Uncovering the antiinflammatory potential of Lactiplantibacillus Plantarum fermented Cannabis Sativa L seeds

Effects of FHS on protein expression in the LPS-induced RAW 264.7 cell

To investigate the anti-inflammatory mechanisms of FHS, we conducted a western blot technology to assess the protein expression levels in cells treated with FHS. As depicted in Fig. 2a, b, LPS stimulation led to increased inflammation and up-regulated expression of TLR4 and NF-κB p65. Both FHS and corticosterone acetate (positive drug) intervention significantly reduced the expressions of TLR4 and NF-κB p65 compared to the LPS-stimulated group. There was no significant difference between the FHS group and the normal cell group, suggesting that FHS did not induce any adverse effects under normal conditions. Under the inflammatory conditions, activated macrophages produce iNOS located downstream of the TLR4/NF-κB pathway, resulting in the generation of massive amounts of NO¹⁶. Therefore, we also investigated the effect of FHS on iNOS expressions. In Fig. ​Fig.2c,2c, the results indicated that both FHS and drug interventions effectively down-regulated the expression of iNOS in cells upon LPS stimulation. To sum up, FHS significantly reduced the inflammatory response by inhibiting the TLR4/NF-κB pathway in the LPS-stimulated Raw 264.7 cells.

In the inflammatory response progression, macrophage activation and elevated interleukin levels are commonly observed at the cellular level¹². Generally, the stimulator LPS could activate the inflammatory reaction by binding to the TLR4 receptor, resulting in the activation of pro-inflammatory cytokines such as IL-6, TNF-α, and enzymes like iNOS¹⁸. The NF-κB signaling pathway is then established in the inflammatory response based on TLR4 receptor activation¹⁸. Our results demonstrated that the treatment of FHS effectively protects macrophages by reducing the levels of pro-inflammatory cytokines and mediators, including TNF-α, IL-1β, IL-6, IL-18, and NO (Fig. ​(Fig.1).1). Further investigation revealed that one of the mechanisms through which FHS combat inflammatory damage is by modulating the TLR4/NF-κB pathway.

Multivariate analysis of UHPLC-Q-TOF/MS-based untargeted metabolomics

As an emerging method, untargeted metabolomics has been used to obtain high-through data and conduct batch correlation analysis based on the complete profiling of metabolites in food materials¹³. Our above results indicated that a significant improvement in the anti-inflammatory effect of hemp seeds after fermentation. Naturally, such enhancement in biological activity is likely accompanied by changes in various metabolites during the fermentation process. Hence, to gain deeper insights, we subjected the metabolomics data to multivariate statistics, aiming to elucidate the specific metabolite changes during fermentation and establish their correlation with the anti-inflammatory properties. The analysis of metabolomics data not only allows us to comprehend the fermentation-induced metabolic profile changes but also facilitates the identification of important potential inflammation-related compounds present in FHS. As a result of significant alterations of these compounds during fermentation, FHS demonstrates a substantial capacity to inhibit the expression of TLR4 and NF-κB p65 (Fig. ​(Fig.2).2). Supplementary Table 1 listed the changes in overall metabolic compounds in FHS and RHS groups, with a total of 107 identified metabolites and their relative content shown for both groups. Upon preliminary comparison, we observed a significant increase in several organic acid compounds after fermentation, including 2-furoic acid, gluconic acid, indolelactic acid, citric acid, and homovanillic acid. Previous studies have reported the beneficial effects of these compounds in suppressing inflammatory responses by inhibiting NF-κB signaling pathway activation, resulting in the reduced production of mediators and pro-inflammatory cytokines in RAW 264.7 cells^19–22. Notably, indolelactic acid has been closely associated with an inflammatory biomarker²³. The PCA result (Fig. ​(Fig.3a)3a) indicated a significant separation between the metabolic profiles of RHS and FHS in the two principal components, signifying distinct differences in the metabolites they contain. These findings highlight the substantial impact of fermentation on the metabolic composition of hemp seeds.

Univariate analysis of UHPLC-Q-TOF/MS-based untargeted metabolomics

To examine the detailed differences in metabolite profiles between groups, the univariate analysis was employed and the results were visualized in volcano plots (Fig. 3b–d). In these plots, up-regulated metabolites are represented by orange dots, while down-regulated metabolites are depicted by blue dots. Specifically, we screened the 20 most important differential metabolites for further correlation analysis based on a comparison of q-value and fold change value between groups. These metabolites include Ala-Thr-Thr-Tyr, ethylparaben, indolelactic acid, citric acid, 2-furoic acid, L-phenylalanine, 9,12,13-trihydroxy-10,15-octadecadienoic acid, N-[(diphenylmethoxy)acetyl]-glycine, gluconic acid, Leu-Val-Thr-Leu-Ala, homovanillic acid, 2-ethoxyethyl acetate, tigecycline hydrate, 2,7-Anhydro-α-N-acetylneuraminic acid, daidzein, [(Coronen-1-yl) methylidene] hydrazine, prolylhydroxyproline, corchorifatty acid F, tuberonic acid glucoside, cannabisin B (Fig. 3b–d). The detailed fragment ions, retention time, and relative content of these compounds are listed in Table ​Table1.1. These 20 differential metabolites play a crucial role in revealing the metabolic changes that occurred during fermentation and may be key contributors to the anti-inflammatory effects of FHS.

Based on these results, the highest proportion of compounds in the FHS belonged to organic acids, alkaloids, and derivatives, followed by amino acids, and peptides. As fermentation progressed, the microbial cells underwent autolysis leading to release of cellular exudates, thus the organic acids, alkaloids, and derivatives compounds pronounced changes²⁴. This observation aligns with previous literature documenting notable alterations in organic acid abundance during the fermentation of hemp seed bran²⁵. Regarding amino acids and peptides, they were attributable to the hydrolysis of proteins in the raw material and bacterial autolysis²⁶. In addition, amines can be derived from amino acids through decarboxylation reactions, producing alkaloids that constitute approximately 20% of plant-based secondary metabolites²⁷. Alkaloids are renowned for their high biological activities and have gained importance in medicine. Thus, the changes observed in amino acids and the subsequent formation of alkaloid compounds during fermentation might account for the observed above phenomenon. In conclusion, fermentation can enhance the utilization value of the food matrix by bringing about changes in these differential metabolites. Furthermore, the alterations in flavor-related metabolites, such as amino acids and peptides during fermentation, hold crucial significance in the development of novel functional food products.

Correlations of differential metabolites with the anti-inflammatory activities

Spearman’s correlation analysis was conducted to investigate the relationship between differential metabolites in hemp seeds during fermentation and their contributions to anti-inflammatory ability. In this analysis, the 20 differential metabolites were selected as the variables and their correlated relationship is depicted in Fig. ​Fig.3e.3e. Among these 20 differential metabolites, 7 potential compounds were found to exhibit a significant positive association with anti-inflammation. These compounds primarily belong to organic acids and peptides. Specifically, the inhibitory activity of IL-6, IL-18, TNF-α, IL-1β, and NO production strongly and positively correlated with Ala-Thr-Thr-Tyr, indolelactic acid, citric acid, 2-furoic acid, gluconic acid, Leu-Val-Thr-Leu-Ala, homovanillic acid, most of which are classified as organic acid compounds (Fig. ​(Fig.3e).3e). Other studies have proved that organic acid compounds can be as major contributors to anti-inflammatory activity. These components have demonstrated their effectiveness in reducing LPS-induced inflammation, and their potential mechanism of action may involve the TLR4/IKK/NF-kB signaling pathway²⁸. Interestingly, our study showed that anti-inflammatory peptides emerged in hemp seeds after fermentation. These peptides likely bind with factors involved in inflammatory pathways, such as TLRs and NF-κB, thus reducing the inflammatory response²⁹. However, it is essential to note that peptides are unstable. Therefore, we need to conduct more detailed investigations to thoroughly validate the potency and molecular stability of these peptides present in FHS.

Confirmation of anti-inflammatory organic acids compounds in FHS

Based on the results obtained from the multivariate, univariate, and correlation analysis presented above, we employed the chromatographic method using HPLC-DAD to identify anti-inflammation-related organic acid compounds present in FHS. By quantifying the strongly anti-inflammation-related compounds identified in the correlation analysis, we found that indolelactic acid (IA) and homovanillic acid (HVA) displayed the most significant changes during the fermentation and exhibited the highest content after fermentation (Fig. 4a, b). In detail, Fig. 4a, b illustrated that the RHS group contained 7.6 µg/mg of IA and 9.2 µg/mg of HVA. However, the content of IA and HVA remarkably increased to 72.7 µg/mg and 57.9 µg/mg after fermentation, respectively.

The significant increase of IA and HVA in FHS is highly likely to be the reason for the enhanced anti-inflammatory activity of FHS. Thus we separately examined the anti-inflammatory activity of these two compounds to better understand their individual contributions. Table ​Table2.2. illustrated the half inhibition concentrations (IC₅₀) of IA and HVA. Both compounds exhibited excellent anti-inflammatory ability. IA exhibited higher inhibitory activity against IL-1β, IL-6, and NO, with IC₅₀ values of 9.11 ± 0.21 µg/mL, 10.67 ± 1.04 µg/mL, 8.74 ± 0.32 µg/mL, respectively. On the other hand, HVA displayed higher inhibitory activity against TNF-α and IL-18, with IC₅₀ values of 11.84 ± 1.02 µg/mL and 10.94 ± 0.31 µg/mL, respectively. These findings suggested that both IA and HVA contribute significantly to the potent anti-inflammatory effects observed in FHS. These findings closely resemble those documented in previous literature, highlighting the significant potential of IA and HVA in biological activities, including antioxidant and anti-inflammatory effects^30,31.

Changes in IA and HVA contents, and antiinflammatory capacity of FHS and RHS during digestion

In this study, hemp seeds exhibited notable anti-inflammatory properties following fermentation with L. plantarum, with IA and HVA identified as the primary metabolites responsible for this effect. Thus, IA and HVA act as pivotal detection indicators for assessing the bioavailability of FHS throughout the entire ingestion. We examine variations in their content and activity at each stage to evaluate FHS availability during digestion. As shown in Fig. ​Fig.5,5, both IA and HVA, along with their levels within FHS, performed relatively stable throughout the simulated digestive process. The exposure of IA and HVA in the simulated oral cavity did not compromise their stability, as the environment had no detrimental impact on their integrity (Fig. 5a, b). This alignment with our previous HPLC validation findings provides additional evidence of IA and HVA stability, even in the presence of α-amylase. During simulated gastric and intestinal digestion, we observed a minor reduction in IA content within FHS, although IA itself exhibited no degradation (Fig. ​(Fig.5a).5a). As a small molecule lipophilic tryptophan metabolite generated by gut microbiota³², IA lacks amide bonds, rendering it stable in simulated gastric and intestinal fluids and conducive to utilization. While reports indicate the accumulation of millimolar indole compounds in the gastrointestinal tract³³, studies specifically addressing the digestion of IA are limited. The slight decrease in IA content in FHS may be attributed to the involvement of tryptophan in the indole derivative pathway, which works together with enzymes to consume IA³². In comparison to IA, HVA within FHS showed relative stability in gastric juice but underwent a significant decrease in the alkaline environment of intestinal juice (Fig. ​(Fig.5b).5b). This could be attributed to HVA’s susceptibility to digestive enzymes and pH variations, which may result in degradation or biotransformation³⁴. An alternative explanation might involve interactions with other constituents of the FHS extract (e.g., amino acids and fatty acids), potentially leading to the formation of complexes that interfere with HVA quantification³⁴.

The fluctuation in anti-inflammatory activity corresponds to changes in the content of IA and HVA, as depicted in Fig. 5c–e. The activity of IA and HVA remains unchanged after salivary digestion, whereas a significant increase is observed following gastric and intestinal digestion (Fig. 5c–e). This could be attributed to the primary action sites of IA and HVA, which are predominantly located in the gastric and intestinal mucosa³⁵. Additionally, as illustrated in Fig. ​Fig.5b,5b, the increase in branching of the alkyl chain resulting from hydrolysis enhances the activity of HVA during the digestion process³⁵. Interestingly, the bioactivity of HVA appears to be independent of its chain length, with 1-methylated compounds demonstrating greater activity³⁵. Saliva, gastric, and intestinal samples of FHS exhibit notable anti-inflammatory activity, which significantly increases after gastric digestion but decreases somewhat after intestinal digestion. However, there is no significant difference compared to the original FHS, as indicated in Fig. 5c–e. The observed activity recovery could be related to the interaction of other components in FHS, such as fatty acids and phenolic compounds, which can be sensitively activated by the acidic and alkaline environment³⁴.

Samples preparation

Extraction

The extracting followed the published procedure³⁹, with minor adjustments. The RHS and freeze-dried FHS powder was extracted in 70% ethanol at a ratio of 1: 20 (W/V) for 24 h at room temperature in an electric shaker (RK-2D, DAIHAN scientific, Korea). Following extraction, the samples were centrifuged at 4000 × g for 10 min (Union 32 R plus, Hanil Science Industrial, Korea). The supernatant and precipitate were collected, respectively. The precipitate was then subjected to a second extraction by adding 70% ethanol in a ratio of 1:20 (W/V) at room temperature for 24 h. The supernatant and precipitate were collected again under the same centrifugal conditions as the first extraction. This process was repeated for a total of three times. The supernatant obtained from the three extractions and freeze-dried (Labconco Corporation, USA). The freeze-dried extract was stored at −20 °C until further analysis.

Cell culture

RAW 264.7 cells were grown in DMEM containing 10% FBS and 1% penicillin. Cells were maintained at 37 °C in a humidified environment with a 5% CO₂ incubator⁴⁰. Sub-cultured cells were treated with 0.05% Trypsin-EDTA (1X)⁴⁰.

Cytotoxicity assay

The cytotoxicity of the extract dissolved in serum-free media was evaluated using the EZ-Cytox assay kit with slight modifications following the protocol⁴¹. RAW 264.7 cells were seeded at a density of 1 × 10⁵ per well in a 96-well plate. Then the cells were treated with extracts of different concentrations and incubated for 24 h. Subsequently, 10 μL EZ-Cytox was added and further incubated for 2 h. A SpectraMax i3 plate reader (Molecular Devices Korea, LLC, Korea) was used to read absorbance at 450 nm. Cell viability was calculated according to the instructions provided with kit.

Anti-inflammatory assays

The effect on inhibiting inflammatory responses of macropahges was performed as previously reported with slightly modified¹³. The levels of TNF-α, IL-1β, IL-6, IL-18, and NO in the cell culture supernatant, were determined using ELISA kits as per the kit’s protocol. Briefly, the cells were seeded at a density of 1 × 10⁵ per well in 96-well plates and incubated for 12 h, followed by stimulation using 50 µL of 10 µg/mL LPS and incubated for 4 h. The 100 µL extract samples were then added to the wells and incubated for 24 h. As a positive control, 10 µg/mL corticosterone acetate dissolved in DMSO and then treated cells.

Immunoblotting analysis

The whole cell lysate was obtained by adding RIPA buffer to the cells, and the mixture was centrifuged (Micro 12, Hanil Science Industrial, Korea) at 7711 × g for 5 min. A modified lowry protein assay kit (Thermo Scientific, USA) was used to evaluate the protein content of the supernatant. The supernatant with uniform protein content was subsequently subjected to Sodium Dodecyl Sulphate Polyacrylamide Gel Electrophoresis (SDS-PAGE) according to the protocol provided by Bio-RAD, USA. The target protein expression was incubated with primary antibodies at a 1:1000 dilution overnight at 4 °C. Subsequently, secondary antibodies with a dilution of 1:3000 were applied and incubated for 1.5 h at room temperature. An enhanced chemiluminescence kit and a chemiluminescence detector (Fusion FX, Vilber Laurmate, France) were utilized to visualize the gel. The data was normalized to the expression levels of the housekeeping protein β-actin.

UHPLC-QTOF-MS identification

The metabolites were identified using UHPLC-QTOF-MS according to methodology with slight modifications⁴². Ultrahigh-pressure liquid chromatography (UHPLC) was equipped with an AD pump, AD autosampler, AD column oven, degasser, controller and photodiode array (PDA) detector (ExionLC), as well as a quadrupole time-of-flight mass spectrometer (QTOF-MS) (SCIEX ExionLC AD machine, MA, USA). Briefly, the extracts dissolved in 70% ethanol (5 mg/mL) and were filtered through 0.22 μm Millex syringe filters. Chromatographic separation was performed on Accucore column C18 (100 × 3 mm, Thermo Fisher Scientific, Waltham, MA, USA). The compounds were eluted using the mobile phase, including solvent A (water containing 0.1% formic acid) and solution B (100% methanol) at a 0.5 mL/min flow rate. The following linear gradient elution program was used: 0–4 min, 9–14% B; 4–5 min, 14–15% B; 5–7 min, 15% B; 7–9 min, 15–17% B; 9–10 min, 17–19% B; 10–11 min, 19% B; 11–13 min, 19–26% B; 13–14 min, 26–28% B; 14–15 min, 28–35% B; 15–17 min, 35–40% B; 17–18 min, 40–48% B; 18–19 min, 48–53% B; 19–23 min, 53–70% B; 23–24 min, 70–9% B; and 24–30 min, 9% B. The mass spectrometry (SCIEX, Framingham, MA, USA) was working under positive (ESI+) and negative (ESI−) electrospray ionization modes. The source temperature was maintained at 320 °C. The spray and tube lens voltage was controlled at 120 V (3.6 kV) and −120 V (2.7 kV) for the positive and negative ion modes, respectively. The sheath gas flow and units and an aux gas flow were controlled at 40 and 8 arb, respectively. The mass scan range was from 115 to 1000 m/z, and the scanning time was set for 1 s.

HPLC-DAD analysis

The procedure was conducted based on a previous study with modifications⁴³. Briefly, the standards and samples were dissolved in ddH₂O. All the standards and samples were prepared in 70% ethanol and then filtered through 0.22 μm Millex syringe filters. HPLC analysis was performed using Agilent 1260 series system equipped with a quaternary gradient pump and a DAD detector. Separation was achieved using a Phenomenex Gemini C18 column (4.6 × 250 mm, 5 μm) from Agilent, USA. The mobile phase (MP) A consisted of 50 mM ammonium formate in water (adjusted to pH 6.55 with formic acid). MPB consisted of 50 mM ammonium formate in methanol (pH 6.55). The gradient was set as follows: 0–10 min with 0% MPB; 10–20 min with 40% MPB; 20–30 min with 100% MPB. The flow rate was set at 1 mL/min, and the injection volume was set to 10 μL.

In-vitro simulated digestion and bioaccessibility

Bioaccessibility of bioactive compounds in Caco-2 cellular model

The bioaccessibility of the samples was performed according to the published protocol^36,37 with minor adjustments. After simulated digestion, intestinal samples from FHS and RHS were marked as FHS-G3 and RHS-G3 and were taken out for bioaccessibility analysis. In this study, the IA and HVA were confirmed as the primary anti-inflammatory compounds during digestion. Hence, the in vitro bioaccessibility (%) was calculated by the ratio of IA and HVA mass in the bioaccessible fraction to the total mass based on the Caco-2 model. Caco-2 cells were cultured at 37 °C in DMEM supplemented with 10% FBS and 1% penicillin-streptomycin. During the measured process, the cells at a density of 1 × 10⁶ cells/ml were seeded on 24-well Transwell plates, and incubated continuously for 14 days to form monolayers. Following washing with HBSS (pH 6.8), the monolayers were exposed to 2 mL of FHS-G3 and RHS-G3 for 2 h. After incubation, the upper liquid and the lower liquid were collected individually for HPLC analysis.

This work was supported by a grant from the fourth Brain Korea (BK) 21 Plus Project (Grant No.4299990913942) financed by the Korean Government, Republic of Korea.