Upregulation of NOXA by 10-Hydroxycamptothecin plays a key role in inducing fibroblasts apoptosis and reducing epidural fibrosis

Reagent

HCPT was purchased from Aladdin Biotechnology Co., Ltd (Shanghai, China). The purity of 10-hydroxycamptothecin was 98%.

Cell culture and treatment

Fibroblasts were obtained from epidural scar tissue isolated from rats that underwent reoperation laminectomies. The cells were cultured at 37°C under 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, Grand Island, NY), containing 15% foetal bovine serum (FBS; Gibco), 0.1 U/L penicillin and 50 µg/ml streptomycin (Gibco, CA, USA). Cells in the exponential growth phase between passages 3 and 6 were used for all experiments. The fibroblasts were seeded onto various dishes and cultured overnight until they reached approximately 60–80% density, and then the cells were washed with phosphate buffered saline and treated with HCPT at various concentrations (0, 1, 2, 4 µg/ml) and for various durations (0, 24, 48, 72 h).

Cell lentiviral infection

Lentiviral infection was used to achieve gene silencing. The target gene NOXA was contained in the lentiviral vectors, which were purchased from Shanghai Genechem Co., Ltd. (Genechem, China). Fibroblasts were treated according to the instructions. Then, successfully transfected cells were used in the experiments evaluating the treatment of HCPT, Western blot analysis and TUNEL assay.

Western blot analysis

Following treatment, the fibroblasts were harvested and lysed in RIPA buffer (Beyotime, Hangzhou, China) on ice. After the lysates were centrifuged in 4°C at 13,000 ×  g for 10 min, the supernatants were collected for Western blot analysis. The protein concentration was determined by the BCA Protein Assay Kit (Beyotime, Hangzhou, China). The proteins were separated by 6%–12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene difluoride membranes (Millipore, Bedford, MA). Following blocking with 5% skimmed milk in TBST for 2 h, the membranes were incubated with the appropriate primary and secondary antibodies successively according to the instructions. The primary antibodies used were anti-NOXA, anti-cleaved-caspase3, anti-cleaved-poly ADP-ribose polymerase (cleaved PARP), anti-caspase3, anti-PARP, anti-Bax, anti-Bcl-2 and anti-β-actin antibodies (Cell Signaling Technology, Beverly, MA, USA). The anti-mouse or anti-rabbit IgG were also purchased from Cell Signaling Technology.

TUNEL assay in fibroblasts

A TUNEL assay was performed to detect the apoptotic effect of HCPT on fibroblasts. The apoptotic rate of fibroblasts was detected using the TdT-mediated dUTP-biotin nick-end labelling (TUNEL) test system (KeyGEN, Nanjing, China). All of the operating steps were according to the manufacturer’s instructions. Following a brief staining procedure, the features of apoptosis were evaluated via a fluorescence microscope. TUNEL-stained fibroblasts were believed to be apoptotic, and the total number of fibroblasts was counted by the DAPI-staining method.

Animals

In all, 72 Sprague-Dawley young adult male rats (mean weight of 280 g), purchased from the experimental animal centre of Yangzhou University (Yangzhou, China) were used for this study. All rats received care in compliance with the principles of Laboratory Animal Care according to international recommendations, and the study was approved by the Animal Care and Research Committee of the Yangzhou University, China. The animals were randomly divided into four groups (18 rats per group) as follows: HCPT (0.2 mg/ml), HCPT (0.1 mg/ml), HCPT (0.05 mg/ml) or control (saline). The rats were allowed to acclimate to the environment for 1 week before the experiment.

Animal model and topical application of drugs

Laminectomy models were performed according to a previous study (Sun et al., 2007). Briefly, following satisfactory anaesthesia by intraperitoneal injection of 1% pentobarbital sodium (40 mg/kg body weight), the fur around the location of L1 and L2 was shaved and the exposed skin was sterilized. After exposing the fascia and the paraspinal muscles by a midline skin incision, the L1 vertebral plate was removed by rongeur forceps. Then, complete 5 × 2 mm areas of dura mater were exposed. All rats underwent a total L1 laminectomy.

After disinfection and haemostasis of the lumbar region, HCPT concentrations of 0.2, 0.1 and 0.05 mg/ml or saline were applied to the laminectomy areas with cotton pads (4 × 4 mm) for 5 min. The surrounding tissues were covered by wet gauze to avoid touching the agent. After the cotton pads were removed, the decorticated areas of the laminectomy site were immediately irrigated with saline to remove the remaining HCPT. After the above operations, the wounds were closed in layers.

Macroscopic assessment of epidural fibrosis

For macroscopic evaluation, six rats were randomly selected from each group 4 weeks after laminectomy. All rats were sacrificed by an overdose of 1% pentobarbital sodium via intravenous administration. Then, the laminectomy sites were reopened, double-blind trials were used to assess epidural fibrosis, and the amount of fibrosis was judged based on the Rydell classification (Rydell, 1970): grade 0, no adhesions were apparent around the dura mater; grade 1, weak adhesions appeared around the dura mater but were easily dissected; grade 2, moderate adhesions appeared around the dura mater and could be dissected with difficulty without disrupting the dura matter; grade 3, dense fibrous adhesions were firmly adherent to the dura mater and could not be dissected.

Hydroxyproline content (HPC) analysis

HPC analysis was performed to assess the amount of collagen in the scar tissue. After macroscopic evaluation, approximately 5 mg of wet-weight scar tissue was obtained from the decorticated areas for HPC analysis according to a previous study (Yan et al., 2010). Briefly, the samples were lyophilised, ground and hydrolysed with 6 mol/l HCl at 130°C for 12 h. Then, 1 ml hydroxyproline developer (β-dimethylaminobenzaldehyde solution) was added to the samples and standards. The absorbances were evaluated at 558 nm using a spectrophotometer. In the end, the HPC/mg scar tissue was calculated according to the standard curve constructed with serial concentrations of commercial hydroxyproline.

Histological analysis

Four weeks after laminectomy, another six rats were picked randomly, and histological analysis was performed. After intraperitoneal injection of 1% pentobarbital sodium (40 mg/kg body weight), all rats received intracardial perfusion with 4% paraformaldehyde. The whole L1 spinal column, including the surrounding muscles and epidural fibrotic tissue, was removed and immersed in 4% paraformaldehyde. After four days of decalcification by 10% buffered formalin, each specimen was further decalcified in an ethylenediamine tetraacetic acid (EDTA) and glycerol solution for 30 days, and then embedded in paraffin again. Successive 4-µm transverse sections were made through the L1 vertebra from the top to the bottom. Twelve continuous transversal sections of 4 µm each from the top to the bottom of the upright L1 vertebra were made. Six odd sections were stained by means of haematoxylin and eosin (HE), and the epidural fibrosis was assessed by light microscopy using a 40×  objective. At a magnification of 400× and with three fields of the laminectomy sites in each section, the fibroblast density was calculated by using Image Pro Plus 6.0.

Scoring system of epidural fibrosis and epidural cell density

To perform impersonal histopathological evaluation, an epidural fibrosis and epidural cell density scoring system was used. The degree of epidural fibrosis (HE × 40) and epidural cell density (HE × 400) was graded according to previous studies (He, Revel & Loty, 1995; Yildiz et al., 2007). The scores of epidural fibrosis (HE × 40) were as follows: Grade 0, there was no fibrosis scar tissue around the dura; Grade 1, thin or weak fibrous bands were found around the dura; Grade 2, continuous adherence was observed in less than two-thirds of the laminectomy defect; and Grade 3, the fibrotic scar tissue adherence is extensive and firm, and more than two-thirds of the laminectomy defect were adherent with fibrous bands, or the nerve roots were also adherent with the fibrosis scar tissue. The epidural cell density (HE × 400) was graded as follows: Grade 1, no more than 100 fibroblasts under 400× magnified visual field; Grade 2, the count of fibroblasts was between 100 to 150 under 400× magnified visual field; and Grade 3, over 150 fibroblasts per 400× field.

TUNEL assay in fibroblasts of epidural tissue

Fibroblast apoptosis in the epidural tissue sections was also identified using the TdT-mediated dUTP-biotin nick-end labelling (TUNEL) test system (KeyGEN, Nanjing, China) according to the manufacturer’s instructions. After staining, the apoptotic fibroblasts were detected under fluorescence microscopy. The final images were merged and analysed by Image Pro Plus 6.0.

Immunohistochemical analysis

The six remaining rats from each group were utilised, and the acquisition of sections was according to the histological analysis above. Following deparaffinization and rehydration through gradient ethanol solutions, the sections were incubated in citrate buffer to activate the antigenicity, and then they were exposed to 3% H2O2 to block endogenous peroxidase activity. The sections were washed with PBS three times, and then the primary antibody (anti-NOXA) was added and incubated at 4°C overnight. The sections were then washed with PBS and incubated with the secondary antibody (biotin-labelled goat anti-immunoglobulin G) at room temperature for 1.5 h. Following treatment with 3,3’-diaminobenzidine solution for 5 min at room temperature, the nuclei were counterstained with haematoxylin for 3 min. The samples were observed under a light microscope.

Statistical analysis

The data from our experiments were analysed using SPSS 19.0 statistical software. All data are expressed as the mean ± standard deviation (SD). Statistical significance was defined as a P value < 0.05.

HCPT induced apoptotic cell death in fibroblasts

To determine the apoptotic effect of HCPT in rat fibroblasts, we treated the fibroblasts with various concentrations (0–4 µg/ml) of HCPT for 24 h. As shown in Fig. 1A, the results from Western blot analysis revealed that HCPT could increase the expression of cell apoptosis markers such as cleaved PARP and Bax, while it decreased the expression of Bcl-2, which was considered as an anti-apoptotic marker. Moreover, we found that the effect of HCPT on these markers was dose-dependent. To further confirm the apoptotic effect of HCPT on fibroblasts, morphological examinations were performed. As shown in Fig. 1B, few TUNEL-positive cells were detected in the control group (1.86% ± 1.85%). Following HCPT treatment, the percentages of TUNEL-positive cells at1 µg/ml, 2 µg/ml and 4 µg/ml were 14.94% ± 1.40%, 20.06% ± 2.64% and 28.26% ± 2.64%, respectively (Fig. 1C). Taken together, these results indicate that HCPT significantly induced apoptosis in fibroblasts.

HCPT increased NOXA expression in fibroblasts

To confirm whether HCPT affected NOXA expression in fibroblasts, the fibroblasts were treated with 2 µg/ml HCPT for 24 h, 48 h and 72 h. Following HCPT treatment, the Western blot analysis showed that HCPT could increase NOXA expression in a time-dependent manner. What’s more, the expression of cell apoptosis markers such as cleaved-caspase3, cleaved PARP and Bax was also increased with the increased expression of NOXA (Fig. 2). The result of the Western blot analysis showed that the application of HCPT could upregulate NOXA expression in fibroblasts and could promote fibroblast apoptosis.

The effect of NOXA on fibroblast apoptosis

To detect the exact effect of NOXA in fibroblast apoptosis, we used lentiviral infection to downregulate NOXA expression to confirm that NOXA is required for HCPT-induced apoptosis in fibroblasts. The levels of NOXA expression revealed from Western blot analysis showed that we knocked down NOXA in the fibroblast lines successfully. NOXA knockdown increased the expression of Bcl-2 and decreased the expression of cleaved PARP and Bax (Fig. 3A). Also, in accordance with the results above, the rate of fibroblast apoptosis that was detected by TUNEL assay was decreased in the NOXA knockdown cells. What is more, the increased expression of NOXA, cleaved PARP and Bax indicated by Western blot analysis and the increased apoptotic rate detected by TUNEL assay, which occurred after HCPT treatment, were partially decreased by NOXA knockdown (Figs. 3B and 3C). Taken together, these results indicate that NOXA played a crucial role in HCPT-induced fibroblast apoptosis.

Macroscopic evaluation of epidural fibrosis

To explore the effect of HCPT on epidural fibrosis, laminectomy models were performed and evaluated. The results of epidural fibrosis were detected via macroscopic observation and graded according to the Rydell classification; the analysis showed that Rydell grade 3 existed in all of the control group rats. Grade 0, 1 and 2 were found in the HCPT-treated groups (Table 1). In the control group, thick and extensive adhesions were found around the dura mater, and it was difficult to remove the adhesions. In the 0.05 and 0.1 mg/ml HCPT-treated groups, weak or moderate adhesions were found around the dura mater, and the adhesions were easily dissected. However, in the 0.2 mg/ml HCPT-treated group, there were few adhesions around the dura mater, and the adhesions were easily removed without bleeding.

Hydroxyproline content (HPC) analysis

HPC analysis was used to detect the amount of collagen in the scar tissue. As shown in Fig. 4, the HPC in the HCPT-treated groups of 0.2 mg/ml, 0.1 mg/ml, 0.05 mg/ml, and the saline-treated group, were 25.78 ± 1.96, 31.21 ± 1.93, 42.82 ± 3.44 and 55.05 ± 4.43 µg/mg, respectively. The HPC levels in the HCPT-treated groups were less than that in the control group (P < 0.05). And the decrease of HPC was in a dose-dependent manner.

The effect of HCPT on epidural fibrosis in rats

To detect the effect of HCPT on reducing epidural fibrosis, histopathological evaluation was performed according to the scoring system of epidural fibrosis and epidural cell density. As shown in Figs. 5 and 6, thick epidural fibrosis with extensive adhesions to dura mater was found in the laminectomy sites of the control group; also, a particularly large number of fibroblasts were found in the scar tissue. Moderate epidural fibrosis was observed and fibroblasts were decreased around the laminectomy sites in the 0.05 mg/ml and 0.1 mg/ml HCPT-treated group compared with the control group. In contrast, little epidural fibrosis with fewer fibroblasts was found in the laminectomy sites of the 0.2 mg/ml HCPT-treated group. All of these results showed that the degree of epidural fibrosis in the HCPT-treated group occurred in a dose-dependent manner (Fig. 7).

HCPT induced fibroblast apoptosis of rats

To confirm whether HCPT could induce apoptosis in rat fibroblasts, the TUNEL assay was used. As shown in Fig. 8, few TUNEL-positive fibroblasts were found in the control group. After HCPT treatment, the number of TUNEL-positive fibroblasts was increased, which occurred in a dose-dependent manner.

The effect of HCPT on NOXA expression in epidural tissue in rats

To confirm whether HCPT could affect NOXA expression in epidural tissue in rats, immunohistochemical staining was performed. As shown in Fig. 9, we found that HCPT could increase NOXA expression in the epidural tissue in rats. Moreover, with the increased concentration of HCPT, the expression of NOXA also gradually increased. All of these results suggested that HCPT treatment could increase NOXA expression in the epidural tissue of rats.