Resveratrol enhances matrix biosynthesis of nucleus pulposus cells through activating autophagy via the PI3K/Akt pathway under oxidative damage

The decrease in nucleus pulposus (NP) matrix production is a classic feature during disc degeneration. Resveratrol (RSV) is reported to play protective effects under many pathological factors.The present study aims to study the effects of RSV on NP matrix homeostasis under oxidative damage and the potential mechanism. Rat NP cells were exposed to H2O2 solution to create an oxidative damage. RSV and the 3-methyladenine (3-MA) were added along with the culture medium to respectively investigate the role of RSV and cellular autophagy. NP matrix synthesis was evaluated by the expression of macromolecules (aggrecan and collagen II) and glycosaminoglycan (GAG) content. Activation of cellular autophagy was assessed by the expression of several molecular markers. Additionally, activity of the PI3K/Akt pathway was also evaluated to study its potential role. Compared with the control group (NP cells treated with H2O2), RSV significantly up-regulated expression of matrix macromolecules (aggrecan and collagen), promoted GAG production, and increased the expression of autophagy-related markers (Beclin-1 and LC-3). Further analysis showed that inhibition of autophagy by 3-MA partly attenuated NP matrix production. Additionally, RSV increased activity of the PI3K/Akt pathway compared with the control NP cells, but it was not affected by the addition of 3-MA. RSV plays a protective role in enhancing NP matrix synthesis under oxidative damage. Mechanistically, activation of the cellular autophagy via the PI3K/Akt pathway may participate in this process. RSV may be an effective drug to attenuate oxidative stress-induced disc degeneration.


Introduction
Intervertebral disc (IVD) degeneration is the major cause of neck or back pain syndrome, which contributes to the drop in life quality and even disability in adults [1]. This disease has a widespread prevalence with an estimated 80% of adults are known to experience low back pain (LBP) at least once in their lifetime [2]. The pathogenesis of disc degeneration contains a complex signaling network and various key molecules [3,4]. Currently, the pathogenesis of disc degeneration has not been clearly elucidated. Therefore, a deeper understanding about the molecular mechanism of disc degeneration will contribute to the development of new treatments for preventing and retarding disc degeneration.
The IVD, a heterogeneous and fibrocartilaginous tissue, is the biggest avascular tissue within the body. It contains three interdependent and structurally connected parts: cartilaginous  endplates (CEPs), annulus fibrosus (AF), and nucleus pulposus (NP). Specifically, the central NP is a gelatinous matrix containing abundant water, proteoglycan, and collagen II [5]. During disc degeneration, the extracellular matrix (ECM) within the NP region was degraded and limited to renew [6]. These degenerative changes in NP matrix ultimately frustrate disc's biomechanical properties and thus affect spinal stability [7]. Therefore, the maintenance of a healthy matrix homeostasis is a key aspect in retarding disc degeneration. Autophagy, a necessary cellular self-eating process, is one of the most important mechanisms for the maintenance of cellular homeostasis by degrading cytosolic macromolecules and damaged organelles under normal physiological and/or pathological conditions [8]. Recently, autophagy has become a research hotspot. Increasing evidence has showed that autophagy dysfunction participates in many degenerative diseases, such as osteoarthritis [9] and neurodegeneration [10]. Previously, a low basal level of autophagy was observed in the NP cells from non-degenerative adults rats, partly revealing the involvement of autophagy in normal disc NP cell biology [11]. However, autophagy is significantly increased in the degenerative rat NP cells [12,13]. Oppositely, Jiang et al. [14] has reported that there are decreased autophagosome, down-regulated expression of Beclin-1, and decreased ratio of LC3-II to LC3-I in NP cells of patients with degenerative discs compared with those in NP cells of patients with non-degenerative discs. Collectively, these reports show sufficient evidence for the existence of autophagy in degenerative disc NP cells at either higher or lower levels, indicating that autophagy may play different roles under the stimulation of different pathological factors in disc NP cells.
Resveratrol (RSV) is a phytoalexin identified in peanuts, grapes, and other plants. Previous studies demonstrated that RSV has a wide range of protective effects in different cell types, such as anti-inflammatory, anti-ageing, anticancer, and cartilage protection [15]. Several studies have shown sufficient rationale for the potential therapeutic use of RSV in retarding and/or regenerating disc degeneration in terms of inhibiting disc cell apoptosis, attenuating disc cell senescence, and decreasing the expression of matrix degradation enzymes [16][17][18][19][20]. As the initiation and progression of disc degeneration are closely associated with oxidative stress [21], the present study was to assess whether RSV has any inhibitory effects on oxidative damage-induced decrease in NP matrix synthesis.

Rat NP cell isolation and culture
All experiments in the present study were approved by the Ethics Committee at The Chinese Medicine Hospital of Liaocheng City, Shandong Traditional Chinese Medicine University [SLK(LU) 2011-0121]. NP cells were isolated from rat lumbar discs (L1-L6) according to a method described in a previous study [22]. Specifically, 32 healthy Sprague-Dawley rats (female, 4-6 weeks old, and 250-270 g weight) were killed by excessive CO 2 inhalation. Then, the spinal column was separated under aseptic conditions and the individual IVDs were collected. After the central gelatinous NP tissue was removed, it was digested with 0.1% collagenase for 2 h at 37 • C, with intermittent shaking every 30 min. Then, the separated NP cells and partially digested tissues were incubated with complete DMEM/F12 culture medium (Gibco, U.S.A.) supplemented with 10% FBS (Gibco, U.S.A.) under standard culture conditions (37 • C, 21% O 2 , and 5% CO 2 ). When the NP cells grew to 80-90% confluence, they were dispersed using 0.25% trypsin (Gibco, U.S.A.) and then re-suspended in appropriate culture plates. In the present study, the second-passage NP cells were used in all experiments.

Design of experiment groups
To reach the study objective, three groups was designed: (i) Control NP cells treated with H 2 O 2 , (ii) NP cells treated with H 2 O 2 and RSV, and (iii) NP cells treated with H 2 O 2 , RSV, and 3-methyladenine (3-MA). All NP cells in each group were treated with test compounds for 48 h. Here, the oxidative damage but not the cytotoxicity was created using the H 2 O 2 (100 μg/ml) according to a recent study [19]. All NP cells were treated with the base level of H 2 O 2 (100 μg/ml). RSV (50 μM) was added along with the culture medium to investigate its protective effects. Additionally, 3-MA (5 mM) was added along with the culture medium to investigate the role of autophagy in this process.

RNA extraction and real-time PCR analysis
Total RNA from the cultured NP cells were isolated using TRIzol reagent (Invitrogen, U.S.A.) according to the manufacturer's instructions. The single-stranded cDNA templates were synthesized using the PrimeScript TM II First Strand cDNA synthesis kit (TaKaRa, Japan). Then, real-time PCR (3 min at 95 • C, followed by 35 amplification cycles of 10 s at 95 • C, 12 s at 56 • C, and 10 s at 72 • C) was performed on a real-time PCR machine (Thermo Scientific, Rockford, U.S.A.) with the method of SYBR Green detection chemistry (TOYOBO, Japan). The primers were shown in the Table  1. β-actin was used to normalize the expression of target genes. Gene expression was calculated with the method of 2 C t .

Western blot assay
Western blot assay was carried out based on the standard procedure. Briefly, after the cultured NP cells were lysed using the RIPA lysis solution (Beyotime, China), the lysates were separated on an SDS/PAGE and transferred on to the PVDF membrane (Beyotime, China). The PVDF membranes were blocked with 5% BSA for 1 h at room temperature. Then, they were incubated with primary antibodies (β-actin: Abcam, ab8226; aggrecan: Norvus, NB600-504; collagen II: Norvus, NB600-844; Beclin-1: Abcam, ab207612; LC3: Abcam, ab48394; Akt: Cell Signaling Technology, #4691; p-Akt: Cell Signaling Technology, #4060; all diluted at 1:1000) overnight at 4 • C and the HRP-conjugated secondary antibodies (goat anti-mouse IgG and goat anti-rabbit IgG, diluted 1:200, Beyotime, China) for 2 h at room temperature. Subsequently, the PVDF membranes were treated with ECL Plus (Thermo, U.S.A.) according to the manufacturer's instructions. Blot intensity was calculated by densitometric analysis using the ImageJ software (National Institutes of Health, U.S.A.). Protein expression of target molecules was normalized to that of β-actin.

Glycosaminoglycan content measurement
Glycosaminoglycan (GAG) content of the cultured NP cells was quantitated by 1,9-dimethyl methylene blue (DMMB) assay as described in a previous study [23]. Briefly, the cultured NP cells were collected and digested by 5 mg/ml papain (Sangon, Biotech Co., Ltd., China) for 6 h at 60 • C. Then, the GAG content in the digested solution was calculated according to the absorbance value at 525 nm. Additionally, the shark cartilage chondroitin sulphate (Sigma, U.S.A.) was used as a standard in the DMMB assay.

Statistical analysis
Each experiment was performed in triplicate. Results are expressed as mean + − S.E.M. and the statistical analysis was performed using SPSS 17.0 software. Intergroup difference was analyzed by the one-way ANOVA. The post hoc test was determined by the LSD test. A statistical significance was regarded when P<0.05.

Real-time PCR and Western blot analysis of matrix macromolecules
Gene expression and protein expression of NP matrix macromolecules (aggrecan and collagen II) were analyzed. Results showed that expression of aggrecan and collagen II in the RSV group was significantly up-regulated at both gene and protein levels compared with the control group. When the cellular autophagy was inhibited by the 3-MA, their expression levels were partly decreased ( Figure 1).

Real-time PCR and Western blot analysis of autophagy-related molecules
Similarly, gene expression and protein expression of cellular autophagy-related markers (Beclin-1, LC3) were analyzed to investigate the role of autophagy. Results showed that expression of these two autophagy-related markers in the RSV group was significantly increased at both gene and protein levels compared with the control group. However, the inhibitor 3-MA partly decreased their expression when it was added along with the culture medium in the RSV-treated NP cells (Figure 2).

ICC analysis of matrix protein deposition
We analyzed matrix protein deposition using the method of immunocytochemical (ICC). Results showed that the staining intensity of these matrix proteins in the RSV group was higher than that in the control group, whereas the inhibitor 3-MA obviously attenuated the positive effects of RSV on matrix protein deposition in the RSV + 3-MA group (Figure 3).

GAG content
GAG is one of main matrix composite within the NP tissue. Results showed that though RSV increased GAG content compared with the control NP cells, it can be partly inhibited by the inhibition of cellular autophagy (Figure 4).

Activity of the PI3K/Akt pathway
To study whether the PI3K/Akt pathway is involved in this process, we evaluated the activity of this pathway by Western blot assay. Results showed that p-Akt expression in the RSV group was higher than that in the control group, however, p-Akt expression was not significantly affected by the addition of 3-MA in the RSV + 3-MA group ( Figure  5).

Discussion
Disc degeneration is a main contributor to LBP [1]. Oxidative stress is known to be closely correlated with disc pathogenesis and progression [21]. Until now, the precise role of cellular autophagy during disc degeneration is controversial [24]. Previous studies have demonstrated the protective effects against oxidative damage in disc tissue and other tissues [15]. However, whether RSV has protective effects against oxidative damage-induced NP matrix decrease is unclear. In the present study, we reported for the first time that RSV enhanced NP matrix biosynthesis under oxidative damage through activating autophagy via the PI3K/Akt pathway. The increase in NP matrix degradation or decrease in NP matrix synthesis is one of the most common features during disc degeneration. Aggrecan and collagen II are the two macromolecules within the disc NP matrix [25]. An adequate NP matrix production is implicated to maintain the disc's mechanical property and thus the spinal stability [26]. It has been well established that oxidative stress is harmful to disc cell's normal biology, including matrix synthesis and cell viability [21]. In the present study, we used the H 2 O 2 to create an oxidative but not cytotoxic environment according to a recent method [19]. However, rat NP tissue contains lots of notochordal cells. There are no accurate molecules to distinguish NP cell from notochordal cells. Therefore, care should be taken when using our results to explain disc degeneration in adults.
Autophagy is proved to be a self-protective process that can maintain homeostasis under certain disadvantageous conditions [27]. Autophagy and oxidative stress are closely related since the former is known to play a protective role against the later [28]. Recently, many studies have demonstrated that some pathological factors, including mechanical loading, high oxygen tension, high glucose, and inflammatory cytokines, are able to induce oxidative stress reaction and thus contribute to catabolic metabolism of disc matrix [29]. A previous study has showed that the rapamycin (an autophagy activator) decreased the level of MMP-3, MMP-9, and ADAMTS-4, whereas the 3-MA (an autophagy inhibitor) produced an opposite effect in the NP cells treated with inflammatory cytokines (TNF-α and IL-1β) [30]. Thus, autophagy may prevent the catabolic metabolism of ECM to play a protective role in retarding disc degeneration.
RSV is a natural activator of cellular autophagy and is often studied in recent studies [18,19]. Studies have reported that RSV attenuates oxidative stress injury in other cells by regulating nuclear factor-κB (NF-κB) and mitogen-activated protein kinase (MAPK) signaling pathways [31,32]. In the present study, we found that RSV enhanced the expression of autophagy-related markers (Beclin-1 and LC-3) and expression of matrix macromolecules (aggrecan and collagen II), and that activity of the PI3K/Akt pathway showed a similar trend. However, we found that when the autophagy was inhibited by the 3-MA, expression of matrix macromolecules (aggrecan and collagen II) decreased whereas the activity of the PI3K/Akt pathway was not changed. Based on our own results and previous reports [8] about the positive effects of autophagy on NP matrix metabolism, it can be deduced that RSV can enhance matrix synthesis through activating cellular autophagy via the PI3K/Akt pathway. In line with us, a previous study showed that RSV-induced autophagy is regulated by the PI3K/Akt/mTOR pathway in prolactinoma cells [33]. However, the present study did not further verify the role of the PI3K/Akt pathway using item specific inhibitor in the RSV group. This is another limitation of the present study.
In conclusion, the present study assessed whether RSV has any inhibitory effects on oxidative damage-induced decrease in NP matrix synthesis. Our results demonstrated that RSV enhanced matrix biosynthesis of NP cells through activating cellular autophagy via the PI3K/Akt pathway. The present study sheds a new light on the protective effects of RSV against oxidative damage and provides that RSV may be a promising drug to attenuate oxidative stress-induced disc degeneration.