Tomatidine suppresses inflammation in primary articular chondrocytes and attenuates cartilage degradation in osteoarthritic rats

In this study, we investigated whether the anti-inflammatory effects of tomatidine alleviate osteoarthritis (OA)-related pathology in primary articular chondrocytes and a rat OA model. STITCH database analysis identified 22 tomatidine-target genes that were enriched in 78 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Moreover,39 of the 105 OA-related KEGG pathways were related to tomatidine-target genes. The top two OA-related KEGG pathways with tomatidine-target genes were the MAPK and neutrophin signaling pathways. Pretreating primary chondrocytes with tomatidine suppressed interleukin-1β (IL-1β)-induced expression of iNOS, COX-2, MMP1, MMP3, MMP13, and ADAMTS-5. Tomatidine also suppressed IL-1β-induced degradation of collagen-II and aggrecan proteins by inhibiting NF-κB and MAPK signaling. In a rat OA model, histological and immunohistochemical analyses showed significantly less cartilage degeneration in thetibiofemoral joints of rats treated for 12 weeks with tomatidine after OA induction (experimental group) than in untreated OA group rats. However, micro-computed tomography (μ-CT) showed that tomatidine did not affect remodeling of the subchondral bone at the tibial plateau. These data shows that tomatidine suppresses IL-1β-induced inflammation in primary chondrocytes by inhibiting the NF-κB and MAPK signaling pathways, and protects against cartilage destruction in a rat OA model.


INTRODUCTION
Osteoarthritis (OA) is a degenerative joint disease associated with chronic pain and disability; it is characterized by articular cartilage breakdown, synovial inflammation, and bone hypertrophy [1]. The constantly increasing numbers of OA patients worldwide incurs an expenditure of 1.0% to 2.5% Gross Domestic Product (GDP) in developed countries [2,3]. OA is a progressive joint disease without any effective curative treatment [4][5][6]. Therefore, it is of paramount importance to identify the underlying pathways involved in OA pathology and discover new therapies that can improve the quality of life for patients with OA.
The hallmark of OA is the imbalance between proinflammatory and anti-inflammatory cytokines [7]. The most critical pro-inflammatory cytokine involved in OA pathogenesis is interleukin-1β (IL-1β), which downregulates key extracellular matrix (ECM) proteins, type-II collagen and aggrecan, and upregulates matrix metalloproteinases (MMPs) and ADAMTS-5, which breakdown the cartilage tissue [8,9]. Therefore, anti-AGING inflammatory drugs are potential therapeutic candidates against OA.
Tomatidine, the major glycoalkaloid produced by the tomato plant, is associated with anti-apoptotic, antiinflammatory, anti-bacterial, and anti-cancer properties [10][11][12][13][14]. Tomatidine suppresses inflammation in LPSstimulated murine macrophages by inhibiting the NF-κB and JNK signaling pathways [15]. However, the therapeutic effects of tomatidine in OA are unknown. Therefore, in this study, we investigated the effects of tomatidine on IL-1β-induced inflammation in primary articular chondrocytes and cartilage degeneration in the rat OA model.

Bioinformatics analysis of tomatidine-target genes and OA-related KEGG pathways
The interaction network shows 22 tomatidine-target genes belonging to three shells based on the STITCH database analysis ( Figure 1A). The first shell contained two genes, MAPK1 and MAPK3; the second shell contained 10 genes, RAF1, PTPRR, RPS6KA1, RPS6KA3, PTPN7, PTPN11, STAT5A, DUSP1, MAP2K1, and MBP; and the third shell contained 10 genes, MAP2K2, DUSP4, JUN, FOS, 3TS1, BCL2, RPS6KA2, TP53, SMAD1, and DUSP6. The weight of MAPK1 was the highest according to the interaction network analysis of the tomatidine-target genes ( Figure  1B). DAVID database analysis identified 78 KEGG pathways (p< 0.05) with tomatidine-target genes, whereas, miRWalk2.0 database analysis showed 105 human OA-related KEGG pathways [16]. As shown in the Venn diagram, 39 OA-related KEGG pathways contained tomatidine-target genes ( Figure 1C). The top five OA-related KEGG pathways with tomatidine-target genes include MAPK signaling, neurotrophin signaling, colorectal cancer, renal cell carcinoma, and long-term potentiation pathways (Table 1). Figure 1D shows the enrichment information of the 22 tomatidine-target genes involved in the top five OArelated KEGG pathways. Among these, MAPK1, MAP2K1, MAPK3, and RAF1 are involved in all the top five KEGG pathways and are considered as hub genes. Figure 1E represents the circular diagrammatic representation highlighting the chromosomal positions and connectivity of the tomatidine-target genes. Among the 22 tomatidine-target genes, MAPK1 shows the greatest degree, betweenness, and closeness centrality. The top 2 shared KEGG pathways with highest p values are associated with inflammation, proliferation, differentiation, pro-survival, and retrograde signaling via MAPK and NF-κB signaling pathways (Figure 2A, 2B).

Tomatidine inhibits IL-1β-induced iNOS and COX-2 expression in primary chondrocytes
We analyzed the effects of tomatidine pre-treatment on the IL-1β-induced expression of iNOS and COX-2 in primary chondrocytes. IL-1β-treatedprimary chondrocytes showed significantly higher levels of iNOS and COX-2 proteins compared to the blank control, but, pre-treatment with 2.5, 5.0 and 10µM tomatidine significantly reduced iNOS and COX-2 expression in a concentration-dependent manner ( Figure  4A, 4B). These data demonstrate that tomatidine inhibits IL-1β-induced iNOS and COX-2 expression in primary chondrocytes.

Tomatidine inhibits IL-1β induced MMPs and ADAMTS-5 in chondrocytes
Next, we analyzed the effects of tomatidine on IL-1βinduced expression of MMPs (MMP1, MMP3, and MMP13) and ADAMTS-5 in primary chondrocytes. IL-1β-treatedprimary chondrocytes showed significantly higher expression of MMPs and ADAMTS-5 compared to the blank controls, but, tomatidine significantly reduced the expression of MMPs and ADAMTS-5 in a concentration-dependent manner ( Figure 5A, 5B). These findings show that tomatidine protects against OA progression by suppressing the expression of MMPs and ADAMTS-5 in primary chondrocytes.

Tomatidine suppresses IL-1β induced degradation of aggrecan and collagen-II in primary chondrocytes
IL-1β-treatedprimary chondrocytes showed significantly higher degradation of ECM proteins, aggrecan and collagen-II, compared with the blank controls based on western blotting and immunofluorescence staining assays ( Figure 6A-6C). However, tomatidine significantly reduced the degradation of aggrecan and collagen-II in the IL-1β-treated primary AGING chondrocytes in a concentration dependent manner ( Figure 6A-6C).

Tomatidine suppresses IL-1β induced activation of NF-κB in primary chondrocytes
We analyzed the effects of tomatidine on IL-1β-induced activation of NF-κB in primary chondrocytes. IL-1βtreated primary chondrocytes showed significantly higher levels of phospho-P65 compared to the blank controls, but, pre-treatment with tomatidine significantly reduced phospho-P65 levels in a concentration-dependent manner (Figure 7A, 7B; P < 0.05). Immunofluorescence assay showed that tomatidine inhibits nuclear translocation of P65 inIL-1β-treated primary chondrocytes in a concentration dependent manner ( Figure 7C). These results demonstrate that tomatidine suppresses activation of the NF-κB in IL-1β-treated primary chondrocytes.

Tomatidine suppresses IL-1β-induced activation of MAPK signaling pathway in primary chondrocytes
Next, we analyzed the effects of tomatidine on the activation status of the MAPK signaling pathway in IL-1β-treated primary chondrocytes. As shown in Figure  8A, 8B, IL-1β-treated primary chondrocytes showed significantly higher levels of phospho-P38, phospho-ERK, and phospho-JNK compared to the blank controls, but, tomatidine suppressed P38 activation in a concentration-dependent manner ( Figure 8A, 8B; P < 0.05). These data demonstrate that tomatidine reduces inflammation in primary chondrocytes by inhibiting the activation of p38 MAPK signaling pathway.
Tomatidine ameliorates OA progression in the rat OA model Next, we analyzed the effects of tomatidine on the progression of OA in the rat model. Histological analysis showed that damage to the articular cartilage was significantly reduced in the experimental group rats (OA model rats treated with tomatidine) compared to the OA group ( Figure 9A). Furthermore, immunohistochemical staining showed that tomatidine significantly reduced INOS and MMP3levels in the articular cartilage of experimental group rats compared to the OA group ( Figure 9B, 9C). This demonstrates that tomatidine ameliorates OA progression in the rat OA model.      AGING separation parameters were similar for both OA and tomatidine-treated experimental group rats ( Figure 10A, 10B; P > 0.05). These results show that tomatidine does not affect subchondral bone remodeling in OA model rats.

DISCUSSION
Our study demonstrates that tomatidine inhibits IL-1βinduced expression of iNOS, COX-2, and MMPs, and degradation of aggrecan and collagen-II in primary chondrocytes by suppressing the activation of NF-κB and MAPK signaling pathways. These results show that tomatidine sustains the levels of ECM proteins by suppressing the expression of proteins involved in inflammation and degradation of ECM proteins. Moreover, tomatidine reduces cartilage degeneration in the rat OA model. These findings reveal that tomatidine is a promising drug for OA treatment.
IL-1β is a major pro-inflammatory cytokine that is involved in OA pathology by inducing inflammatory responses and catabolic effects [9,[17][18][19]. IL-1β promotes the expression of MMPs, which are involved in the degradation of cartilage components [20]. IL-1β also induces ADAMTS-5, which promotes breakdown of aggrecan [21,22]. IL-1β stimulates the expression of iNOS and COX-2, which promote the release of well known inflammatory mediators, NO and PGE2 [23,24]. Several studies suggest that inhibition of these proinflammatory mediators hinders OA progression [19,20].
In this study, we hypothesized that the antiinflammatory properties of tomatidine will delay or abrogate OA progression in the primary chondrocytes and cartilage tissue. Chiu et al. reported that tomatidine inhibits the inflammatory response in LPS-stimulated mouse macrophages by suppressing the NF-κB and JNK signaling pathways [15]. The activation of NF-κB signaling pathway is implicated in the onset and progression of OA [25]. NF-κB pathway also regulates the expression of iNOS, COX-2, and MMPs [26,27]. Therefore, targeted inhibition of the NF-κB signaling pathway is a potential therapy for patients with OA. We demonstrate that tomatidine significantly inhibits IL-1βinduced expression of inflammatory factors by suppressing NF-κB activation. MAPK signaling is another key pathway involved in OA pathogenesis [28]. The OA chondrocytes demonstrate significantly higher levels of phospho-p38, phospho-JNK, and phospho-ERK compared to normal chondrocytes [29]. Our study demonstrates that tomatidine blocks IL-1β-stimulated MAPK pathway activation in primary articular chondrocytes. Overall, our study demonstrates that tomatidine exerts anti-inflammatory effects on IL-1βstimulated chondrocytes by suppressing the NF-κB and MAPK signaling pathways.
Our study has several limitations. Firstly, it is not clear if tomatidine inhibits NF-κB and MAPK pathways directly or indirectly through mediators. Secondly, a higher concentration of tomatidine does not demonstrate anti-inflammatory and cell survival effects, probably because it exceeds the optimal dosage.

CONCLUSIONS
In summary, our study shows that tomatidine inhibits IL-1β-stimulated inflammation in primary chondrocytes by suppressing the NF-κB and MAPK signaling pathways ( Figure 11). Moreover, tomatidine decreases the articular cartilage pathology in the rat OA model. These findings suggest that tomatidine is a potential treatment for OA, which needs to be ascertained by further investigations.

Identification of tomatidine-target genes, their interaction network and the OA-related KEGG pathways containing tomatidine-target genes
We identified and constructed an interaction network between tomatidine-target genes in first, second, and third shells using the Search Tool for Interacting Chemicals (STITCH) database [30]. Tomatidine was imported into the STITCH, and the predicted gargetgenes were abtained with default settings. Then, we calculated the degree, betweenness, and closeness of the tomatidine-related genes in the network using Cytoscape 3.7.2 [31]. Degree Centrality represents the number of links a node has with other nodes in the network. Closeness Centrality is a measure of the centrality of a node in a network, that is, more central the node is, the closer it is to other nodes in the network. Betweenness Centrality measures how important a node is based on all the number of shorter paths between a pair of nodes. We used the miRWalk2.0 database to retrieve KEGG pathways involved in human OA [16]. We used the Database for Annotation, Visualization, and Integrated Discovery (DAVID) to identify KEGG pathways enriched (p<0.05) with the tomatidine-target genes [32,33]. The Venn Diagram (Venny 2.1, http://bioinfogp.cnb.csic.es/tools/venny/index.html) was used to show the KEGG pathways that are common to OA and tomatidine-target genes.

Identification of the hub genes and KEGG pathways related to tomatidine-target genes
The GO plots were used to present the enrichment information of the top five KEGG pathways [34]. The genes in the top five KEGG pathways were considered as the hub genes. The circlize R package was used to AGING determine the centrality of all the tomatidine-targeted genes in the network [35]. The top 5 KEGG pathways with the smallest p values were selected and their relationship with the tomatidine-target genes was established using the Pathway Builder Tool 20 (https://www.proteinlounge.com).

Chemicals and reagents
Tomatidine (

Primary chondrocyte isolation and culture
The primary chondrocytes were isolated as described previously [36]. Briefly, the articular cartilage was obtained from the knees of male Sprague-Dawley rats, cut into 1 mm 3 pieces, and digested with trypsin for 30 min at 37°C. Then, the primary chondrocytes were obtained by digestion with 0.2% collagenase II for 4-6 h at 37°C. The primary chondrocytes were cultured in DMEM/F12 medium supplemented with 10% FBS and 1% penicillin/streptomycin at 37°C and 5% CO2. The medium was changed every two days. The primary chondrocytes from 2 nd and 3 rd passages were used for further experiments.

Cell viability assay
Tomatidine cytotoxicity was analyzed using the CCK-8 assay according to manufacturerʼs instructions. Briefly, the primary chondrocytes were seeded in a 96-well plate (5 × 10 3 /well) for 24h, pre-treated with different concentrations of tomatidine (2.5, 5, 10 or 20μM) for 2 h then treated with10ng/mlIL-1β for 24h. Primary chondrocytes treated with DMSO, tomatidine or IL-1β alone were used as controls. Subsequently, the cells were incubated with 10μl CCK-8 solution for 1h at 37°C. The absorbance was measured at 450nm using a plate reader (Bio-Rad, Richmond, CA, USA).

Immunofluorescence staining
The primary chondrocytes were seeded in a 24-well plate and treated with 10ng/ml IL-1β alone or in combination with 10μM tomatidine. The cells were then fixed with 4% paraformaldehyde for 10min, permeated with 0.1% Triton X-100 for 10min, and blocked with 1% BSA for 30min. Subsequently, the cells were incubated overnight at 4 °C with antibodies against collagen-II, aggrecan, and P65 at 1:200 dilutions. Then, after washing thrice with PBS, the cells were incubated with Cy3-conjugated goat anti-rabbit or Cy3conjugated goat anti-rat secondary antibodies at a 1:75 dilution for 1h. The cells were incubated with 4-6-Diamidino-2-phenylindole (DAPI) for 10 min to stain the nuclei, and examined using a automated EVOS fluorescence microscope (Life Technologies, USA).

Animal experiments and histological analysis
The animal protocol was approved by the Ethics and Animal Research Committee of Huazhong University of Science and Technology. We purchased male Sprague-Dawley rats weighing 250-300g from the Experimental Animal Center of Tongji Medical College (Wuhan, AGING China) and induced OA as described previously [37]. The animals (n=5/group) were randomly divided into blank control, OA, and experimental groups. OA pathogenesis was induced by surgery involving anterior cruciate ligament transaction and partial medial meniscectomy [37]. The experimental group rats were fed with 25 mg/ Kg/ day tomatidine as previously described [38]. After 12 weeks, the animals were sacrificed, the tibiofemoral joints were collected, fixed with 4% paraformaldehyde for 24h, and decalcified by incubation in 10% EDTA for 20 days. Then, the specimens were embedded in wax blocks, cut into 3-5μm thick sections, and stained with safranin O-fast green(S-O) and toluidine blue. The stained specimens from all 3 rat groups were scored according to the Osteoarthritis Research Society International (OARSI) recommendations and graded in a blinded manner. Another set of tibiofemoral joint sections from the 3 groups were incubated with antibodies against INOS and MMP3. The numbers of INOS and MMP3 positive cells were determined using the Image-J software.

Micro-computed tomography (μ-CT)
We scanned the right knee joints of the rats using the μ-Computed Tomography system (Scanco Medical, Bassersdorf, Switzerland). We obtained 21.0μm resolution μ-CT images at 70kV and 113μA and generated 3D images using the CT system software. The same region of interest (ROI) was drawn on 100 consecutive slices of 2.1mm total thickness representing the subchondral bone at the tibial plateau and the values for bone parameters such as bone volume/tissue volume (BV/TV), trabecular numbers (Tb.N) and trabecular separation (Tb.Sp) were determined.

Statistical analysis
All experiments were performed in triplicate and the data presented as means ± standard deviation (SD). All data were analyzed using the Shapiro-Wilk test for normal distribution and Levene's test for homogeneity of variances. The statistical differences between normally distributed groups were analyzed by one-way analysis of variance (ANOVA) and Studentʼs t-test. The nonparametric data (OARSI scores) were compared using the Kruskal-Wallis H test. P-value < 0.05 was considered statistically significant.

AUTHOR CONTRIBUTIONS
HY designed the experiments; XC, TY, XH and YX performed the experiments;BN performed instrument operations; RZ analyzed the data; XC and TY prepared the manuscript. All authors read and consented to the final manuscript.