treatment group. Nitric oxide has been shown to be inhibited in the presence of cannabinoids previously (Mbvundula et al., 2005; Mbvundula et al., 2006). Additionally, NO is a critical mediator of chondrocyte responses to inflammatory and mechanical stress, and its effects on cartilage homeostasis are concentration- and exposure-dependent (Xian Bo et al., 2022). Under inflammatory conditions, NO is primarily generated via inducible nitric oxide synthase (iNOS), and excessive production promotes cartilage catabolism by stimulating PGE 2 and matrix metalloproteinase expression while suppressing proteoglycan and collagen synthesis (Ahmad et al., 2020). In disease states such as OA, sustained NO elevation contributes to extracellular matrix degradation and progressive loss of cartilage integrity, while inhibition of NO signaling has been associated with increased expression of anabolic extracellular matrix markers, including aggrecan (ACAN) and type II collagen (COL2A1) (Xian Bo et al., 2022). Therefore, this marked reduction of NO in response to HPE could be due to the downstream signaling pathways preventing extracellular matrix breakdown and supporting the preservation of cartilage matrix integrity by limiting GAG loss following explant harvest.

Proteoglycan breakdown is an early event in cartilage degradation, and this represents an important point of intervention (Little et al., 2002). The findings of this experiment provide support for the use of this HPE to support cartilage health. Although the current experimental design does not distinguish between retained native tGAG and GAG synthesized post-harvest, interpreting mGAG and tGAG results together provides a better understanding of how cartilage tissue seeks to establish homeostasis following a mechanical disruption. Future studies could incorporate approaches that differentiate pre-existing GAGs from those synthesized post-harvest, such as isotopic labeling (e.g., [S 35 ]-methionine) or click-chemistry techniques that label newly synthesized GAGs and chondrocyte DNA, enabling more precise evaluation of supplement-driven effects on cartilage matrix homeostasis and turnover dynamics (Porter et al., 2022).

In the second study, HPE treatment T3 and T5 significantly reduced PGE 2 production relative to controls at 24 and 48 h. While PGE 2 concentrations exhibited a time-dependent increase in control and T1 explants, concentrations in T3 and T5 groups did not change over time. This stabilization suggests that the inclusion of HPE at these concentrations (7.32 and 12.2 mg/mL) inhibited the progression of inflammatory signaling within cartilage explants. In support of this, the bioactive components of HPE, primarily CBD, have been investigated in the context of modulating the PGE 2 pathway. CBD has previously been investigated in the context of endometriosis and has been shown to reduce PGE 2 production in rat uterine cells (Genovese et al., 2022). In contrast, other studies using human gingival fibroblasts reported increased PGE 2 production following CBD treatment alone, suggesting that CBD-mediated modulation of PGE 2 is likely context- and dose-dependent (Abidi et al., 2022). Future studies should evaluate cartilage explant cultures supplemented with HPE at comparable dose levels under inflammatory challenge conditions to determine whether the observed stabilizing effects on PGE 2 production are maintained in a stimulated inflammatory environment. Together with the reductions in NO, increased GAG retention within explants, and reduced GAG loss these findings provide strong support to the hypothesis that HPE modulates multiple inflammatory axes involved in cartilage extracellular matrix homeostasis.

Beyond attenuation of pro-inflammatory mediators such as NO and PGE 2, HPE supplementation also promoted inflammation resolution. One of the key findings of the present study was the increased RvD1 production in explants cultured with 12.2 mg/mL (T5) HPE. Resolvins are specialized pro-resolving lipid mediators generated during the resolution phase of inflammation (Sun et al., 2007). Importantly, these mediators not only exert anti-inflammatory effects but also actively promote return to tissue homeostasis (Yanoshita et al., 2025). RvD1 functions to limit the magnitude and duration of inflammatory responses by suppressing pro-inflammatory gene expression and mediator production, thereby preventing excessive inflammation that can be harmful to tissues while supporting accelerated recovery and tissue adaptation (Yanoshita et al., 2025). In the context of cartilage health, RvD1 is elevated in synovial fluid from OA joints and has been shown to alleviate bone and joint destruction in arthritic models by suppressing cartilage- and bone-degrading enzymes, which are key contributors to osteoarthritic pathogenesis (Peltner et al., 2023). The current study’s findings of elevated RvD1 following HPE supplementation support a role for HPE in promoting pro-resolving inflammatory processes. Mechanistically, CBD has recently shown to be a potent lipid mediator class switching agent that stimulates the production of pre-resolving mediators such as RvD1 and suppressing the pro-inflammatory biosynthesis (Peltner et al., 2023). This evidence supports what is observed with our reduction in PGE 2 and increased presence of RvD1 with HPE at a concentration of 12.2 mg/mL. Future investigations could expound these effects in the context of an active inflammatory challenge and a similar concentration of HPE to quantify HPE effects on the RvD1 pathway in a different physiological context.

Chondrocyte viability results indicated that there is a physiologically relevant range of concentrations at which HPE has beneficial effects on cartilage health. Concentration of 24.4 mg/mL of HPE significantly reduced cell viability, allowing us to establish an upper limit for HPE inclusion. Mean viability in this treatment group was approximately 90%, whereas all other treatment groups maintained viability exceeding 98%. As a novel nutraceutical product, these studies represent some of the first investigations into the biological efficacy of HPE, and the establishment of a safe range of inclusion is critical to inform future in-vitro and in vivo research.

HPLC analysis indicated that CBD and CBDA are quantitatively the most prominent bioactive compounds present in the HPE digested filtrate compared to CFA and CFB. Consequently, the observed effects of fully digested HPE in the current study is likely attributable to the combined actions of CBD and CBDA. Cannflavins have been reported to exhibit anti-inflammatory activity (Barrett et al., 1986), as well as antioxidant and cytoprotective properties; however, compound-specific oral bioavailability, pharmacokinetics, and metabolite formation remain poorly characterized (Li et al., 2023). Within the present simulated digestion model, the detectable abundance of CFA was markedly reduced following both the digestion process and subsequent filtration. CFB also demonstrated decreased recovery across digestion, although the magnitude of reduction appeared less pronounced relative to CFA. These observations may suggest differential stability, solubility, or partitioning behavior between cannflavin species during gastrointestinal simulation, though the mechanisms underlying these differences remain unclear and warrant further investigation. Analysis of both the retentate and the filtrate following the final 50 kDa filtration step allowed for improved characterization of bioactive reaching the target tissues. Additionally, the digestion model supports the occurrence of hepatic bioactivation and/or bioactive transformations in the product following the addition of rat liver microsomes (Zhou et al., 2004a). Although direct measurement of specific metabolites was not undertaken in this study, the metabolites of CBD have been extensively researched and quantified in other pharmacokinetic studies (Elder et al., 2025). Thus, the observed reduction in CBD levels as reported in our HPLC results could potentially be attributed, in part, to its metabolism. Furthermore, HPE contains a diverse array of phenolic constituents and secondary metabolites that were not quantified in the present HPLC analysis. The potential presence of these uncharacterized compounds, as well as digestion-derived transformation products, may contribute to the overall bioactive profile of the extract.

The HPLC results provide evidence that CBD exposure was likely a primary contributor to the cytotoxic effects observed at an HPE concentration of 24.4 mg/mL, where CBD and CBDA exposure were estimated at 0.64 μM and 49.40 μM, respectively. Although CBD has been studied for its therapeutic applications and is widely used in the management of chronic pain (Montero et al., 2025), it has also been investigated for its anti-cancer properties, where high concentrations have been shown to induce programmed cell death (5–65 μM) (Melo et al., 2025). For these reasons, it is plausible that in the current experiment, high concentrations of CBD induced apoptosis, resulting in reduced cell viability.

Beyond the observed reduction in cell viability, additional biomarker outcomes indicate that chondrocytes exposed to 24.4 mg/mL were under increased physiological stress. Specifically, PGE 2 results demonstrated that explants conditioned at this concentration exhibited significantly elevated basal PGE 2 production at 0 h. This elevation suggests that T10 treatment alone was sufficient to induce a pro-inflammatory or stress-associated response in chondrocytes before the application of an exogenous inflammatory challenge. While CBD supplementation has been associated with reduced PGE 2 production (Genovese et al., 2022), exposure to high concentrations may induce alternative cellular stress responses that counteract these effects in chondrocytes. Notably, PGE 2 levels did not further increase in the T10 group following LPS stimulation, suggesting an impaired inflammatory responsiveness. Rather than reflecting an active inflammatory response, the lack of LPS-induced PGE 2 production is consistent with reduced cellular metabolic activity or progression toward apoptosis (Melo et al., 2025). A similar pattern was observed for NO production, where all other treatment groups exhibited increased NO levels following LPS exposure, whereas the T10 group did not. This trend persisted at 48 h, indicating sustained cellular dysfunction throughout the experimental period. Taken together with the PGE 2 levels and cytotoxicity results, these findings indicate that high levels of HPE supplementation (24.4 mg/mL) induce cellular stress and cytotoxicity in cartilage explants, emphasizing the importance of biologically effective and safe concentrations when evaluating CBD-containing nutraceuticals in tissues with limited regenerative capacity, such as articular cartilage.

Limitations

Several limitations of this study should be acknowledged. The results of this trial are derived from in-vitro models and may not accurately represent in-vivo conditions. Firstly, in vivo, oral nutraceuticals are subject to batch variability, incomplete digestion, variable gastrointestinal absorption, first-pass metabolism, and tissue-specific distribution. Together, these factors may substantially reduce systemic and intra-articular exposure to bioactive compounds found in the HPE product. While the current simulated digestion protocol accounts for upper gastrointestinal tract digestion, hepatic biotransformation, and ultrafiltration to approximate particle sizes capable of entering the joint space, it does not incorporate lower gastrointestinal tract digestion or microbial metabolism, which may further modify the bioactivity of constituent compounds. Species-specific differences in gastrointestinal physiology and pharmacokinetics are also not considered, which may not fully reflect biological complexity in vivo.

Additionally, in trial 2, stimulation with LPS (O55:B5) was insufficient to induce a statistically significant increase in inflammatory outcome measures, limiting the interpretation of treatment effects under inflammatory challenge. This lack of stimulation may reflect reduced LPS activity due to manufacturing, storage, or handling conditions, or alternatively, differential cellular responses, as structurally distinct LPS serotypes are known to exhibit variable potency and downstream signaling profiles (Pieterse et al., 2016).

This highlights the importance of careful selection of inflammatory stimulus used in in-vitro models, as variability in LPS strain can influence the magnitude and expression of the inflammatory response (Pieterse et al., 2016). Ensuring a consistent and sufficiently robust inflammatory challenge is essential for accurately evaluating the treatment effects of novel nutraceuticals with this explant model.

Despite these limitations, the data presented herein provide evidence for anti-inflammatory and anti-catabolic effects of HPE.

The results of this trial indicate that the simulated digested hemp polyphenol extract at 12.2 mg/mL (T5) reduced GAG loss from cartilage matrix and reduced key mediators of inflammation, including PGE 2 and NO. This concentration in vitro also promoted increased chondrocyte production of the pro-resolving mediator RvD1, suggesting a potential shift towards supporting inflammation resolution rather than simple suppression. In contrast to higher concentrations, 12.2 mg/mL did not compromise chondrocyte viability, supporting its safety and efficacy at this inclusion level.