Characterization of patients. The present study was approved by The Fifth Affiliated Hospital, Southern Medical University, Shanghai Tenth People’s Hospital, Tongji University and Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (approval no. XHEC-D-2015-112). The primary endpoint is LC3 mRNA relative expression level between OPLL and non-OPLL patients. According to our previous preliminary test (n=5 for each group), the relative expression of LC3 mRNA in the control group was 0.28±0.13 (mean±standard deviation). The relative mean expression in OPLL group was 0.51± 0.21. Based on a two-sided 5% significance level and a power of 90%, sample size was calculated using PASS 11 statistical software (version 11.0.8 for Windows 7). A sample size of 12 subjects in each group was necessary to detect a difference expression of LC3 mRNA between OPLL and control group. From December 2015 to December 2018, X-ray, CT images and MRI of the cervical spine were evaluated preoperatively in each patient. According to inclusion and exclusion criteria for patients referenced previous methods [19,20], actually, in this study, 21 cases of OPLL patients and 16 cases of non-OPLL patients were enrolled. All patients underwent anterior cervical decompression surgery and all surgical procedures were performed by the same five surgeons.
All patients provided written informed consent and spinal ligament tissues were obtained during surgery using intraoperative aseptic techniques. The OPLL patients with a radiographic diagnosis of OPLL involving the cervical spine, as well as symptoms, such as neck pain and numbness in the extremities. OPLL patients included 10 males and 11 females, with a mean age of 48 ± 5.9 years. Ossified lesions were distributed from C1 to C7, including 4 cases of continuous type, 2 cases of mixed type, 1 case of segmental type and 14 case of local type. By contrast, 16 non-OPLL patients with cervical trauma spinal fracture were enrolled, who did not have OPLL, cervical spondylosis or stenosis. In non-OPLL group, the patients included 9 males and 7 females, with a mean age of 42.1 ± 13.1 years.
To investigate this hypothesis, autophagy was assessed by reverse transcription-quantitative (RT-q)PCR, western blotting and immunofluorescence in ligament clinical specimens and ligament fibroblasts. Subsequently, Pearson’s correlation was used to compare the expression of osteogenic-specific genes with autophagy. Besides, small interfering (si)RNA was used to knockdown the expression of Beclin1 in spinal ligament fibroblasts, following which the expression levels of osteocalcin (OCN), alkaline phosphatase, biomineralization associated (ALP) and collagen type 1 (COL 1) were compared in OPLL ligament fibroblasts and non-OPLL ligament fibroblasts.
Reagents and antibodies. Monodansylcadaverine (MDC) was purchased from Sigma-Aldrich; Merck KGaA. Polyclonal antibodies against Beclin1, and LC3 were obtained from Cell Signaling Technologies, Inc. β-actin, vimentin and keratin antibodies were obtained from Abcam.
Specimen processing and cell isolation. During the anterior cervical decompression surgery, 37 posterior longitudinal ligament specimens were obtained. To avoid contamination with osteoblasts or osteocytes, the ligaments were extracted carefully, cut up into 1 mm2 pieces and washed with PBS several times. Subsequently, the specimens were divided into two parts. One was stored in liquid nitrogen for RT-qPCR analysis. The remaining were washed with normal saline, plated in 35-mm culture dishes, and maintained in Dulbecco’s modified Eagle’s media supplemented with 10% fetal bovine serum (Gibco, USA). All assays were carried out on the fifth passage cell cultures.
Measurement of cell viability. Cell viability was examined using 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfo-phenyl)-2H-tetrazolium (Beyotime Institute of Biotechnology), according to the manufacturer’s protocols. Briefly, the ligament fibroblasts were plated in 96-well culture plates and cultured in the osteogenic differentiating medium for 1, 2, 3, 4, 5, 6, 7 or 8 days. Cell viability assays were performed and the absorbance of optical densities were measured at each time point and detected by a microplate spectrophotometer at 450 nm.
RT-qPCR. Total RNA was obtained from the posterior longitudinal ligament specimens or the ligament fibroblasts, according to our previous study design. Expression of Beclin1, LC3, ULK1, COL 1, OCN and ALP were examined by RT-qPCR, according to our previous method . For PCR amplification, specific oligonucleotide primers of rat sequences were designed on the basis of sequences in GenBank (Table 1).
Western blotting. The protein expression of Beclin1, LC3-II/I, β-actin and Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was detected by western blotting, according to our previous paper . Briefly, total protein was extracted using a western blot kit (Beyotime Institute of Biotechnology). Each sample (50 μg/well) was separated on a 10% sodium dodecyl sulfate-polyacrylamide gel for Beclin1 protein and 12% sodium dodecyl sulfate-polyacrylamide gel for LC3 protein. Next, separated proteins were transferred to polyvinylidene difluoride membranes, followed by blocking with 5% non-fat milk for 1 h. Next, the polyvinylidene difluoride membranes were incubated with anti-β-actin (Cat.#4970; CST, USA) (dilution 1:1000), GAPDH (Cat.#2118; CST, USA) (dilution 1:1000), anti-Beclin1(Cat.#3495; CST, USA) (dilution 1:500) or anti-LC3 antibodies (Cat. #83506; CST, USA) (dilution 1:200) and detected using the ECL detection kit (Beyotime Institute of Biotechnology). The blots were quantiﬁed by densitometry using Image Lab version 2.1 software (Bio-Rad Laboratories, Inc.).
Knockdown of Beclin1 in ligament fibroblasts. siRNAs specifically targeting Beclin1 were constructed and designed by superbiotek (Shanghai, China). To target Beclin1, the following sequences were used: Forward, 5’- GAGCGAUGGUAGUUCUGGAGG-3’ and reverse, 3’-UCCAGAACUACCAUCGCUCUG-5’. A missense siRNA vector (lack of complementary sequences) served as a non-silencing control (siControl). All transfections were carried out using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer’s instructions.
Levels of autophagy and apoptosis quantified by flow cytometry. The rates of apoptosis were analyzed by flow cytometry in the present study, according to our previous method . The rates of autophagy were analyzed by flow cytometry according to the method of Bursch et al  and Shen et al . Briefly, the cells were incubated with 0.05mM MDC (Cat. #30432; Sigma-Aldrich, USA) in PBS at 37°C for 10 minutes and then the intracellular MDC was measured by flow cytometry within 30 minutes. The autophagy incidence was determined as the percent of MDC positive cells automatically analyzed using FlowJo software (Tree Star, San Carlos, CA) and the unstained cells were used as control.
Immunofluorescence for vimentin and keratin. The ligament fibroblasts were seeded on 6-well plates with sterile glass cover slip, fixed with 4% paraformaldehyde and permeabilized in 0.1% Triton X-100 for 20 mins. Subsequently, the ligament fibroblasts were incubated with primary anti-vimentin (Cat.#5741; CST, USA) (dilution 1:200) or anti-keratin antibodies (Cat.#13063; CST, USA) (dilution 1:200) for 2 h. This was followed by incubation with an Alexa Fluor® 488-conjugated secondary antibody (Cat. #4408; CST, USA) (dilution 1:200) for 1 h, and images were taken with a fluorescence microscope (Olympus Corporation).
Immunofluorescence for LC3 and Beclin1 protein. The ligament fibroblasts were seeded on 6-well plates with sterile glass cover slips and transfection of LC3 protein fused with green fluorescent protein plasmid (GFP-LC3) was carried out using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), as previously described . Subsequently, cells were incubated for 72 h and then fixed with 4% paraformaldehyde for 15 mins, followed by incubation with anti-Beclin1 (Cat.#3495; CST, USA) (dilution 1:200) in 5% bovine serum albumin (Cat. # ST023; Beyotime, China) overnight at 4˚C. Subsequently, the ligament fibroblasts were incubated with Alexa Fluor® 594-conjugated secondary antibody (Cat.#8889; CST, USA) (1:100) and then observed by fluorescence microscopy (Olympus Corporation).
Statistical analysis. All data were presented as the mean ± SD. Differences in mRNA expression among the OPLL and non-OPLL groups were analyzed using a t-test, with SPSS 13 (SPSS, Inc.). Multiple comparisons of data among the groups were determined by the one-way ANOVA followed by Dunnett’s test. Pearson correlation coefﬁcients were used to analyze the correlation between parameters of autophagy level and osteogenic makers. P<0.05 was considered to indicate a statistically significant difference.