Chondroprotection of PPARα activation by WY14643 via autophagy involving Akt and ERK in LPS‐treated mouse chondrocytes and osteoarthritis model

Abstract Autophagy maintains cellular homoeostasis. The enhancement of autophagy in chondrocytes could prevent osteoarthritis (OA) progression in articular cartilage. Peroxisome proliferator‐activated receptor α (PPARα) activation may also protect articular chondrocytes against cartilage degradation in OA. However, whether the protective effect of activated PPARα is associated with autophagy induction in chondrocytes is not determined. In this study, we investigated the effect of PPARα activation by its agonist, WY14643, on the protein expression level of Aggrecan and ADAMTS5, and the protein expression level of autophagy biomarkers, including LC3B and P62, using Western blotting analysis in isolated mouse chondrocytes pre‐treated with lipopolysaccharides (LPS, mimicking OA chondrocytes) with or without the autophagy inhibitor chloroquine diphosphate salt. Furthermore, Akt and ERK phosphorylation was detected in LPS‐treated chondrocytes in response to WY14643. In addition, the effect of intra‐articularly injected WY14643 on articular cartilage in a mouse OA model established by the destabilization of the medial meniscus was assessed using the Osteoarthritis Research Society International (OARSI) histopathology assessment system, along with the detection of Aggrecan, ADAMTS5, LC3B and P62 protein levels using immunohistochemistry assay. The results indicated that PPARα activation by WY14643 promoted proteoglycan synthesis by autophagy enhancement in OA chondrocytes in vivo and in vitro concomitant with the elevation of Akt and ERK phosphorylation. Therefore, autophagy could contribute to the chondroprotection of PPARα activation by WY14643, with the implication that PPARα activation by WY14643 may be a potential approach for OA therapy.

biosynthesis is an effective therapeutic approach for OA. 4 Apple procyanidins are promising food components that inhibit OA progression by promoting mitochondrial biogenesis and proteoglycan homoeostasis with the up-regulation of Aggrecan in primary chondrocytes. 5 FoxO transcription factors modulate proteoglycan in cartilage homoeostasis and OA, protecting against OA-associated cartilage damage. 6 Suramin increases cartilage proteoglycan accumulation in vitro and protects against joint damage triggered by papain injection in mouse knees in vivo. 7 As one of three subtypes of peroxisome proliferator-activated receptors (PPARs, including PPARα，PPARβ/δ and PPARγ), PPARα responds to specific ligands by altering gene expression to regulate lipid and lipoprotein metabolism, apoptosis and inflammatory responses in the liver and other organs of the human body. [8][9][10] Some studies have reported that PPARα plays an important role in chondrocyte metabolism. Activation of PPARs α, β/δ and γ impairs TGF-β1-induced collagen production and modulates the TIMP-1/MMPs balance in three-dimensional cultured chondrocytes. 11 Agonists of PPARα, β/δ and γ reduce TGF-β1-induced proteoglycans' production in chondrocytes. 12 PPARα down-regulates AGE-induced TGF-β and MMP-9 expression in chondrocytes. 13 PPARα activation pathway potentiates interleukin-1 receptor antagonist production in cytokine-treated chondrocytes. 14 In 2011, Clockaerts et al report that PPARα activation by its agonist, WY14643, decreases inflammatory and destructive responses in OA cartilage, although it does not have an effect on COL2A1 or Aggrecan mRNA expression. 15 These studies indicate that PPARα activation may have a protective effect on articular cartilage against OA progression.
Autophagy is a highly conserved homoeostatic process, maintaining cellular homoeostasis by sequestering and degrading cytosolic macromolecules, damaged organelles and some pathogens. 16 Moreover, the relationship between PPARα and autophagy and its effect on cell metabolism have been reported. 9,17-19 TAK1 regulates hepatic lipid metabolism and tumorigenesis via the AMPK/ mTORC1 axis, affecting both autophagy and PPARα activity. 17 Inhibition of glycogen synthase kinase 3β promotes autophagy to protect mice from acute liver failure (ALF) mediated by PPARα. 18 PPARα-mediated induction of autophagy ameliorates liver injury in cases of ALF by attenuating inflammatory responses, indicating a potential therapeutic application for ALF treatment. 9 Pharmacologic activation of PPARα reverses the normal suppression of autophagy in the fed state, inducing autophagic lipid degradation, or lipophagy. 19 These studies demonstrate that PPARα could induce autophagy to protect liver against pathological damage. Lots of evidence indicates that autophagy also participates in OA pathological progression and the enhancement of autophagy could protect articular cartilage from OA progression. [20][21][22][23] For example, autophagy is a protective mechanism in normal cartilage, and its ageing-related loss is linked with cell death and OA. 20 Autophagy activation by rapamycin and Torin 1 reduces severity of experimental OA. [21][22][23] However, whether PPARα might regulate autophagy to protect cartilage against OA pathological progression is not determined.
In this study, we detected the effect of PPARα activation by its agonist WY14643 on autophagy and proteoglycan synthesis in LPS-treated mouse articular chondrocytes (mimicking OA chondrocytes) 24

| Mouse chondrocyte isolation and culture
Chondrocytes of knee cartilage were isolated from neonatal male

| Western blotting analysis
Protein extracts were subjected to SDS-PAGE (8%-12%) and transferred to a PVDF membrane (GE Healthcare, Herfordshire, UK) as previously described. 26,27 The membrane was incubated with various primary antibodies as required at 4°C overnight, followed by the addition of the corresponding secondary antibodies at room temperature for 1-2 hours. An enhanced chemiluminescence detection kit was used to detect antibody reactivity (Pierce, Rockford, IL, USA).

| Established mouse model of experimental OA
Six-week-old male mice (average weight 28 g) were acclimatized to the laboratory environment for 2 weeks before the following experiments.
Mouse joints were randomly divided into four groups (16 mice, 32 knee joints), including sham (four mice, eight knee joints), control (OA, four mice, eight knee joints), OA+DMSO (four mice, eight knee joints), and OA+WY14643 (four mice, eight knee joints). The mouse experimental model of OA was induced by DMM as previously described. 28 After the first week post-surgery, DMSO and WY14643 (5 × 10 −3 g/ kg) were intra-articularly injected once every 3 days, respectively, and mice were not killed until 4 weeks after injection. The protocol was approved by the committee on the Ethics of Animal Experiments of Xiamen University (ID no. 20170301).

| Histopathological scores
The fresh samples were fixed in 4% paraformaldehyde for 48 hours followed by decalcification in 10% EDTA-2Na for 3 weeks, and then paraffin-embedded. Three-micrometre-thick sections of the samples were cut in sagittal plane from the medial side of joint and were stained by Safranin O-Fast green stain. The articular cartilage thickness in medial femur condyle and tibial plateau (from superficial zone to subchondral bone) was measured using Image-Pro Plus 6.0 software by the two blinded observers. The degree of articular cartilage degeneration was then assessed using the OARSI histopathology assessment system. 29 Grade 0 in OARSI system represents normal cartilage and increasing grade (from 1 to 6) indicates a more biologically cartilage degeneration.

| Immunohistochemistry assay
Three-micrometre-thick sections were obtained from the paraffin-embedded specimens prior to immunohistochemistry assay.

| Statistics analysis
Data were presented as means ± SEM at least three separate experiments for each group. After a test for normal distribution and Variance homogeneity detection, differences between the groups were examined for statistical significance using one-way analysis of variance following Tukey's post hoc tests with GraphPad Prism version 5 (GraphPad Software, Inc, San Diego, CA. USA). OARSI Scoring was performed using Kruskal-Wallis test with spss 19.0 software (SPSS, Chicago, IL). Statistical significance was determined at level of P < 0.05.*P < 0.05;**P < 0.01.

| PPARα activation by WY14643 promoted proteoglycan synthesis by the enhancement of autophagy in LPS-treated mouse articular chondrocytes
To investigate whether autophagy linked to PPARα participated in OA pathological progression, the biomarkers of proteoglycan synthesis and autophagy were detected with Western blotting in LPS-

| AKT and ERK were involved in autophagy enhancement and proteoglycan synthesis by WY14643-stimulated PPARα activation in LPStreated mouse articular chondrocytes
As shown in Figure 2A

| Intra-articular injection of WY14643 ameliorated articular cartilage degeneration in a mouse OA model
As shown in Figure 3A

| Intra-articular injection of WY14643 promoted proteoglycan synthesis in a mouse OA model
To confirm whether WY14643 ameliorated articular cartilage degeneration via enhancing proteoglycan synthesis, Aggrecan and ADAMTS5 expression levels in chondrocytes of medial femur condyle and tibial plateau were measured using immunohistochemistry assay. As shown in Figure 4A

| Intra-articular injection of WY14643 enhanced autophagy in a mouse OA model
As shown in Figure 5A, the LC3B expression level in either OA or OA+DMSO group was lower than Sham and OA+WY14643 groups. Compared with OA+DMSO group, intra-articular

| Intra-articular injection of WY14643 enhanced p-AKT and p-ERK in a mouse OA model
As shown in Figure 6A, p-Akt in either OA or OA+DMSO group was lower than Sham and OA+WY14643 groups. Intra-articular injection of WY14643 significantly elevated p-Akt level, near to Sham group (*P < 0.05, vs OA+DMSO group). P-ERK mainly expressed at a F I G U R E 3 Effect of intra-articular injection of WY14643 on articular cartilage in a mouse OA model. A. Sections in different treated groups were examined using Safranin O-Fast green stain (original magnification ×20). B. Bar graph shows the change in articular cartilage thickness in medial femur condyle and tibial plateau. Data are mean ± SEM of eight joints per group (*P < 0.05). C. Bar graphs show the histopathological scores performed by OARSI histopathology assessment system in medial femur condyle and tibial plateau. Data are mean ± SEM of eight joints per group (*P < 0.05) lower level in Sham, OA and OA+DMSO groups ( Figure 6B). Similar to p-Akt, p-ERK was significantly increased in the WY14643-treated group, compared with OA+DMSO group ( Figure 6B,*P < 0.05).

| D ISCUSS I ON
Our findings demonstrate that PPARα activation by WY14643 pro- The positively regulatory effect of PPARα activation on Akt phosphorylation has previously been reported. PPARα activation protects the type 2 diabetic rat myocardium against ischaemiareperfusion injury via the activation of the PI3K/Akt and nitric oxide pathway. 33 An enhanced Akt phosphorylation is observed in the ischaemic myocardium of WY14643-treated rats. 34 We also ob- As a key regulator of mTOR (a suppressor of autophagy), Akt is always also considered to be an autophagy suppressor. For example, sucrose induces autophagy via inhibiting Akt/mTOR pathway in human chondrocytes. 35 In OA human cartilage, increased PTEN  37 In this study, we also found that WY14643 induced autophagy concomitant with the elevation of Akt phosphorylation. TCN, an Akt inhibitor, partially reduced WY14643-induced autophagy. It might be due to the effect of one of the upstream regulators of Akt, the class IA phosphatidylinositol 3-kinase p110-β subunit, which has been reported to be a positive regulator of autophagy. 38 Hence, Akt might not consistently function as an autophagy suppressor and regulate autophagy in a context-dependent manner, 37 and it is conceivable that Akt activation could promote autophagy induction in WY14643-treated OA chondrocytes.
Activation of JNK and ERK is attenuated by WY-14643 treatment in the acute liver injury mouse model. 9 Inactivation of PPARα results from the activation of MAPK pathway during cardiac hypertrophic growth. 39 These studies indicate the reciprocal inhibition between PPARα and MAPK. However, PPARα ligands have been shown to activate MAPK family members in a variety of different cell types. 40 Gardner et al have previously reported that the PPARα agonist, nafenopin, rapidly induces ERK and/or p38 phosphorylation in rat liver epithelial cells (GN4). 41 Consistent with Gardner et al's study, we also observed that WY14643, one of the PPARα ligands, elevated ERK phosphorylation in LPS-treated chondrocytes. The contradictory results might be because of the property of PPARα.
As a nuclear receptor, PPARα controls a variety of target genes to regulate cell metabolism. Meanwhile, as a phosphoprotein, PPARα is involved in cell events in a nongenomic manner through crosstalking to lots of signal molecules. 8,40 Therefore, we speculated that the elevation of ERK phosphorylation by WY14643 in LPS-treated chondrocytes might be as a result of the crosstalk among PPARα, NF-κB and ERK as well as previous studies in other cells. 9,[42][43][44] Cisplatin treatment activates ERK and subsequently promotes autophagy in ovarian cancer cells. 45 Inhibition of MEK/ERK activation attenuates autophagy and potentiates pemetrexed-induced activity against HepG2 hepatocellular carcinoma cells. 46 Platycodin D induces apoptosis and triggers ERK-and JNK-mediated autophagy in human hepatocellular carcinoma BEL-7402 cells. 47 Consistent with the aforementioned studies, our data that PD98059, a MAPK inhibitor, partially reduced autophagy enhancement by WY14643-stimulated PPARα activation indicate the promoted effect of p-ERK on autophagy in LPS-treated chondrocytes.
Therefore, we suggest that PPARα activation by WY14643 enhanced autophagy through the elevation of ERK phosphorylation in LPS-treated chondrocytes.
The PPARα agonists, such as fibrates, fenofibrate, could effectively ameliorate OA symptoms in cell, animal OA model and OA patients. For example, fibrates modulate inflammatory properties of IPFP (infrapatellar fat pad) and synovium, representing a potential disease-modifying drug for OA therapy. 48 In erosive hand OA patients, fenofibrate treatment was associated with significant decreases in pain score, tender joint count, duration of morning stiffness, disease activity score, Cochin index and ESR. 49 In this study, we found that WY14643 ameliorated cartilage degeneration in a mouse OA model concomitant with autophagy enhancement.
Similarly, WY14643 inhibits LPS-induced inflammation in synovial fibroblasts via NF-kB pathway 32 and inhibits the inflammatory and destructive responses in human OA cartilage explants. 15 Hence, WY14643 may also ameliorate OA symptoms like other PPARα agonists.
In conclusion, PPARα activation by WY14643 could promote chondrocyte proteoglycan synthesis via the enhancement of autophagy involving Akt and ERK phosphorylation, exhibiting an effective chondroprotection on articular cartilage in OA pathological progression (Figure 7).

ACK N OWLED G EM ENTS
The present study was supported by the National Science Foundation of China (nos 81572189 and 81371952) and the Natural Science Foundation of Xiamen, China (no. 3502Z20159015).

CO N FLI C T S O F I NTE R E S T
The authors declare that they have no competing interests.

Chun Xia
https://orcid.org/0000-0001-8585-9313 F I G U R E 7 Model of PPARα activation by WY14643 promotes proteoglycan synthesis via the enhancement of autophagy in OA chondrocytes