Avenanthramide C as a novel candidate to alleviate osteoarthritic pathogenesis

Abstract

Osteoarthritis (OA) is a degenerative disorder that can result in the loss of articular cartilage. No effective treatment against OA is currently available. Thus, interest in natural health products to relieve OA symptoms is increasing. However, their qualities such as efficacy, toxicity, and mechanism are poorly understood. In this study, we determined the efficacy of avenanthramide (Avn)-C extracted from oats as a promising candidate to prevent OA progression and its mechanism of action to prevent the expression of matrix-metalloproteinases (MMPs) in OA pathogenesis. Interleukin-1 beta (IL-1β), a proinflammatory cytokine as a main causing factor of cartilage destruction, was used to induce OA-like condition of chondrocytes in vitro. Avn-C restrained IL-1β-mediated expression and activity of MMPs, such as MMP-3, -12, and -13 in mouse articular chondrocytes. Moreover, Avn-C alleviated cartilage destruction in experimental OA mouse model induced by destabilization of the medial meniscus (DMM) surgery. However, Avn-C did not affect the expression of inflammatory mediators (Ptgs2 and Nos) or anabolic factors (Col2a1, Aggrecan, and Sox9), although expression levels of these genes were upregulated or downregulated by IL-1β, respectively. The inhibition of MMP expression by Avn-C in articular chondrocytes was mediated by p38 kinase and c-Jun N-terminal kinase (JNK) signaling, but not by ERK or NF-κB. Interestingly, Avn-C added with SB203580 and SP600125 as specific inhibitors of p38 kinase and JNK, respectively, enhanced its inhibitory effect on the expression of MMPs in IL-1β treated chondrocytes. Taken together, these results suggest that Avn-C is an effective candidate to prevent OA progression and a natural health product to relieve OA pathogenesis.

INTRODUCTION

Osteoarthritis (OA) is the most prevalent degenerative arthritis. It is characterized by a progressive destruction of articular cartilage accompanying phenotypes of whole-joint disease, such as cartilage destruction, synovitis, osteophyte formation, and subchondral bone sclerosis (1). In the cartilage, chondrocytes are unique resident cells that can produce cartilage-specific extra-cellular matrix (ECM) components as well as various catabolic and anabolic factors. OA pathogenesis is mainly caused by an imbalance between anabolic and catabolic factors due to de-differentiation and apoptosis of chondrocytes (2).

Various biochemical signals in chondrocytes can lead to the imbalance by modulating proinflammatory cytokine production, the cessation of ECM synthesis, and the irreversible degradation of the ECM by the action of matrix-metalloproteinases (MMPs) (3, 4). Aging and mechanical stresses are potential OA-causing factors that can promote the expression of pro-inflammatory cytokines as well as pro-inflammatory mediators such as cyclo-oxygenase-2 (COX-2) encoded by Ptgs2 and inducible nitric oxide synthase (iNOS) encoded by Nos2 in chondrocytes (5). Interleukin-1 beta (IL-1β), IL-6, and tumor necrosis factor alpha (TNF-α) are well-known pro-inflammatory cytokines causing OA. They are responsible for the production of matrix-degrading enzymes such as MMPs and ADAMTS (aggrecanases). Previous studies have shown that hypoxia-inducible factor (HIF)-2α encoded by Epas1 (2) and zinc importer ZIP8 encoded by Slc39a8 (6) play crucial roles in OA pathogenesis by upregulating MMPs (such as MMP-3, MMP-12, and MMP-13) and ADAMTS.

Numerous signaling pathways are involved in inflammatory responses and ECM degradation in OA pathogenesis. It has been reported that phosphorylated levels of mitogen-activated protein (MAP) kinases, including extracellular signal-regulated kinase (ERK), p38 kinase, and c-Jun N-terminal kinase (JNK), are up-regulated in osteoarthritic cartilage (7). In fact, MAP kinases play a crucial role in OA pathogenesis by promoting cytokine-mediated MMP expression. For instance, IL-1β can stimulate the induction of MMP-3 and MMP-13 via activation of p38 kinase and JNK in articular chondrocytes (7). ERK activity is predominantly required for the expression of MMP-1 and MMP-3 in IL-1β-stimulated SW1353 human chondrosarcoma cells (8). It has been reported that IL-1β expression in osteoarthritic cartilage and synovial fluids is mediated by nuclear factor-κB (NF-κB) (9). NF-κB dimers are sequestered by an inhibitor of κB (IκB) in the cytoplasm in an inactive form. Signal-induced degradation of IκB initiates the activation of NF-κB, leading to its translocation into the nucleus and up-regulation of inflammation-related genes, including NOS2 and COX-2 (9, 10).

Avenanthramides (Avns) are unique polyphenolic compounds exclusively found in oats (11). It has been scientifically verified that oats (Avena sativa L.) have health-promoting effects including cholesterol-lowering potential of β-glucan (12). Oats also have anti-oxidant effects due to the presence of phenolic compounds Avns (13). Avns have been isolated in three major isoforms: Avn-A, -B, and -C. They have shown essential functions in several biological activities, including anti-oxidant, anti-proliferative, anti-histamine, and anti-inflammatory functions (14). Among them, previous studies have largely focused on the high anti-oxidant effects of Avn-C to demonstrate the potential of using this natural health product as an anti-cancer drug (15). Avn-C also has anti-inflammatory effects by blocking the expression of pro-inflammatory cytokines such as IL-6 and TNF-α through decreased NF-κB activity in endothelial cells (16) and skeletal muscle cells (17). In addition, Avn-C can prevent lung inflammation under hypoxia by inhibiting SIRT1-mediated COX-2 expression in lung cancer cell lines (18).

In the current study, we determined the potential of using Avn-C to prevent OA pathogenesis. Avn-C markedly blocked cartilage destruction by restraining the expression and activities of MMPs in articular chondrocytes. The inhibitory effects of Avn-C on cartilage destruction were mediated by p38 kinase and JNK signaling. Results of this study suggest possible functions and underlying molecular mechanisms of Avn-C in alleviating OA pathogenesis.

RESULTS

Avn-C reduces catabolic gene expression during IL-1β-mediated in vitro OA-like pathogenesis

To determine effects of Avn-C on OA pathogenesis, we induced an in vitro arthritis condition in mouse articular chondrocytes by treating cells with IL-1β, a main factor causing arthritis. Expression levels of well-known catabolic and pro-inflammatory factors of OA pathogenesis as well as anabolic factors were examined by reverse transcription-polymerase chain reaction (RT-PCR). Avn-C did not recover IL-1β-mediated reduction of anabolic factors such as Aggrecan and Col2a1, main constituents of cartilage ECM, or Sox9, a major transcription factor of Col2a1 (Fig. 1A). Moreover, Avn-C treatment did not inhibit IL-1β-mediated induction of pro-inflammatory factors Il-6, Ptgs2, or Nos2 (Fig. 1B). On the other hand, transcript levels of Mmp-3, -9, -12, -13, and Adamts-4 were dramatically increased by IL-1β, showing a well-induced OA pathologic condition. They were significantly reduced by Avn-C in a dose-dependent manner in articular chondrocytes (Fig. 1C). In consistent with a previous study (2), IL-1β did not induce the expression of Mmp-14 or Adamts-5 (Fig. 1C). We focused on inhibitory effects of Avn-C on the expression of Mmp-3, -12, and -13 as these genes were inhibited the most effectively by Avn-C. Both transcriptional levels and protein levels of MMP-3, -12, and -13 were inhibited by Avn-C in IL-1β-treated mouse articular chondrocytes as determined by quantitative real-time (qRT)-PCR (Fig. 2A) and Western blotting (Fig. 2B), respectively. Matrix-degrading activities of secreted MMP-3, -12, and -13 were also significantly reduced by Avn-C treatment (Fig. 2C). These data show that Avn-C can inhibit IL-1β-induced in vitro OA pathogenesis by restraining the expression of catabolic factors, especially MMP-3, -12, and -13.

Fig. 1.

Avn-C reduces the expression of catabolic factors during IL-1β-mediated in vitro OA-like pathogenesis. Primary cultured mouse articular chondrocytes were treated with 2 ng/ml IL-1β for 24 h in the presence of the indicated concentration of Avn-C. Expression level of each gene was then analyzed by RT-PCR. (A) Anabolic factors of OA pathogenesis: Aggrecan, Col2a1, and Sox9. (B) Pro-inflammatory factors: Il-6, Ptgs2, and Nos2. (C) Catabolic factors: Mmp-3, -9, -12, -13, -14, Adamts4, and Adamts5.

Fig. 2.

Avn-C inhibits the expression and activity of MMPs in IL-1β-treated chondrocytes. Cells were cultured without serum and treated with 2 ng/ml IL-1β for 24 h in the presence of the indicated concentration of Avn-C. (A) Transcriptional levels of Mmp-3, -12, and -13 were analyzed by qRT-PCR (n = 3). (B) Protein levels were examined by Western blot analysis (n = 3). (C) Cultured media were subjected to ELISA to measure activities of MMP-3, -12, and -13 (n = 4). Values are presented as means ± SEM (*P < 0.05; **P < 0.01; and ***P < 0.005).

Avn-C blocks destabilization of the medial meniscus (DMM)-induced cartilage destruction

We attempted to determine in vivo effects of Avn-C in the surgical model of DMM, a type of experimental OA model in mice. After 200 mM Avn-C or dimethyl sulfoxide (DMSO) as a control was administered by intra-articular injection every week for six weeks to DMM or sham (control) mice, cartilage destruction was histologically analyzed by Safranin-O staining (Fig. 3A) and scored according to the Osteoarthritis Research Society International (OARSI) system (Fig. 3B). Two control groups of sham mice injected with DMSO or Avn-C showed relatively intact cartilages, whereas DMM mice with Avn-C treatment presented less cartilage destruction than DMSO-injected DMM mice which exhibited severe progression of cartilage destruction (Fig. 3A). The OARSI grade of Avn-C-injected DMM mice was decreased about 55% compared to that of DMSO-injected DMM mice (Fig. 3B). It has been reported that the more severe the cartilage degeneration, the thicker the subchondral bone plate (SBP) (19). SBP thickness of articular cartilage was increased in DMM mice. However, it was slightly inhibited in Avn-C injected DMM mice compared to DMSO-injected DMM mice (Fig. 3B). In addition, osteophyte formation seemed to be slightly prevented by Avn-C in DMM-induced OA mice model, although such prevention was not significant (Fig. 3B). These data indicate that Avn-C has a protective effect on DMM-induced cartilage destruction.

Fig. 3.

Avn-C blocks DMM-induced cartilage destruction. Mice were subjected to sham operation or DMM surgery and injected with DMSO (vehicle) or Avn-C into knee joints (sham with DMSO; n = 7, DMM with DMSO; n = 7, sham with Avn-C; n = 11, and DMM with Avn-C; n = 11). (A) DMM-induced experimental OA was analyz-ed by safranin-O staining for the cartilage section. Scale bars, 100 μm. (B) OA parameters were scored, including OARSI grade for cartilage destruction, thickening of the subchondral bone plate (SBP), and osteophyte maturity. (C-E) Immunohistochemical staining for MMP-3 (C), MMP-12 (D), and MMP-13 (E) in cartilage tissues from sham- or DMM-operated mice with intra-articular injection of 10 μl of DMSO or 200 μM Avn-C. Scale bars, 25 μm. Values are presented as means ± SEM (NS, not significant; **P < 0.01; and ***P < 0.005).

Avn-C reduces the expression of MMPs in experimental mouse OA cartilages

Next, to examine the in vivo expression of MMPs, we performed immunohistochemical analysis using cartilage tissue sections from experimental OA mouse models which we analyzed in Fig. 3. Expression levels of MMP-3 (Fig. 3C), MMP-12 (Fig. 3D), and MMP-13 (Fig. 3E) were definitely reduced in Avn-C treated DMM cartilages than in sham and DMSO-treated DMM mice, consistent with in vitro results from IL-1β-mediated arthritic chondrocytes shown in Fig. 1C. Altogether, these data indicate that Avn-C has a protective effect on DMM-induced cartilage destruction by preventing the expression of matrix-degrading enzymes such as MMP-3, -12, and -13.

Avn-C inhibits IL-1β signaling by blocking p38 MAP kinase and JNK signaling pathways

To determine the molecular mechanisms involved in the protective effect of Avn-C on the pathogenesis of arthritis, we examined signaling pathways involved in IL-1β mediated MMP expression. It has been reported that MAP kinases including ERK, p38 kinase, and JNK are activated by IL-1β and the activation of these signaling pathways is involved in the pathogenesis of arthritis, including degradation of cartilage ECM, inflammation, and apoptosis of chondrocytes (20). The maximal activation of all three MAP kinases in 15-30 min after IL-1β treatment was verified in current experiments by determining phosphorylation on Thr202/Tyr204 of ERK, Thr180/Tyr182 of p38 kinase, and Thr183/Tyr185 of JNK (Fig. 4A). Pre-treatment with Avn-C significantly blocked p38 kinase and JNK signaling induced by IL-1β, but not ERK signaling (Fig. 4B). NF-κB is a major transcription factor that controls gene expression involved in the production of cytokines such as IL-6 and TNF-α (21). IL-1β also activated NF-κB signaling in mouse chondrocytes as determined by the degradation of IκB in 15-30 min after IL-1β stimulation (Fig. 4A). However, degradation of IκB was sustained by the addition of Avn-C in IL-1β stimulated articular chondrocytes, indicating that Avn-C did not modulate IL-1β-mediated NF-κB signaling (Fig. 4B). Next, we examined whether IL-1β-induced p38 kinase and JNK signaling regulated the expression of MMP-3, -12, and -13. Treatment with SB203580, a specific inhibitor of p38 kinase, blocked IL-1β-induced expression of Mmp-3 and -13, but not Mmp-12 (Fig. 4C). SP600125, a specific inhibitor of JNK, more effectively prevented the expression of all three Mmps (Fig. 4C). In addition, we tested synergistic effects of Avn-C on the inhibition of each signaling pathway with specific inhibitors. Co-treatment with Avn-C and SB203580 or SP600125 more dramatically inhibited the expression of all three Mmps tested in this study (Fig. 4D). Interestingly, Mmp-12 expression was also inhibited by co-treatment with Avn-C and SB200125, although IL-1β -induced Mmp-12 up-regulation was not blocked by only p38 kinase inhibition (Figs. 4C, D). These data support that the protective effect of Avn-C on OA pathogenesis is due to inhibition of p38 kinase and JNK signaling pathways causing expression of MMPs.

Fig. 4.

Avn-C inhibits IL-1β signaling by blocking p38 MAP kinase and JNK. (A) Primary cultured mouse articular chondrocytes were treated with 2 ng/ml IL-1β for the indicated time. Phosphorylation of ERK, p38 kinase, and JNK was determined by Western blot analysis. Degradation of IκBα was detected to determine the activation of NF-κB. (B) Cultured cells were treated with IL-1β for 15 min in the presence of Avn-C at indicated concentration and then subjected to Western blot analysis. (C) Chondrocytes were treated with IL-1β in the presence of indicated dose of p38 kinase inhibitor, SB203580, or JNK inhibitor, SP600125. Mmp-3, -12, and -13 expression levels were analyzed by RT-PCR. (D) Cells were co-treated with IL-1β, Avn-C, and SB203580 or SP600125. Transcriptional levels of Mmp-3, -12, and -13 were determined by RT-PCR.

DISCUSSION

OA is typically characterized by progressive degradation and loss of articular cartilage resulting from the action of mechanical and biochemical factors (1, 22). Although various therapeutic approaches have been attempted to prevent OA progression, no effective treatment is currently available. The use of natural health product by OA patients to alleviate symptoms is rising globally. In this study, we evaluated the efficacy of Avn-C extracted from oats as a promising candidate to prevent OA progression. Here, we found that Avn-C prevented cartilage destruction in experimental OA mouse model induced by DMM surgery and inhibited the expression of MMPs, including MMP-3, -12, and -13, by regulating p38 kinase and JNK signaling.

Previous papers have reported that Avn-C exhibits anti-inflammatory effects by inhibiting the expression of pro-inflammatory cytokines via NF-κB signaling in skeletal muscle cells (17) and anti-oxidant effects on the inhibition of colon cancer cell growth (15, 18). However, in our current study, Avn-C had no effect on the expression of inflammatory mediators such as Ptgs2 or Nos2 during OA pathogenesis (Fig. 1B). Moreover, Avn-C did not affect NF-κB activation stimulated by IL-1β (Fig. 4B). However, Avn-C effectively inhibited Mmps and Adamts4 expression in OA cartilage.

The pro-inflammatory cytokine IL-1β is one of the most critical catabolic factors in the development of OA. It can induce the release of inflammatory mediators and MMPs. Excessive amount of IL-1β is found in cartilage, synovial fluid, the synovial membrane, and the subchondral bone (23). Herein, IL-1β was used to develop an in vitro OA inflammatory microenvironment. Results of the present study confirmed that IL-1β dramatically stimulated the expression of catabolic factors and inflammatory mediators, such as Il-6, Ptgs2, Nos2, Mmps, and Adamts that play crucial roles in the degeneration of articular cartilage. Of these genes, expression levels of Mmp-3, -9, -12, -13, and Adamts4 were significantly inhibited by Avn-C in a dose-dependent manner in IL-1β-mediated in vitro mouse OA-like pathogenic articular chondrocytes. In vivo results from mouse experimental OA model induced by DMM supported the inhibitory effect of Avn-C on OA progression by inhibiting the expression of MMPs such as MMP-3, -12, and -13.

OA is considered as a whole joint disease with pathological changes, including synovial inflammation, subchondral bone sclerosis, osteophyte formation, and cartilage destruction. Avn-C prevented pathological changes of the whole joint. We scored the grade of cartilage degeneration according to the OARSI system. DMM induced OA cartilage showed an average OARSI grade of 2.4 (P < 0.005), denoting increased subchondral sclerosis and osteophyte formation (Fig. 3B). However, Avn-C injection decreased the severity of OA phenotypes induced by DMM, which showed an OARSI grade 1.1 with thinner subchondral bone plate than in DMM control mice (Fig. 3B).

A previous study has demonstrated that HIF-2α as a critical transcription factor plays an essential role in IL-1β-induced MMP expression during OA pathogenesis (2). Our current results were consistent with results of the previous study showing induction of MMPs except MMP-2, -14, and -15 by IL-1β treatment or HIF-2α overexpression (2). The 5’-flanking regions of Mmp-2, -14, and -15 are distinct from those of other MMPs, although these genes have HIF-2α binding sites on their promoter regions. This indicates that Mmp-2, -14, and -15 are regulated by other factors rather than HIF-2α (24, 25). However, in the current study, inhibitory effect of Avn-C on IL-1β-induced MMPs expression did not seem to be mediated by HIF-2α because other catabolic factors such as Ptgs2 and Nos2 were not regulated by Avn-C (Fig. 1B). Anabolic factors such as Col2a1, Aggrecan, and Sox9 were not affected by Avn-C either (Fig. 1A), although these anabolic genes were suppressed by IL-1β. These results provide a possibility that Avn-C can inhibit certain downstream signaling pathways among multiple signaling pathways activated by IL-1β. Thus, we considered various signaling pathways involved in IL-1β induced MMP production. It is well-known that pro-inflammatory cytokines such as IL-1β, TNF-α, and IL-6 are responsible for the production of MMPs by upregulating NF-κB and MAP kinases (ERK, p38 kinase, and JNK) (23, 26, 27). Of them, p38 kinase and JNK signaling, but not ERK or NF-κB, were responsible for the inhibitory effect of Avn-C on IL-1β induced MMPs production in mouse articular chondrocytes. In addition, Avn-C enhanced effects of specific inhibitors of p38 kinase (SB203580) and JNK (SP600125). These results support that Avn-C can be used to relieve OA progression as a natural health product. Our current study demonstrates that Avn-C has potential to alleviate OA progression.  

MATERIALS AND METHODS

Experimental OA model by DMM surgery in mice

Cartilage destruction in mice was induced by DMM surgery. Sham-operated mice as controls were also prepared. At two weeks after DMM operation, mice were injected with 10 μl of 200 μM Avn-C or DMSO every week for six weeks by intra-articular injection. Knee joints were processed for histological analysis at 8 weeks after DMM surgery. All experiments were approved by Chonnam National University Animal Care and Use Committee.

Histological analysis

Mouse joint tissues were decalcified with 0.5 M EDTA (pH 8.0) for two weeks, embedded in paraffin, and sectioned at 5 μm in thickness. Cartilage destruction was evaluated by Safranin-O staining and was scored according to the OARSI grade system. Lateral sections were placed on slides for immunohistochemical staining and incubated in 3% H₂O₂ for 10 min to block endogenous peroxidase activity. Sections were then incubated with 0.1% trypsin for 30 min at 37°C to retrieve antigen. After blocking with 1% BSA for 30 min, slides were incubated with rabbit anti-MMP-3 (Abcam, Cambridge, UK), rabbit anti-MMP-12 (Epitomics, California, USA), and rabbit anti-MMP-13 (Abcam) followed by staining using the EnVision + System-HRP (Dako, Denmark) and AEC + substrate (Dako). Sections were counter-stained with Hematoxylin (Dako).

Primary culture of articular chondrocytes

Primary chondrocytes were isolated from femoral condyles and tibial plateaus of postnatal day 5 mice. The articular cartilage was pre-incubated for 2 h at 37°C with 0.2% trypsin and 0.2% type II collagenase and further digested with 0.2% type II collagenase for an additional 90 min. Chondrocytes were main-tained as a monolayer in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and appropriate antibiotics. On culture day 3, cells were treated as indicated for each experiment. Chondrocytes were treated with indicated concentration of IL-1β (GeneScript, Piscataway, NJ, USA) for 24 h with or without Avn-C (SI-033-053-1, Sigma-Aldrich, St Louis, MO, USA) dissolved in DMSO, SB203580 (p38 inhibitor), or SP6000125 (JNK inhibitor) (Santa Cruz, CA, USA).

RT-PCR and qRT-PCR

Total RNA was isolated from cultured articular chondrocytes using TRI reagent (TR118, Molecular Research Center Inc., Cincinnati, OH, USA). The RNA was reverse transcribed to cDNA. The resulting cDNA was amplified by PCR and quantified by qRT-PCR. qRT-PCR was performed using SYBR premix Ex Taq (RR420, TaKaRa, Japan). All qRT-PCR reactions were performed in duplicate and the threshold cycle value of each gene was normalized against that of glyceraldehyde-3 phosphate dehydrogenase (Gapdh) as an internal control. Primers and experimental condition are summarized in Supplementary Table 1.

Western blot analysis

Cultured mouse articular chondrocytes were washed with cold PBS and lysed in lysis buffer containing 50 mM Tris-HCl, pH8.0, 150 mM NaCl, 5 mM NaF, 1% NP-40, 0.2% SDS, 0.5% deoxycholate, a protease inhibitor cocktail, and a phosphatase inhibitor cocktail (Roche, Basel, Switzerland). Cell lysates were loaded and separated by SDS-PAGE. Proteins were transferred to nitrocellulose membranes. After blocking with 5% skim milk at room temperature for 1 h, membranes were incubated with primary antibodies overnight at 4°C with indicated antibodies. The following antibodies were used: rabbit anti-MMP-3 (Abcam), rabbit anti-MMP-12 (Epitomics), rabbit anti-MMP-13 (Abcam), mouse anti-ERK1 (BD Biosciences, NJ, USA), rabbit anti-phospho-Erk1/2 (Cell Signaling Technology, Danvers, MA, USA), rabbit anti-JNK (Cell Signaling Technology), mouse anti-phospho-JNK (Cell Signaling Technology), rabbit anti-p38 (Santa Cruz), mouse anti-phospho-p38 (Cell Signaling Technology), and rabbit anti-IκBα (Santa Cruz). Membranes were then incubated with horse-radish peroxidase conjugated secondary antibodies, anti-mouse (Sigma-Aldrich) or anti-rabbit IgG (Sigma-Aldrich), and detected using an ECL solution (GE Healthcare, Pittsburgh, PA, USA). Images were detected using ImageSaver6 software via an EZ-capture MG system (ATTO, Tokyo, Japan). Protein level was normalized against the level of β-actin (Sigma-Aldrich) as a loading control.

Enzyme-linked immunosorbent assay (ELISA)

To quantify activities of MMP-3, -12, and -13 in the culture medium, primary chondrocytes were cultured without serum. They were then treated with IL-1β and indicated dose of Avn-C. Supernatants were collected after 24 h and stored at −80°C until ELISA. ELISA kits #71130, #71157 and #71156 for analyzing substrate degrading activities of MMP-3, -12, and -13, respectively, were purchased from Anaspec (CA, USA). ELISAs were performed according to the manufacturer’s specifications using duplicate wells for each sample.

Statistical analysis

All experiments were repeated at least three times. Data obtained from qRT-PCR and enzymatic activity assays were first tested for conformation to a normal distribution using the Shapiro-Wilk test, followed by analysis with Student’s t-test (pairwise com-parisons) or analysis of variance with post-hoc tests (multiple comparisons) as appropriate. The n-value is the number of independent experiments or mice. Threshold for significance was set at the 0.05 level of probability (P < 0.05).