SNP, RV, dimethyl sulfoxide (DMSO), N-acetyl cysteine (NAC), carboxy-PTIO (PTIO), type II collagenase, L-ascorbic acid, Hoechst 33258, 4′,6-diamidino-2-phenylindole (DAPI) and insulin-transferrin-selenium were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco's modified Eagle's medium/nutrient mixture F-12 (DMEM/F12), trypsin, penicillin/streptomycin and fetal bovine serum (FBS) was purchased from Gibco (Carlsbad, CA, USA). Cell Counting Kit-8 (CCK-8) was purchased from Dojindo (Kumamoto, Japan). Caspase-3, -8, and -9 activity assay kits, 2′,7′-dichlorofluorescin diacetate (DCFH-DA), 3-amino,4-aminomethyl-2′,7′-difluorescein, diacetate (DAF-FM DA), Actin-Tracker Green and Tubulin-Trakcer Red were from purchased from Beyotime Institute of Biotechnology (Jiangsu, China). Annexin V-FITC/propidium iodide (PI) apoptosis detection kit and tetramethylrhodamine methyl ester (TMRM) was purchased from Invitrogen Life Technologies (Carlsbad, CA, USA). In situ cell death detection kit was purchased from Roche Diagnostics (Basel, Switzerland).

SNP was dissolved with phosphate-buffered saline (PBS) and NAC was dissolved with ultrapure water just before experiment. RV and PTIO were dissolved with DMSO. It is ensured that the working concentration of DMSO in medium was <1% throughout all experiments. The treatment time and concentration of RV and SNP were determined by our experiments. The pretreatment time of PTIO and NAC is 4 h, while their concentration is respectively 100 μM and 2 mM.

This present study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Our experiment protocol was approved by the Animal Care and Experiment Committee of Shanghai Jiao Tong University School of Medicine. Ten 3-month-old Sprague-Dawley rats were provided by Experimental Animal Center of Shanghai Ninth People's Hospital for in vitro experiments. NP cell isolation and culture were carried out as we previously described (35). The second-generation NP cells were adopted in the following experiments. NP cells were seeded into 96-well plates (5×10³/well for cell viability assay) or 6-well plates (1×10⁵/well for apoptosis assay, caspases activity assay, intracellular ROS and NO measurement and mitochondrial membrane potential assay) for at least 24 h before treatment of any reagent.

According to manufacturer's instructions, cell viability was measured with CCK-8 (Dojindo) using a microplate reader (Omega Bio-Tek, Inc., Norcross, GA, USA). All cell viability experiments were performed 6 times. Flow cytometry (FCM) was carried out for analysis of apoptosis rates with Annexin V-FITC/PI apoptosis detection kit following instructions, while 10,000 NP cells were collected for each FCM analysis. Apoptosis rates were calculated as Q2 (Annexin V-FITC-positive and PI-positive) + Q3 (Annexin V-FITC-positive and PI-negative). NP cells were stained with 0.5 μg/ml Hoechst 33258 for 20 min in dark and then imaged by a fluorescent microscope (IX71; Olmypus, Tokyo, Japan). Caspase-3, -8 and -9 activities were detected with caspase-3, -8, and -9 activity assay kits following manufacturer's instructions using a microplate reader (Omega Bio-Tek, Inc.). The caspase activity levels were expressed as relative activity with control as standard.

Mitochondrial membrane potential was measured with TMRM staining. NP cells were incubated with 100 nM TMRM at 37°C in dark for 30 min, washed with PBS 3 times and covered with fresh medium. NP cells were subsequently imaged using microscope (IX71, Olmypus). The excitation wavelength for TMRM was 549 nm. Three images (×200) of each kind of treatment were obtained for quantitative analysis of fluorescence intensity of the TMRM using IPP version 6.0 software (Media Cybernetics, Bethesda, MD, USA). The fluorescence intensity of the TMRM was expressed as mean density in IPP version 6.0 software.

To measure intracellular ROS or NO level, NP cells was incubated with DCFH-DA (10 μM) for 30 min or with DAF-FM DA (5 μM) for 20 min in dark at 37°C. After washed with PBS 3 times, NP cells were collected and suspended in fresh medium. The fluorescence intensity was detected using a microplate reader (Omega Bio-Tek, Inc.). The excitation wavelengths for DCFH-DA and DAF-FM DA are 488 and 495 nm respectively. The experiments were repeated 3 times.

Imaging of cytoskeletal structure was carried out as previously described (29). After staining with Actin-Tracker Green and Tubulin-Trakcer Red, cytoskeletons (×400) were imaged using a fluorescent microscope (LEICA DM4000B; Leica Microsystems GmbH, Wetzlar, Germany). The excitation wavelengths for Actin-Tracker Green and Tubulin-Trakcer Red are 488 and 543 nm respectively. After cell treatment, NP cells (×200) were imaged under transmitted light illumination using microscope (IX71; Olmypus).

Rat disc organ culture was carried out as we previously described (35). Ex vivo, discs were pretreated with 100 μM RV for 24 h or 2 mM NAC for 12 h or 100 μM PTIO for 12 h, then treated with or without 1 mM SNP for 18 h. Then the harvested discs were fixed in 4% paraformaldehyde, and then decalcified with EDTA for 2 weeks. After embedding with paraffin, 5-μm thick serial mid-sagittal sections of discs were made for slides. Mid-sagittal sections of discs were analyzed for apoptosis using in situ cell death detection kit according to manufacturer's instructions. DAPI staining was conducted for indication of total cells. TUNEL-positive apoptotic cells and DAPI-positive total cells of NP area on mid-sagittal sections of discs were identified by fluorescent microscope (IX71; Olmypus), and apoptosis rate was calculated as the percentage of numbers of TUNEL-positive cells to the numbers of total cells using IPP version 6.0 software (Media Cybernetics). The quantitative analysis was performed on three ×200 fields/section (three sections/disc and three discs for each kinds of treatment).

All statistical data were expressed as mean ± standard deviation. Results were statistically analyzed by a one-way analysis of variance (ANOVA) with multiple comparisons using SPSS 19.0 (IBM, Inc., New York, NY, USA). P-values <0.05 were considered to indicate statistically significant difference.

As shown by CCK-8 assays, SNP induced NP cells cytotoxicity in a dose and time-dependent manner (Fig. 1A and B). Treatment concentration of SNP was set as 1 mM and treatment time of SNP was set as 6 h throughout the following experiments without indication. Treatment of RV for 24 h did not exert obvious cytotoxicity on NP cells when RV concentration was no higher than 200 μM (Fig. 1C). Treatment of 100 μM RV did not exert significant cytotoxicity on NP cells when treatment time was no longer than 24 h (Fig. 1D). Pretreatment of RV for 18 h showed dose-dependent protective effects on SNP induced cytotoxicity (Fig. 1E). Pretreatment of 100 μM RV protected NP cells from SNP induced cytotoxicity in a time-dependent manner (Fig. 1F). Treatment concentration of RV was set as 100 μM and treatment time of RV was set as 18 h throughout the following experiments without indication.

In order to investigate the mechanism of protective effects of RV, NAC as an established ROS scanvenger and PTIO as an established NO scanvenger were adopted in the following experiments. FCM assay with Annexin V-FITC/PI staining showed that SNP induced significant NP cell apoptosis, which was significantly rescued by RV and NAC instead of PTIO (Fig. 2A and B). CCK-8 assays demonstrated that RV and NAC instead of PTIO significantly suppressed SNP induced cytotoxicity, while NAC or PTIO alone did not show obvious cytotoxicity (Fig. 2C). Meanwhile, RV and NAC instead of PTIO markedly inhibited activation of caspase-3 and -9 induced by SNP (Fig. 2D). However, SNP did not significantly activate caspase-8 (Fig. 2D). Hoechst 33258 staining images showed that RV and NAC instead of PTIO effectively decreased apoptotic nuclei containing condensed or fragmented chromatin induced by SNP, which is consistent with FCM assay with Annexin V-FITC/PI staining (Fig. 2E). As to mitochondrial membrane potential, TMRM staining and quantitative analysis demonstrated that significant loss of ΔΨm was caused by SNP and effectively rescued by RV and NAC but not PTIO (Fig. 3A and B). As shown by intracellular ROS and NO analysis, SNP significantly induced production of intracellular ROS (Fig. 3C) and NO (Fig. 3D). Pretreatment with RV obviously suppressed SNP induced excessive intracellular ROS production, but did not suppress SNP induced NO production (Fig. 3C and D). As expected, NAC significantly decreased the SNP induced high intracellular ROS and NO production, while PTIO significantly decreased the SNP induced high intracellular NO production but not ROS production.

Overall, ROS, not NO, production is probably the key mechanism of SNP induced NP cell apoptosis. RV can effectively scavenge intracellular ROS instead of NO, subsequently protecting NP cells from SNP induced apoptosis.

Actin-Tracker Green and Tubulin-Trakcer Red were used to mark F-actin and α-tubulin respectively under fluorescent microscope. In control and RV treated NP cells, normal cytoskeleton was observed with regularly distributed F-actin filament and microtubule (Fig. 4A). SNP treatment caused cytoskeleton shrinkage, by curling up F-actin filaments and disrupting microtubule structure (Fig. 4A). The SNP induced cytoskeleton shrinkage was significantly rescued by RV and NAC instead of PTIO (Fig. 4A). Under microscope with transmitted light illumination, SNP induced obvious cell shrinkage compared to control, which was significantly prevented by RV and NAC but not PTIO (Fig. 4B).

Organ culture was also carried out to analyze effects of RV on NP cells ex vivo. Treatment of SNP ex vivo significantly increased the TUNEL-positive cells percentage of NP area on mid-sagittal sections of discs compared to control (Fig. 5A and B). Consistent with in vitro results, RV and NAC significantly decreased TUNEL-positive cells percentage on mid-sagittal sections of discs compared to SNP treated group, while PTIO exerted no significant effects (Fig. 5A and B).