Dietary Oleocanthal Supplementation Prevents Inflammation and Oxidative Stress in Collagen-Induced Arthritis in Mice

Oleocanthal (OLE), a characteristic and exclusive secoiridoid of Oleoaceae family, is mainly found in extra virgin olive oil (EVOO). Previous studies have reported its antioxidant, anti-inflammatory, antimicrobial, anticancer and neuroprotective effects. Since the pathogenesis of rheumatoid arthritis (RA) involves inflammatory and oxidative components, this study was designed to evaluate the preventive role of dietary OLE-supplemented effects in collagen-induced arthritis (CIA) murine model. Animals were fed with a preventive OLE-enriched dietary during 6 weeks previous to CIA induction and until the end of experiment time. At day 43 after first immunization, mice were sacrificed: blood was recollected and paws were histological and biochemically processed. Dietary OLE prevented bone, joint and cartilage rheumatic affections induced by collagen. Levels of circulatory matrix metalloproteinase (MMP)-3 and pro-inflammatory cytokines (IL-6, IL-1β, TNF-α, IL-17, IFN-γ) were significantly decreased in secoiridoid fed animals. Besides, dietary OLE was able to diminish COX-2, mPGES-1 and iNOS protein expressions and, also, PGE2 levels. The mechanisms underlying these protective effects could be related to Nrf-2/HO-1 axis activation and the inhibition of relevant signaling pathways including JAK-STAT, MAPKs and NF-κB, thus controlling the production of inflammatory and oxidative mediators. Overall, our results exhibit preliminary evidences about OLE, as a novel dietary tool for the prevention of autoimmune and inflammatory disorders, such as RA.


Introduction
Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease characterized by severe synovitis leading to progressive joints destruction accompanied by systemic inflammation. RA is often considered an autoimmune condition expressing certain characteristic autoantibodies such as rheumatoid factors and antibodies to citrullinated proteins [1].
The etiology of RA involves genetics and environmental factors association. Nevertheless, geography, socioeconomic status, diet/nutrients, alcohol, smoking, and hostmicrobiome also contribute to the risk of developing RA [2,3].
Although the pathogenesis of RA is multifactorial, various inflammatory pathways lead to an altered immune function. Particularly, there are massive immune cells infiltration in the synovial membrane, including CD4 + T, B and natural killer cells, in addition to macrophages, dendritic cells, neutrophils and mast cells. All of them induce aberrant cytokines production such as tumor necrosis factor alpha (TNF)-α, interferon (IFN)-γ and interleukins (IL) (IL-6, IL-8, IL-12/23, IL-17 or IL-18), which perpetuate and ease joint inflammation, rising matrix metalloproteases (MMPs), and consequently activating osteoclasts at joint site. As a result, pannus invades subchondral tissue, destroying cartilage and generating ankyloses [4]. Furthermore, RA includes extra-articular manifestations, Essentially, OLE exerts several protective functions in neurodegeneration through its ability to modulate oxidative stress and apoptosis in neuronal cells [23], suggesting it as a potential adjunct to drugs against Alzheimer and Parkinson diseases [17]. Besides, a large number of studies supports its anticancer potential tested on different in vitro and in vivo cancer assays [17,24,25]. Complementary, other bioactive properties in individual health have been attributed to this secoiridoid as cardio-protection and antimicrobial actions [13,18,26].
Indeed, OLE has shown potent anti-inflammatory effects inhibiting the activities of COX-1 and COX-2, and also 5-lipoxygenase (5-LOX), the rate-determining enzymes for the synthesis of prostaglandins (PGs) and pro-inflammatory leukotrienes, respectively. Surprisingly, OLE exerted better anti-inflammatory activity than the non-steroidal antiinflammatory drug, (NSAID), ibuprofen, at the same concentration. These results promoted later studies with OLE, which were useful to assign its health-benefiting properties, suggesting OLE as a possible key supplement to several inflammatory diseases and improving the side-effects of classic pharmacological therapy [18,[27][28][29]. Recently, our research group has demonstrated the potential anti-inflammatory and antioxidant modulation of OLE in lipopolysaccharide (LPS)-stimulated murine peritoneal macrophages which represents a well-established model for validate anti-inflammatory drugs [30].
Although the studies about the antioxidant activity of OLE are very limited, it has been demonstrated that OLE could inhibit nicotinamide adenine dinucleotide phosphate oxidase (NOX) in isolated human monocytes, and diminish intracellular ROS levels in SH-SY5Y cells [13].
In terms of rheumatic diseases, OLE has shown to ameliorate osteoarthritis and RA in vitro. However, to date, there are not data about the possible beneficial effects of OLE for an in vivo model of RA. Thus, this study was designed to evaluate an OLE-enriched diet preventive effects on collagen-induced arthritis (CIA) murine model. For this purpose, we analyzed macroscopic and histological damage, production of inflammatory mediators and, explored possible signaling pathways involved.

Chemicals
OLE was isolated from EVOO Cornicabra cultivar (1000 g). Oil was mixed with acetonitrile (400 mL), the mixture was shaken vigorously in a separating funnel and the two phases were allowed to separate. The process was repeated with extra acetonitrile (2 × 400 mL) to ensure complete extraction of phenols.
The combined polar phases were concentrated to an oily residue using a rotary evaporator at residue pressure. The residue was washed with cyclohexane (2 × 10 mL) to eliminate residual triglycerides, and purified by column chromatography using silica gel 60 (40-63 mm) (Merck ® , Darmstadt, Germany), as the stationery phase and a gradient EtOAc/cyclohexane (0:1 → 1:2) as the eluent.
The purity of the extracted OLE was based on the 1 H and 13 C NMR spectra and HPLC analyses. NMR spectra were registered in CDCl3 in a Bruker Avance-300 spectrometer; 1   The gradient, at a flow rate of 1.0 mL/min, was as follows: 2 min from 100% to 95% of A; 8 min from 95% to 75% of A; 10 min from 75% to 60% of A; 10 min from 60% to 50% of A. The analyses showed that the isolated OLE had purity higher than 95% (data are shown in Supplementary Materials).

Animals and Diet
Thirty-four three-weeks-old male DBA-1/j mice (Janvier ® , Le Genest St Isle, France) were housed under standard conditions (24-25 • C, humidity 70-75%, lighting regimen 12L/12D) in our Animal Laboratory Center. All mice were acclimated for 4 weeks prior to the initiation of experiments, feeding with standard diet (SD) and water ad libitum. At 4weeks-aged animals were divided into three experimental groups: (1) Naïve group received a SD (n = 10), (2) CIA group received a SD (SD-CIA) (n = 12) and, (3) CIA-OLE group received a SD supplemented with OLE 0.025% (dietary enrichment percentage selection based on our previous reports [31,32]) (n = 12). All diets were elaborated according to the basis of the American Institute of Nutrition (AIN) standard reference diet, mixing OLE or no, and stored them at -80 • C. Experimental groups were fed with fresh diet daily during six weeks, previously to first immunization and until the day of sacrifice. Animal care and procedures agreed with protocols approved by Animal Ethics Committee of the Universidad de Sevilla (ethical approval number 23/07/2018/119), and recommendations of European Union regarding animal experimentation (Directive of the European Counsel 2012/707/EU).

Collagen Type II Induction of Arthritis Disease
The first immunization of disease was injected at the basis of the tail when DBA-1/j mice were ten-weeks-old (Day O). Bovine type II collagen (CII) was diluted in acetic acid (2 mg/mL) (MD Bioproducts ® , Zurich, Switzerland) and was emulsified with an equal volume of complete Freund's adjuvant (2 mg/mL Mycobacterium tuberculosis, strain H37Ra; Difco ® , Detroit, MI, USA). On day 21, a second intraperitoneal immunization was injected with 100 mg of CII dissolved in phosphate buffered saline (PBS).

Clinical Assessment of Joints
The development of arthritis was considered and evaluates daily when mice exhibited redness and/or swelling affections in digits, joints or entire paws. To evaluate the arthritic index for each paw of mice, a macroscopic score system was used: 0, no signs; 1, mild inflammation of joint; 1.5, marked inflammation of joint and digits; 2, severe inflammation of paw. Scoring was performed by two observers-blind.

Histological Analysis
On day 43, mice were sacrificed and knees joints were removed and fixed in 4% formalin. Paws were decalcified in 10% EDTA for 30 days and, then, were dehydrated and embedded in paraffin. Samples sectioned (7 mm) were stained with hematoxylin and eosin (H&E) or tartrate-resistant acid phosphate (TRAP) using naphtol-AS-MX phosphate (Sigma ® , St. Louis, MO, USA) and fat red violet LB salt (Sigma ® , St. Louis, MO, USA), to perform histological analysis.

Data Analysis
All values are expressed as arithmetic means ± standard error (SEM) in figures and text. Results were evaluated using Graph Pad Prism version 5.01 software (San Diego, CA, USA), analyzing the statistical significance by one-way analysis of variance (ANOVA), followed by Tukey's multiple comparisons test. p-Values < 0.05 were considered statistically significant. Figures from densitometry experiments are representatives of different experiments performing on a different day.

Effects of OLE Supplemented Diet on Collagen-Induced Arthritis Model
The similarity of CIA and RA reside in the loss of tolerance and self-antigens production against collagen, one of the main self-antigens also observed in human RA. Immunization of mice with an emulsion of CII on complete Freund's adjuvant exhibits an erosive polyarthritis with an autoimmune response against cartilage mediated by both auto-reactive B and T cells. Consequently, these pathogenic antibodies detect their endogenous antigen in the joint triggering a local inflammatory response, synovial inflammatory cell infiltration, hyperplasia, cartilage destruction and bone erosion [34].
The model requires at least 6-8 weeks for the appreciation of clinical sings consistently. Indeed, the sings of arthritis were evaluated since day 29. We observed a progressive development of clinical symptoms and swelling from day 33 in CIA control mice fed with a normal standard diet (SD-CIA) ( Figure 1A).
On the contrary, animals that were fed with OLE showed a decrease in arthritis severity, reducing footpad thickness and inflammation of forelegs and hind legs in comparison with SD-CIA group. Representative images of hind paws from different experimental animal groups supported these results. In effect, OLE dietary not only retarded the development, but also exhibited an effective therapeutic activity on disease onset ( Figure 1B). progressive development of clinical symptoms and swelling from day 33 in CIA control mice fed with a normal standard diet (SD-CIA) ( Figure 1A).
On the contrary, animals that were fed with OLE showed a decrease in arthritis severity, reducing footpad thickness and inflammation of forelegs and hind legs in comparison with SD-CIA group. Representative images of hind paws from different experimental animal groups supported these results. In effect, OLE dietary not only retarded the development, but also exhibited an effective therapeutic activity on disease onset ( Figure 1B). As shown in Figure 2, H&E staining revealed infiltration of inflammatory cells into articular tissues, exudation into the synovial tissue, synovial hyperplasia and cartilage and bone damage in SD-CIA group, compared to Naïve group. However, the histological features of joints from mice fed with OLE resembled Naïve control (Figure 2A-C). TRAP, is considered to be just a histochemical marker of osteoclasts and is expressed by osteoclasts, macrophages, dendritic cells and a number of other cell types [35]. Consequently, TRAP staining showed larger quantities of osteoclasts and bone erosion in joint sections from CIA mice whereas joint sections from mice fed with OLE diet presented a significant reduction of TRAP-positive osteoclasts ( Figure 2D-F). As shown in Figure 2, H&E staining revealed infiltration of inflammatory cells into articular tissues, exudation into the synovial tissue, synovial hyperplasia and cartilage and bone damage in SD-CIA group, compared to Naïve group. However, the histological features of joints from mice fed with OLE resembled Naïve control (Figure 2A-C). TRAP, is considered to be just a histochemical marker of osteoclasts and is expressed by osteoclasts, macrophages, dendritic cells and a number of other cell types [35]. Consequently, TRAP staining showed larger quantities of osteoclasts and bone erosion in joint sections from CIA mice whereas joint sections from mice fed with OLE diet presented a significant reduction of TRAP-positive osteoclasts ( Figure 2D-F).

Effects of Dietary OLE on MMP-3 and Joint Inflammatory Biomarkers Levels
It has been reported that pro-inflammatory cytokines as TNF-α, IL-1β, IFN-γ, IL-6, or IL-17 play a critical role on the pathogenesis of RA [36]. Besides, pro-inflammatory cytokines, including TNF-α, IL-1β and IL-6, may activate MMP genes through the binding of several different transcription factors [37]. In order to address whether OLE-supplemented diet could regulate these joint inflammatory biomarkers, we measured cytokine levels in paw homogenates and MMP-3 in serum. As shown in Figure 3A, IL-1β, IFN-γ, TNF-α, IL-6 and IL-17 levels were markedly increased in paw homogenates from SD-CIA animals in comparison with sham group suggesting its relationship with the synovial tissue inflammation (p < 0.001 vs. Naïve). Nevertheless, our data indicated a significant reduction of pro-inflammatory cytokines production in paw homogenates from animals fed with OLE-diet (IL-1β: p < 0.01; IFN-γ: p < 0.001; TNF-α: p < 0.05; IL-6: p < 0.001; IL-17: p < 0.001 vs. CIA) ( Figure 3A).
In addition, circulating levels of MMP-3 were raised in SD-CIA animals, in parallel to the disease severity (p < 0.01 vs. Naïve) ( Figure 3B). On the contrary, a significant reduction of MMP-3 levels was observed in OLE-enriched diet fed animals (p < 0.01 vs. CIA). In addition, circulating levels of MMP-3 were raised in SD-CIA animals, in to the disease severity (p < 0.01 vs. Naïve) ( Figure 3B). On the contrary, a signif duction of MMP-3 levels was observed in OLE-enriched diet fed animals (p < 0.01 v

Effects of Dietary OLE on COX-2 Protein, mPGES-1 Expression and PGE2 Product
COX-2 and mPGES-1 expressions and levels of PGE2 were determined by blot and ELISA, respectively, in paw homogenates. Additionally, COX-2 express studied by immunohistochemistry in joints sections (Figures 4 and 5).

Effects of Dietary OLE on COX-2 Protein, mPGES-1 Expression and PGE 2 Production
COX-2 and mPGES-1 expressions and levels of PGE 2 were determined by Western blot and ELISA, respectively, in paw homogenates. Additionally, COX-2 expression was studied by immunohistochemistry in joints sections (Figures 4 and 5).
Arthritic control animal group showed an overexpression of both pro-inflammatory enzymes (COX-2: p < 0.01; mPGES-1: p < 0.001 vs. Naïve) in parallel with an increase of PGE 2 levels (p < 0.01 vs. Naïve). On the contrary, dietary OLE was able to diminish the protein expression of both proteins (COX-2: p < 0.01; mPGES-1: p < 0.001 vs. CIA) and PGE 2 levels (p < 0.01 vs. CIA) in CIA-induced mice ( Figure 4).  Similar results were found after the immunohistochemical analysis. We could observe overexpression of COX-2 positive cells in SD-CIA control group, whereas OLE experimental diet reduced remarkably the immunoreactivity for this pro-inflammatory enzyme ( Figure 5).   Similar results were found after the immunohistochemical analysis. We could observe overexpression of COX-2 positive cells in SD-CIA control group, whereas OLE experimental diet reduced remarkably the immunoreactivity for this pro-inflammatory enzyme ( Figure 5).  Similar results were found after the immunohistochemical analysis. We could observe overexpression of COX-2 positive cells in SD-CIA control group, whereas OLE experimental diet reduced remarkably the immunoreactivity for this pro-inflammatory enzyme ( Figure 5).

OLE Experimental Diet Attenuated iNOS Overexpression
iNOS is considered a critical oxidative stress marker involved in nitric oxide (NO) production, a free radical inductor of oxidation [8]. Down-regulation of iNOS expression was observed in hind paw of CIA mice group submitted to treatment with OLE-supplemented diet when compared with CIA mice received SD diet (p < 0.001 vs. CIA), as shown in Figure 6.

OLE Experimental Diet Attenuated iNOS Overexpression
iNOS is considered a critical oxidative stress marker involved in production, a free radical inductor of oxidation [8]. Down-regulation o was observed in hind paw of CIA mice group submitted to treatment mented diet when compared with CIA mice received SD diet (p < 0.001 in Figure 6.

Effects of Dietary OLE on pSTAT-3 Protein Expression
STAT-3 has been described as a critical transcription factor involv esis of RA and is activated by upstream cytokines, such as IL-6 and IL [38]. We evaluated pSTAT-3 protein expression by Western blot from nates. Statistical analysis revealed that pSTAT-3 was overexpressed i pared with Naïve group (p < 0.01 vs. Naïve) whereas, nutritional ther nificantly suppressed STAT-3 phosphorylation in arthritic CIA mice (Figure 7).

Effects of Dietary OLE on pSTAT-3 Protein Expression
STAT-3 has been described as a critical transcription factor involved in the pathogenesis of RA and is activated by upstream cytokines, such as IL-6 and IL-17, among others [38]. We evaluated pSTAT-3 protein expression by Western blot from hind paw homogenates. Statistical analysis revealed that pSTAT-3 was overexpressed in CIA group compared with Naïve group (p < 0.01 vs. Naïve) whereas, nutritional therapy with OLE significantly suppressed STAT-3 phosphorylation in arthritic CIA mice (p < 0.01 vs. CIA) (Figure 7).

OLE Experimental Diet Attenuated iNOS Overexpression
iNOS is considered a critical oxidative stress marker involved in nitric oxid production, a free radical inductor of oxidation [8]. Down-regulation of iNOS exp was observed in hind paw of CIA mice group submitted to treatment with OLE-s mented diet when compared with CIA mice received SD diet (p < 0.001 vs. CIA), as in Figure 6.

Effects of Dietary OLE on pSTAT-3 Protein Expression
STAT-3 has been described as a critical transcription factor involved in the pat esis of RA and is activated by upstream cytokines, such as IL-6 and IL-17, among [38]. We evaluated pSTAT-3 protein expression by Western blot from hind paw ho nates. Statistical analysis revealed that pSTAT-3 was overexpressed in CIA grou pared with Naïve group (p < 0.01 vs. Naïve) whereas, nutritional therapy with O nificantly suppressed STAT-3 phosphorylation in arthritic CIA mice (p < 0.01 v (Figure 7).

Dietary OLE Induces Nrf-2/HO-1 Antioxidant Pathway Activation
The expression of the proteins HO-1 and Nrf-2 were also evaluated in paw homogenates by Western blotting. Balance of Nrf-2 maintains the regulation of oxidative stress, inflammation, immune response, and cartilage and bone metabolism through activation of antioxidant enzymes, as HO-1. Thus, Nrf-2 regulates redox status and plays key roles in cellular defense by enhancing the removal of ROS [10]. As shows Figure 7, CIA mice fed with SD diet presented a reduced expression of both proteins, in comparison with Naïve control group (p < 0.05 vs. Naïve). However, a significant overexpression of both proteins, Nrf-2 and HO-1, were observed in paw homogenates from animals fed with OLE diet (p < 0.001 and p < 0.05 vs. CIA, respectively) ( Figure 8).
The expression of the proteins HO-1 and Nrf-2 wer nates by Western blotting. Balance of Nrf-2 maintains t inflammation, immune response, and cartilage and bon of antioxidant enzymes, as HO-1. Thus, Nrf-2 regulates in cellular defense by enhancing the removal of ROS [1 fed with SD diet presented a reduced expression of bo Naïve control group (p < 0.05 vs. Naïve). However, a si proteins, Nrf-2 and HO-1, were observed in paw homoge diet (p < 0.001 and p < 0.05 vs. CIA, respectively) ( Figure   Figure 8. OLE dietary treatment induced up-regulation of Nrfpaws homogenates. Densitometry analysis was normalized wi ing). Western blot images are representative of separate experi group, non-arthritic control animals fed with SD; CIA, control with SD; OLE induced arthritic animals fed with OLE enriched sented mean ± S.E.M. (n = 8), + p < 0.05 vs. Naïve control group group.

OLE Diet Modulates MAPKs Signaling Pathway
It has been reported an important role of MAPKs i pro-inflammatory gene expression which initiates infla present study, we measured phosphorylation, and there JNK MAPKs. Our data showed that phosphorylation of

OLE Diet Modulates MAPKs Signaling Pathway
It has been reported an important role of MAPKs in the RA pathogenesis, inducing pro-inflammatory gene expression which initiates inflammatory responses [39]. In the present study, we measured phosphorylation, and therefore, activation of p38, ERK and JNK MAPKs. Our data showed that phosphorylation of JNK, p38 and ERK proteins were increased significantly in cytosolic extracts from SD-CIA mice paw homogenates, compared with Naïve animals (p < 0.05 vs. Naïve). Again, OLE diet showed its effectiveness along the experimental period. In fact, the protein expressions of all phosphorylated MAPKs proteins, p-JNK, p-p38 and p-ERK, were significantly improved after dietary OLE treatment (p < 0.05 vs. CIA) (Figure 9). oxidants 2021, 10, x FOR PEER REVIEW Figure 9. OLE diet modulated MAPKs signaling pathways in hind paws homogenates ylated p38, JNK and ERK was analyzed with Western blot and normalized with respec non-phosphorylated total proteins. Western blot images are representative of separate ments with similar results. Naïve group, non-arthritic control animals fed with SD; CIA group induced arthritic animals fed with SD; OLE induced arthritic animals fed with O riched diet (0.025% w/w). Values represented mean ± S.E.M. (n = 8), + p < 0.05 vs. Naïve thritic group; * p < 0.05 vs. CIA arthritic group.

Effects of Dietary OLE on NF-κB Signaling Pathway
We analyzed the preventive effects of OLE supplemented diet on IκB-α d and nuclear translocation of the subunits p65 and p50 from paw homogenates observed that in SD-CIA animals' group, IκB-α expression was decreased, rev radation, and therefore allowing NF-κB translocation into the nucleus (p < 0.05 Contrary, IκB-α was overexpressed in cytoplasmic extracts from mice fed wit so there was not degradation (p < 0.05 vs. CIA).
These results were corroborated, after analysis of both p65 and p50 subu cleus extracts from paw homogenates. As shown Figure 10, the nuclear p65 an tein expressions were significantly increased in CIA group (p50: p < 0.01; p65: p Naïve); however, dietary OLE treatment could prevent the CIA-induced nucl cation of both subunits in comparison with arthritic animals fed with SD (p5 p65: p < 0.001 vs. CIA), avoiding the NF-κB-mediated transcriptional activat 10).  . OLE diet modulated MAPKs signaling pathways in hind paws homogenates. Phosphorylated p38, JNK and ERK was analyzed with Western blot and normalized with respect to specific non-phosphorylated total proteins. Western blot images are representative of separate experiments with similar results. Naïve group, non-arthritic control animals fed with SD; CIA, control group induced arthritic animals fed with SD; OLE induced arthritic animals fed with OLE enriched diet (0.025% w/w). Values represented mean ± S.E.M. (n = 8), + p < 0.05 vs. Naïve non-arthritic group; * p < 0.05 vs. CIA arthritic group.

Effects of Dietary OLE on NF-κB Signaling Pathway
We analyzed the preventive effects of OLE supplemented diet on IκB-α degradation and nuclear translocation of the subunits p65 and p50 from paw homogenates. Firstly, we observed that in SD-CIA animals' group, IκB-α expression was decreased, revealing degradation, and therefore allowing NF-κB translocation into the nucleus (p < 0.05 vs. Naïve). Contrary, IκB-α was overexpressed in cytoplasmic extracts from mice fed with OLE diet, so there was not degradation (p < 0.05 vs. CIA).
These results were corroborated, after analysis of both p65 and p50 subunits in nucleus extracts from paw homogenates. As shown Figure 10, the nuclear p65 and p50 protein expressions were significantly increased in CIA group (p50: p < 0.01; p65: p < 0.001 vs. Naïve); however, dietary OLE treatment could prevent the CIA-induced nuclear translocation of both subunits in comparison with arthritic animals fed with SD (p50: p < 0.05; p65: p < 0.001 vs. CIA), avoiding the NF-κB-mediated transcriptional activation ( Figure 10). ts 2021, 10, x FOR PEER REVIEW Figure 10. OLE diet-enriched controlled NF-κB signaling pathway p and p65 and p50 subunits migration to the nucleus. The expression try analysis and normalized with respect to β-actin (house-keeping resentative of separate experiments with similar results. Naïve grou mals fed with SD; CIA, control group induced arthritic animals fed animals fed with OLE enriched diet (0.025% w/w). Values represent 0.05; ++ p < 0.01; +++ p < 0.001 vs. Naïve group; * p < 0.05; *** p < 0.001 v

Discussion
Our study has revealed, for the first time, a preventive the arthritic process through a CIA murine model. This is a evaluate important RA pathogenic mechanisms and novel an described in 1977 and is characterized by the production of C feature that has also been described in RA.
In RA or CIA, the normal organization of synovium app ing infiltrated with T and B cells, macrophages and neutrop hyperplasic membrane of synoviocytes that invades and d Conversely, dietary OLE supplementation could control the a ering infiltration of immune cells, an appreciable reduction plasia, cartilage degradation and bone erosion compared wit The development and progression of RA are closely rela

Discussion
Our study has revealed, for the first time, a preventive role of OLE-enriched diet on the arthritic process through a CIA murine model. This is a useful model to study and evaluate important RA pathogenic mechanisms and novel antiarthritic drugs. It was first described in 1977 and is characterized by the production of CII-specific antibodies [40], a feature that has also been described in RA.
In RA or CIA, the normal organization of synovium appears imbalanced and becoming infiltrated with T and B cells, macrophages and neutrophils [41]. This fact triggers a hyperplasic membrane of synoviocytes that invades and destroys bone and cartilage. Conversely, dietary OLE supplementation could control the arthritis score through a lowering infiltration of immune cells, an appreciable reduction of oedema, synovial hyperplasia, cartilage degradation and bone erosion compared with CIA mice fed with a SD.
The development and progression of RA are closely related to production of MMPs, pro-inflammatory cytokines and oxidation products. Th-17 cells produce cytokines such as IL-17, considered a major player in the pathogenesis of CIA. IL-17 promotes inflammation by enhancing the production of crucial pro-inflammatory cytokines, mainly TNF-α, IL-1β, IL-6 and RANLK, among others [42]. At the pannus cartilage junction, these proinflammatory mediators can induce the production of additional cytokines, chemokines and MMPs, assisting cartilage degradation [43]. Moreover, TNF-α activates osteoclastogenesis, suppresses the osteoblasts recruitment and inhibits the expression of matrix genes [44]. Additionally, IL-1β is one of best serum markers of RA patients, which is correlated with disease activity and implicated in degradation and destruction of matrix cartilage and articular joints. Likewise, IL-6 can be detected in synovial fluid of RA patients. This cytokine not only acts as a growth factor for T and B cells and, antibody production but also, induces synoviocytes proliferation, osteoclast differentiation and increment of MMPs levels [43].
In terms of MMPs, MMP-3 is a proteinase secreted by synovial fibroblasts and chondrocytes in the synovium and, is responsible for the degradation of proteoglycan, various type of collagen types, fibronectin, aggrecan core protein, among others [32,45]. Our results agree with above studies and showed that high IL-1β, IL-6, IFN-γ, IL-17, TNF-α and MMP-3 levels were associated with disease onset and joint progression in CIA mice. However, arthritic mice fed with OLE-enriched diets showed a significant reduction in serum MMP-3 levels as well as in IL-17, TNF-α, IL1-β, IFN-γ and IL-6 pro-inflammatory cytokines in paw homogenates in comparison with CIA control group, correlating in parallel with the macroscopic and histological findings. Therefore, these data suggest that dietary OLE exerts anti-inflammatory effects through the modulated production of these key RA biomarkers. Similar data were found in our recent study, which reported that OLE controlled the production of inflammatory mediators decreasing IL-1β, TNF-α, INF-γ, IL-6, IL-17 and IL-18 in LPS-stimulated murine peritoneal macrophages [30] and, in previous works reported by Scotece et al., (2012; 2019) using a chondrogenic and macrophage cell line or human primary osteoarthritis (OA) chondrocytes LPS-stimulated [28,29].
RA patients present a significant ROS production [46]. This inflammation-associated action could induce osteoclast differentiation that could act in feedback way rising ROS levels [47]. Moreover, IL-1β, IL-6 or TNF-α promote osteoclasts formation. Although the presence of osteoclasts in bone loss has been scarcely documented in RA, it has been widely accepted as responsible for bone erosion in RA patients [48]. Actually, TRAP staining is considered one of the best histochemical markers of osteoclasts and Suzuki et al. described TRAP staining-positive multinucleated cells in CIA as a validated method [49]. Our data showed that joint sections from mice fed with OLE diet presented a marked reduction of TRAP-positive osteoclasts, suggesting a decrease of osteoclast activity in OLE-supplemented diet-fed mice.
To further evaluate the antioxidant role involved in OLE effects, iNOS expression was studied. iNOS is an inducible enzyme implicates in NO production of RA patients, inducing oxidative damage of arthritic joint [50]. Several works have reported how iNOS expression was induced in murine models of CIA, relating to NF-κB signaling pathway [8,32,51]. In our study, OLE-enriched diet showed comparable results to iNOS modulation in LPS-induced NO and ROS production, and iNOS overexpression in murine peritoneal macrophages [30]. Consequently, these data suggest that dietary OLE treatment could attenuate incidence of CIA through down-regulation of NF-kB and iNOS expression, regulating oxidative damage of cartilage and improving clinical signs.
COX-2 and mPGES-1, enzymes responsible for the overproduction of PGE 2 implicated in inflammation and pain hypersensitivity associated with RA, are up-regulated contributing to the disease progression through EP 4 receptor activation [45,52]. COX-2 enzyme expression is induced by cytokines such as TNF-α, IL-1β and IL-6. Our data are in concordance with other authors who reported increased levels of COX-2 in CIA models [32,45,53]. In the present study, we have demonstrated that dietary OLE treatment reduced COX-2 expression in knee joints from DBA/1j mice CIA model. These results are in accordance with those reported by Montoya et al. (2019) in LPS-induced murine peritoneal macrophages [30], in human monocytes stimulated with LPS [54] and in LPSactivated human primary OA chondrocytes [29]. Complementary, we have shown that dietary OLE supplementation reduced mPGES-1 expression, which was accompanied by a significant decrease in PGE 2 levels possibly due to decreased expression of COX-2 in the joint. Therefore, regulation of these pro-inflammatory biomarkers could represent a potential molecular target susceptible to OLE modulation that has not been previously established in vivo.
Signal transduction pathways strongly implicated in inflammation and oxidative stress include MAPKs, JAK-STAT and NF-κB. NF-κB, a dimeric transcription factor, activates multiple pro-inflammatory genes expression in the microenvironment of arthritic joints, such as Th-1 and Th-17, COX-2, iNOS, in addition to MAPKs [7,26]. In resting cells, NF-κB, p65 and p50 heterodimers composition is present in cytoplasm blocked by inhibitor protein IκB-α. However, pro-inflammatory stimulus undergoes phosphorylation and, subsequently, ubiquitination, then allowing NF-κB p65 and p50 heterodimers migrate to the nucleus and bind to specific promoter sequences [7]. Consequently, suppression of the NF-κB pathway could be a novel strategy for delaying the progress of RA. Our data agree with Rosillo et al. (2016), who reported that, in control CIA mice, IκB-α was significantly degraded and p65 and p50 subunits were translocated to the nucleus [55]. Nevertheless, dietary OLE treatment prevented nuclear p65-and p50-NF-κB translocations by blocking IκB-α degradation. These facts resulted in an ameliorated pro-inflammatory markers production and, then, reducing the joint inflammatory injury. Similar data were found recently by Carpi et al. (2019) who investigated the impact of OLE on NF-κB the activation and the expression of molecules associated with inflammatory and dysmetabolic responses in Simpson-Golabi-Behmel syndrome adipocytes TNF-α-stimulated. OLE attenuated the activation of NF-κB pathway and prevented NF-κB activity by directly interacting with the p65 subunit [56]. Complementary, LPS-induced inflammation in human primary OA chondrocytes also exhibited a significant regulation of IκB and NF-κB-p65 subunit after OLE treatment [29].
MAPK family members, including p38 kinases, ERK and JNK, play critical roles in many important cell processes, including inflammation. They have been detected in tissue from RA patients, suggesting an implication in pathogenesis of this disease [39]. Phosphorylated forms of MAPKs can induce transcriptional and post-transcriptional activation of different cytokines involved in RA. Moreover, ERK promotes pannus formation; JNK regulates the production of collagenases by synovial fibroblast; and, p38 modulates MMP-3 expression in fibroblasts and osteoclasts [39]. In addition to that, MAPKs may activate the JAK/STAT signaling pathway, which has been described in RA patients, contributing to fibroblast proliferation and osteoclastogenesis [57]. Our data agree with previous reports [45,55,57] where phosphorylation of MAPKs and STAT-3 were increased in CIA mouse group and, positively related to the severity of synovitis, whereas dietary OLE supplementation reduced significantly both MAPKs and STAT-3 activation at transcriptional level. Collectively, our data suggest that dietary OLE may repress cytokines production interfering negatively with pJNK, pp38, pERK MAPKs and pSTAT-3 signaling pathways.
Nrf-2 is a key regulator factor in anti-oxidant response, which plays key roles in cellular defense by modulating HO-1 transcription, enhances ROS removal and mitigates Th-17 cell-mediated inflammation [53]. Besides, it has been documented that in knockout mouse Nrf-2 deficiency accelerates the degeneration and joint damage in arthritis disease [58]. In the present study, accordingly with previous reports [45,55,[57][58][59], expression of both, Nrf-2 and HO-1, was decreased in the arthritic group, by contrast, dietary OLE could restore Nrf-2 and HO-1 expressions, conferring a role for Nrf-2/HO-1 axis in the beneficial effects of dietary OLE in this murine model of RA.

Conclusions
Collectively, our study has demonstrated, for the first time, the antioxidant, antiinflammatory and immunomodulatory effects of dietary OLE in CIA model, which were accompanied by an important attenuation of RA biomarkers (MMP-3), pro-inflammatory cytokines (IL-6, IL-1β, IL-17, IFN-γ, TNF-α), and inhibition of COX-2, mPGES-1 and iNOS overexpression. The mechanisms underlying these protective effects could be related to the activation of Nrf-2/HO-1 axis and the inhibition of relevant signaling pathways, such as JAK-STAT, MAPKs and NF-κB, controlling the inflammatory mediator's production.
Overall, our results exhibit preliminary evidence for OLE as a new dietary tool against autoimmune and inflammatory disorders, including RA. Nevertheless, further investigations are needed to provide insights into full biological significance of these results and the influence of secoiridoids and their properties on human autoimmune disorders.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/antiox10050650/s1, Figure S1: The purity of the extracted OLE was based on the 1 H and 13 C NMR spectra and HPLC analyses.