Eruca sativa seeds were supplied by SAIS spa (Cesena, Italy; license: no. 12, 6/7/1973; authorization from Regione Emilia Romagna 43) and guaranteed by the producer for the quality and the genetic homogeneity of the product. Seeds were soaked for 15 min in 1% sodium hypochlorite and washed with tap water. Subsequently, the seeds were air dried at a temperature of 30°C for approximately 8–12 min, were ground and again air dried with forced ventilation. ESE was characterized by HPLC analysis.

For cell culture experiments, ESE was dissolved in dimethyl sulfoxide (DMSO) and diluted in the cell culture medium.

RAW 264.7, a murine macrophage cell line obtained from ‘Centro Substrati Cellulari, Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia’ (Brescia, Italy) were cultured in monolayer at 37°C in a moisturized atmosphere of 5% CO₂ and 95% air using DMEM-high glucose medium (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich; Merck KGaA). In order to induce an inflammatory state, cells were grown until 70–80% confluence was reached, and after they were incubated with 1 µg/ml LPS from Escherichia coli 0111:B4 (Sigma-Aldrich; Merck KGaA) for 24 h. Untreated cells were used as control. At the end of the treatment, cells were either fixed or harvested for other analyses, while the culture medium was collected to carry out experiments with NSC-34 cells. All the experiments were made in triplicates and repeated three independent times.

NSC-34 motor neurons, a murine cell line acquired from Cedarlane/CELLutions Biosystems Inc. (Burlington, ON, Canada), were cultured in DMEM-high glucose medium (Sigma-Aldrich; Merck KGaA) with the addiction of 10% FBS (Sigma-Aldrich; Merck KGaA), at 37°C in a moisturized atmosphere of 5% CO₂ and 95% air. We evaluated the cytotoxicity of ESE, exposing NSC-34 cells for 24 h to different concentrations of ESE. In particular, we tested the following concentrations of ESE: 0.1, 0.2, 0.3 and 0.4 µg/ml. To examine anti-inflammatory effects of ESE against LPS-induced inflammation, NSC-34 cells were pre-treated for 24 h with ESE. At the end of pre-treatment, medium was replaced with cell-free conditioned medium from LPS-stimulated RAW 264.7 macrophage, and incubated for 24 h. As controls, NSC-34 were incubated with the medium of unstimulated RAW 264.7. Cells treated with vehicle (<0.1% DMSO) or with the ESE alone were also included as controls. Then, motor neuronal cells were fixed for eosin/hematoxylin staining or immunocytochemistry analysis. All the experiments were done in triplicate and repeated for three independent times.

RAW 264.7 and NSC-34 morphological changes were evaluated by eosin and hematoxylin staining applying a standard protocol. Cells, grown on coverslips (10 mm; Thermo Fisher Scientific, Darmstadt, Germany), were fixed with 4% paraformaldehyde (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) for 15 min at room temperature. Cells were washed with 1X phosphate-buffered saline (PBS) and stained with hematoxylin Harris (Bio-Optica, Milan, Italy) for 1 min, rinsed with tap water, stained with eosin (Bio-Optica) for 5 min, and rinsed with distilled water. After, coverslips were dehydrated with ethanol series (Carlo Erba Reagents, Valde-Reuil, France; 50, 70, 80, 96, and 100%) and xylene (J.T. Baker, Deventer, The Netherlands). Subsequently, coverslips were mounted on microscope slides with mounting medium (Eukitt, Carlo Erba Reagents, Valde-Reuil, France) and allowed to dry. Microscopy was performed using a light microscope (Leica DM 2000 combined with Leica ICC50 HD camera) with the objective ×20 (for RAW 264.7 macrophages) or ×40 (for NSC-34 motoneurons). All images are representative of three independent experiments.

RAW 264.7 macrophages and NSC-34 motor neurons were grown on coverslips (10 mm; Thermo Fisher Scientific, Oberhausen, Germany) and, at the end of the treatments, they were fixed with 4% paraformaldehyde at room temperature for 15 min and after washed with PBS (pH 7.5). Then, cells were incubated with 3% hydrogen peroxide (H₂O₂) at room temperature for 15 min to suppress the endogenous peroxidase activity. After three washes with PBS, cells were blocked with horse serum +0.1% Triton X-100 for 20 min and incubated overnight at 4°C with primary antibodies. In particular, RAW 264.7 cells were incubated with the following antibodies against examined proteins: mouse monoclonal antibody against interleukin (IL)-1β (cat. no. 12242; 1:100; Cell Signaling Technology, Inc., Danvers, MA, USA), rabbit polyclonal antibody against Fas Ligand (FasL; cat. no. ab15285; 1:100; Abcam, Cambridge, UK), goat polyclonal antibody against P-selectin (cat. no. sc-6943; 1:50; Santa Cruz Biotechnology, Inc.) and rat monoclonal antibody against IL-10 (cat. no. sc-73309; 1:100; Santa Cruz Biotechnology, Inc.).

NSC-34 were incubated with the following primary antibodies against examined proteins: rabbit polyclonal antibody against tumor necrosis factor-α (TNF-α; cat. no. 3707; 1:250; Cell Signaling Technology, Inc.), rat monoclonal antibody against IL-10 (cat. no. sc-73309; 1:100; Santa Cruz Biotechnology, Inc.), rabbit polyclonal antibody against FasL (cat. no. ab15285; 1:100; Abcam), mouse monoclonal antibody against cyclooxygenase 2 (COX2; cat. no. sc-166475; 1:100; Santa Cruz Biotechnology, Inc.), rabbit monoclonal antibody against caspase 1 (cat. no. ab108362; 1:100; Abcam), rat monoclonal antibody against NLR family pyrin domain containing 3 (NLRP3; cat. no. MAB7578; 1:100; R&D Systems, Minneapolis, MN, USA), rabbit polyclonal antibody against IL-18 (cat. no. ab191152; 1:100; Abcam), mouse monoclonal antibody against IL-1β (cat. no. 12242; 1:100; Cell Signaling Technology, Inc.) and mouse monoclonal antibody against Toll-like receptor 4 (TLR4; cat. no. ab22048; 1:100; Abcam).

After, cells were washed with PBS and incubated with biotinylated secondary antibody (1:200; Vector Laboratories, Inc., Burlingame, CA, USA) and streptavidin AB Complex-HRP (ABC-kit from Dako, Glostrup, Denmark). The immunostaining was developed with the DAB peroxidase substrate kit (Vector Laboratories, DBA Italia S.r.l., Milan, Italy; brown color; positive staining) and counterstaining with nuclear fast red (Vector Laboratories, DBA Italia S.r.l.; pink background; negative staining).

The immunocytochemical assays were repeated three times and each experimental group was plated in duplicate, for a total of 6 coverslips for each antibody. With the aim of calculating the percentage of positive cells stained, the images were captured using a light microscopy (LEICA DM 2000 combined with LEICA ICC50 HD camera) with an objective ×40 and the densitometric analysis was carried out using the software LEICA Application Suite ver. 4.2.0. Quantitative analysis was performed on 6 coverslips by covering approximately 90% of the total area.

At the end of the treatment, RAW 264.7 cells were harvested and lysed using buffer A [320 mM sucrose, 10 mM, 1 mM EGTA, 2 mM EDTA, 5 mM NaN3, 50 mM NaF, β-mercaptoethanol, and protease/phosphatase inhibitor cocktail (Roche Molecular Diagnostics, Branchburg, NJ, USA)] in ice for 15 min, and centrifuged at 1,000 × g for 10 min at 4°C. The supernatant was collected as cytosolic extract. The obtained pellet was lysed using buffer B [150 mM NaCl, 10 mM Tris-HCl (pH 7.4), 1 mM EGTA, 1 mM EDTA, Triton X-100, and protease/phosphatase inhibitor cocktail (Roche Molecular Diagnostics)] in ice for 15 min and centrifuged at 15,000 × g for 30 min at 4°C. The supernatant was harvested as nuclear extract. Protein concentrations were analyzed using the Bradford assay (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Twenty µg of proteins were heated for 5 min at 95°C and resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and after transferred onto a PVDF membrane (Immobilon-P, Millipore, DBA Italia S.r.l., Milan, Italy).

After, membranes were blocked with 5% skim milk in PBS for 1 h at room temperature and incubated overnight at 4°C with the following primary antibodies: Rabbit polyclonal antibody against TNF-α (cat. no. 3707; 1:500; Cell Signaling Technology, Inc.) and rat monoclonal antibody against IL-10 (cat. no. sc-73309; 1:500; Santa Cruz Biotechnology, Inc.). After the membranes were washed with PBS 1X and incubated with horse radish peroxidase (HRP)-conjugated goat anti-rabbit or chicken anti-rat IgG secondary antibody (cat. nos. sc-2004 and sc-2956, respectively; 1:2,000; Santa Cruz Biotechnology, Inc.) for 1 h at room temperature. In order to evaluate if blots were loaded with equal amounts of proteins, membranes were incubated with antibody for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) HRP Conjugated (cat. no. 3683; 1:1,000; Cell Signaling Technology, Inc.). The relative expression of protein bands was analyzed using an enhanced chemiluminescence system (Luminata Western HRP Substrates; Millipore, DBA Italia S.r.l.) and protein bands were acquired with ChemiDoc™ MP System (Bio-Rad Laboratories, Inc.) and quantified using the computer program ImageJ software. All blots are representative of three independent experiments.

Statistical analysis was performed using GraphPad Prism version 6.0 computer software program (GraphPad Software, Inc., La Jolla, CA, USA). The data were statistically analyzed by Student's t-test for the comparison between 2 groups or by one-way ANOVA test and Bonferroni post hoc test for multiple comparisons. P<0.05 was considered to indicate a statistically significant difference. Results are reported as mean ± SEM of N experiments.

ESE was chemically characterized by HPLC. Analysis showed that glucoerucin was the GLS present in higher quantity in ESE, and in particular the extract contained 46.36 mg glucoerucin/g of ESE. Moreover, the ESE contained 0.16 mg/g dry weight of ascorbic acid and 1.99 mg/g dry weight of total flavonoids.

At first we induced inflammation in RAW 264.7 cells, exposing them to LPS. Eosin and hematoxylin staining of LPS stimulated RAW 264.7 showed morphological changes, such as an increase in cell size and the extension of lamellipodia and filopodia (Fig. 1). We observed a positive staining for FasL in LPS treated cells compared to controls (Fig. 1), indicating the induction of apoptosis in LPS-treated macrophages. Furthermore, RAW 264.7 macrophages exposed to LPS presented a positive staining for P-selectin (Fig. 1).

LPS-induced inflammation was associated with increased protein levels of the pro-inflammatory cytokines TNF-α and IL-1β (Fig. 2). Instead, IL-10 levels, a well known anti-inflammatory cytokine, decreased in LPS stimulated macrophages compared to control cells (Fig. 2).

After confirming the presence of inflammation in macrophages, we evaluated whether ESE was able to counteract the inflammation induced in NSC-34 by the medium of LPS-treated RAW 264.7. At first we assessed whether the ESE caused cytotoxicity or morphological changes in NSC-34, using the concentrations of the extract 0.1, 0.2, 0.3 and 0.4 µg/ml. Our results showed no cytotoxicity for the concentrations 0.1, 0.2 and 0.3 µg/ml of ESE, while the highest concentration showed a moderate cytotoxicity, associated with the presence of cell debris (Fig. 3). Given that the extract was dissolved in DMSO, we included in our analysis also NSC-34 motor neurons incubated with a concentration of DMSO comparable to that used for the highest concentration of ESE, but no cytotoxicity was observed (Fig. 3).

We evaluated the morphological changes induced in NSC-34 by the medium of LPS-stimulated RAW 264.7. By eosin/hematoxylin staining, we observed that NSC-34 motor neurons exposed to the medium of LPS-treated RAW 264.7 showed cell death and degeneration. However, only the pre-treatment with ESE 0.3 µg/ml was able to prevent LPS-induced morphological changes, preserving the neural morphology that was similar to that of control cells (Fig. 4A). Instead, lower concentrations of the extract were not able to counteract LPS induced effects (Fig. 4A), while we did not evaluate the concentration 0.4 µg/ml of ESE, given its moderate cytotoxicity. For this reason we decided to continue the experiments with the concentration 0.3 µg/ml of ESE.

The treatment of NSC-34 motor neurons with the medium of LPS-stimulated RAW 264.7, induced apoptosis associated with FasL positive staining (Fig. 4B). However, our results showed that the pre-treatment with ESE was able to counteract apoptosis, as demonstrated by FasL negative staining (Fig. 4B and C). NSC-34 control cells, and those incubated with DMSO or with ESE alone did not express FasL (data not shown).

We observed TLR4 positive staining in NSC-34 motor neurons exposed to the culture medium of LPS-stimulated macrophages (Fig. 5). Moreover, we evaluated also COX2 expression, and we found that NSC-34 motor neurons stimulated with the medium of LPS-treated RAW 264.7 macrophages showed a positive staining for COX2 (Fig. 5). These results suggested that inflammation depends on both TLR4 and COX2 pathway activation. Interestingly, NSC-34 pre-treated with ESE did not express TLR4 and COX2 proteins. NSC-34 motor neurons used as controls or incubated with DMSO, or with the extract alone did not show expression of neither TLR4 nor COX2 (data not shown), indicating the absence of inflammation.

Moreover, we evaluated cytokine levels and observed that NSC-34 cells exposed to the culture medium of LPS-stimulated RAW 264.7 did not show the expression of IL-10, but they showed a positive staining for the pro-inflammatory cytokine TNF-α (Fig. 6). Pre-treatment with ESE recovered the expression of IL-10, while the expression of the pro-inflammatory cytokine was abolished. Cells incubated with DMSO or with the seed extract alone expressed IL-10, but not TNF-α (data not shown).

Inflammation may be also mediated by a protein complex known as inflammasome, a cytoplasmic protein complex that mediates pro-inflammatory responses. It is formed by a pattern recognition receptor, acting like sensor of damage, such as of pathogen associated molecular patterns, that when activated, induced the production of active caspase 1 (16,17). The NLRP3 inflammasome, formed by NLRP3 that belongs to the NOD-like receptor (NLR) family, is the most studied. When the inflammasome is activated, the cleavage of pro-caspase 1 is induced to form the active caspase 1, that in turn converts the inactive IL-18 and IL-1β into the active forms, triggering inflammation (16,17). We found NLRP3 and caspase 1 positive staining in NSC-34 stimulated with the LPS-conditioned medium of RAW 264.7 macrophages indicating the activation of inflammasome. According to the activation of caspase 1, we observed the expression of the proinflammatory cytokines IL-18 and IL-1β (Fig. 7). However, the pre-treatment of NSC-34 with ESE was able to abolish the expression of NLRP3 and caspase 1 and, consequently, the production of IL-18 and IL-1β (Fig. 7). NSC-34 control cells, or incubated with DMSO or with the extract alone, did not express inflammasome proteins (data not shown).