A total of 30 mature male New Zealand white rabbits (age, 6 months; weight, 1.84±0.89 kg), purchased from the Animal Breeding Center (Zhejiang University, Hanhzhou, China), were randomly divided into three equal groups: Group 1, saline treatment; group 2, chitosan treatment (Sigma-Aldrich, St. Louis, MO, USA); and group 3, chitosan + nicotinamide treatment (nicotinamide is an inhibitor of SIRT1; Sigma-Aldrich). The left hind limbs of the rabbits in each group were designated as the injured tendon, and the right hind limbs were designated as the normal flexor tendons for each group. The rabbits were maintained in individual standard rabbit cages, with access to a standard rabbit diet and water ad libitum. The study was approved by the ethics committee of The First Affiliated Hospital of Medical School of Zhejiang University (Hangzhou, China).

During surgery, the rabbits were administered an intramuscular injection of 2% xylazine (1 mg/kg; Sigma-Aldrich) as premedication, and 10% Ketamin-HCl (60 mg/kg; Sigma-Aldrich) for anesthesia. The thigh of the rabbit was bound using an elastic bandage as a tourniquet, and a 2-cm longitudinal incision was made in the plantar skin between the proximal and distal interphalangeal joints of the toes. The flexor digitorum profundus tendons were transected using a blade, through the incision in the tendon sheath. Finally, the injured tendons were surgically repaired using 6-0 sutures (Hanhzhou AiPu Medical Instrument, Co., Ltd., Hangzhou, China) and the skin covering the lesion was sealed with a simple continuous pattern, using non-absorbable 5-0 silk sutures (Hanhzhou AiPu Medical Instrument, Co., Ltd). A rectangular window was created in the cast at the site of injury (plantar surface of the hind limb) for injection of the reagents. At 6 h post-surgery, 0.2 ml saline, 0.2 ml chitosan, or 0.2 ml chitosan + 5 mM/kg nicotinamide was injected into the hind limb, through the rectangular window, once per day for six weeks.

At 6 weeks post-surgery, the rabbits (n=10/group) were sacrificed via 10 ml air via the ear intravenously and the flexor tendons were harvested and stored at −20°C for subsequent morphological and histological analyses. Gross and histological evaluation of tendon adhesion was initially performed. The standard degree of adhesion was determined, as mentioned previously (12,19). Gross evaluation for adhesion degree was determined as follows: None (no adhesions), filmy (separable from the surrounding tissue), mild (not separable from the surrounding tissue), moderate (35–60% of the injured area) and severe (>60% of the injured area). Evaluation was performed by an investigator in a blinded-manner. Histological evaluation for adhesion formation was determined as follows: None (no adhesions), mild (<33% of the tendon), moderate (33–66% of the tendon surface) and severe (>66% of the tendon surface). The average grade from the grades of 10 slides was calculated. In addition, the flexor tendons were fixed on biomechanical instruments, in order to assess mechanical strain. Pretension was set at 10 N and the maximum tensile breakage of the tendons was recorded at a 20 mm/min drawing speed.

Human tenocytes were derived from five healthy toe tendon explants, which were collected fresh from surgery at the First Affiliated Hospital of Zhenjiang University. In brief, the tendons were stripped aponeurosis under a microscope (DM5500 B; Leica Microsystems, Wetzlar, Germany) and were washed with Hank's solution. The tendon was then subjected to digestion with 0.25% trypsin and 0.1% collagenase for 20 min at of 37°C. The digested tendons were collected, washed with Hank's solution and cut into 1–2 mm³ fragments. The fragments were exposed to 0.25% trypsin and 0.1% collagenase for 1 h at 37°C in order for second digestion. The digested mixture was filtered and centrifuged. The sediment was washed and made into a cell suspension. The cells were counted and inoculated at 5×10⁵/ml. Finally, the cells were cultured in phenol red-free Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% charcoal stripped fetal bovine serum prior to the subsequent experimental protocols The cells were pretreated with chitosan (5, 10 or 50 µg/ml) for 30 min, and were then split into two groups, those that were activated by interleukin (IL)-1β (10 ng/ml), and those that were not. High purity chitosan and IL-Iβ were purchased from Sigma-Aldrich, dissolved in normal saline and added to the culture medium according to the indicated concentrations. All reagents were obtained from Invitrogen Life Technologies, (Carlsbad, CA, USA). All subsequent experiments were conducted 1 h after IL-1β supplementation, unless otherwise stated.

The effects of chitosan and IL-1β on cell viability were detected using an MTT assay (Sigma-Aldrich), which is based on the uptake of 3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyltetrazolium bromide. The tenocytes (2×10³ cells/well) were cultured in 96-well plates 24 h prior to treatment with chitosan. Following treatment with the reagents, the MTT solution was added to each well and incubated for 2 h at 37°C. SDS buffer (10%) supplemented with 0.01 M HCl was then added to the cells and left for 12 h. Subsequently, the absorbance at 570 nm was measured using a spectrophotometer (Perkin Elmer LS-55; PerkinElmer Inc., Waltham, MA, USA). Independent experiments were performed in triplicate.

The rabbit flexor tendons or human tenocytes were lysed using ice-cold lysis buffer containing: 50 mmol/l Tris-HCl (pH 7.4); 1% NP-40; 150 mmol/l NaCl; 1 mmol/l EDTA; 1 mmol/l phenylmethylsulfonyl fluoride; and complete proteinase inhibitor mixture (one tablet/10 ml; Roche Molecular Biochemicals, Indianapolis, IN, USA). A Bicinchoninic Acid Protein Assay kit (Beyotime Institute of Biotechnology, Haimen, China) was used to quantify the protein concentration within the lysate and western blot analysis was subsequently performed. Proteins (40 µg) were separated by 10% SDS-PAGE and transferred to a nitrocellulose membrane (EMD Millipore, Bilerica, MA, USA). The membrane was then probed with mouse anti-goat polyclonal antibody against SIRT1 (cat. no. sc-19857; Santa Cruz Biotechnology, Inc., Dallas, TX, CA), rabbit polyclonal antibody against acetylated p65 (cat. no. 3045s; Cell Signaling Technology, Inc., Beverly, MA, USA), rabbit polyclonal antibody against acetylated p53 (cat. no. 2570s; Cell Signaling Technology, Inc) and rabbit polyclonal antibody against β-actin (cat. no. A2066; Sigma-Aldrich), followed by incubation with anti-goat (cat. no. ab157532) or anti-rabbit (cat. no. ab191866) secondary antibodies (Abcam, Cambridge, MA, USA). All antibodies were used at concentrations and dilutions as recommended by the manufacturer's instructions (dilutions ranged between 1:100 and 1:10,000 for western blot analysis). The blots were visualized by enhanced chemiluminescence (ECL) using ECL reagent (Pierce Biotechnology, Inc., Rockford, IL, USA) and images were captured on X-ray films (ChampGel 6000; Beijing Sage Creation Science, Co., Ltd., Beijing, China).

Detection and quantification of apoptosis was performed using flow cytometry. After treatment with the reagents, the tenocytes were cultured with binding buffer, Annexin V-Enhanced Green Fluorescent Protein and prop-idium iodide (Sigma-Aldrich). The mixture was then incubated at room temperature for 15 min in the dark, followed by flow cytometry (FACScan; BD Biosciences, Franklin Lakes, NJ, USA). The percentage of apoptotic cells was quantified using CellQuest software version 5.1 (BD Biosciences).

Double-stranded oligonucleotides 5′-GAT CCC GTT GGA TGA TAT GAC ACT GTT CAA GAG ACA GTG TCA TAT CAT CCA ACT TTT TTG GAA A-3′ (SIRT1 target sequence underlined) were cloned into the pSuperiorRetroPuro vector (Oligoengine, Seattle, WA, USA). The plasmid was packaged into a retrovirus by transfection of the amphotropic packaging cell line LA (Cell Biolabs, Inc. USA). A virus expressing a scrambled small interfering (si)RNA (5′-GATCCCGCCGTCGTCGATAAGCAATATTTGATATC CGATATTGCTTATCGACGACGGCTTTTTTA-3′; pSuperi-orRetroPuro vector) served as a control. Human tenocytes were infected with the SIRT1 siRNA retrovirus and selected using 0.5 g/ml puromycin (Sigma-Aldrich) for 10 days. Cells were harvested for the subsequent apoptotic assay and western blot analysis.

Statistical analysis was performed using SPSS 16.0 statistical analysis software (SPSS Inc., Chicago, IL, USA). Data was analyzed using one-way analysis of variance. Statistical analyses of cell viability and apoptosis were conducted using Student's t-test. All data were expressed as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

Following surgery, the flexor tendons were harvested from the rabbits, in order to determine the state of adhesion. Based on the gross (10) and histological evaluations of adhesion (Table I), the numbers of rabbits exhibiting severe and moderate levels of adhesion werelowest in the chitosan-treated rabbits (group 2; 0 severe and 2 moderate; n=10), when compared with the levels in the normal saline-treated rabbits (group 1; 6 severe and 3 moderate; n=10), and the effect of chitosan was reversed following nicotinamide treatment (group 3; 5 severe and 3 moderate; n=10). Furthermore, analysis of the maximum force generated by the tendon anastomoses demonstrated that the greatest forces were observed in the chitosan group (Fig. 1). These results indicated that chitosan may possess anti-adhesive bioactivity.

In addition to examination of the physiological adhesion status, the flexor tendons were randomly selected and disrupted using liquid nitrogen, in order to detect the expression levels of specific signaling molecules. As shown in Fig. 2, the expression levels of SIRT1 in the chitosan group (group 2; n=3/group) were elevated, when compared with the control group; chitosan-induced SIRT1 protein expression was observed to be inhibited by nicotinamide. Conversely, the expression levels of the SIRT1-targeted proteins, acetylated p65 and p53, decreased with the upregulation of SIRT1 and increased with the downregulation of SIRT1, which was induced by nicotinamide.

An in vitro experiment was performed in human tenocytes to further ascertain the protective effects of chitosan on tendon repair. Following chitosan pretreatment, the tenocytes were incubated with IL-1β. Fig. 3 demonstrates that chitosan influenced the survival of tenocytes. Treatment with 10 µg/ml or 50 µg/ml chitosan attenuated IL-1β-induced cell proliferation (Fig. 3A) and apoptosis (Fig. 3B) in a dose-dependent manner, compared with the IL-1β group.

Following chitosan pretreatment, the cells were exposed to IL-1β, and the expression levels of various signaling molecules were detected. As shown in Fig. 4, the expression levels of SIRT1 were downregulated and the expression levels of acetylated p65 and p53 were upregulated in the tenocytes stimulated with IL-1β. However, chitosan reversed the changes in protein expression that had been induced by IL-1β. These results demonstrate the fundamental regulatory role of SIRT1 in tendon repair by chitosan.

To determine the role of SIRT1 in the protection of tendons from adhesion, the tenocytes were transfected with small interfering (si)-SIRT1. As shown in Fig. 5A, transfection with si-SIRT1 resulted in a down-regulation of SIRT1 expression, indicating the successful knockdown of SIRT1. As shown in Fig. 4, chitosan reversed the IL-1β-induced downregulation of acetylated p65 and p53. Notably, the action of chitosan on the IL-1β-induced expression of acetylated p65 and p53 was abrogated following SIRT1 knockdown (Fig. 5B). Furthermore, silencing of SIRT1 expression abolished the effects of chitosan on IL-1β-induced cell proliferation (Fig. 5C) and apoptosis (Fig. 5D), when compared with the IL-1β + chitosan negative control group.