Animal care and experimental procedures were performed in accordance with the Guidelines for Animal Experimentation of Anhui Medical University and were approved by the Institutional Animal Care and Use Committee (LLSC20190761). Forty 4-month-old male, skeletally mature New Zealand white rabbits were obtained from Anhui Medical University Experimental Animal Center (Hefei, China). The rabbits were randomly divided into the following eight groups: Control, I-4w, SR-1w, SR-2w, SR-4w, rESWT-1w, rESWT-2w, and rESWT-4w, and housed one per cage at 22°C–24°C with a 12-hr light/dark cycle and were allowed access to food and water ad libitum. All rabbits were anesthetized by intravenous administration of 30 mg/kg sodium pentobarbital in an ear vein.
In group Control, the left knee joint was not fixed to serve as a control. In group I-4w, the left knee joint was only fixed for four weeks at full extension using a plaster cast from groin to proximal toes as described previously (Figure 2. A) , and did not undergo recovery. In groups SR-1w, SR-2w, and SR-4w, the left knee joint was fixed for four weeks before the rabbits underwent one, two, and four weeks of self-recovery, respectively, without any special rehabilitation protocol. In groups rESWT-1w, rESWT-2w, and rESWT-4w, the left knee joint was fixed for four weeks before the rabbits underwent one, two, and four weeks of rESWT, respectively. As in our previous study, A radial shock wave device (DolorClast, Switzerland) was used in the experiment, once in a week (Figure 2. C) . Firstly, each rabbit received less than 500 times impulses of shock waves at a small dose of 1.5 bar, 4 Hz to the left quadriceps femoris area, in order to adapt the rabbits to the stimulation. Secondarily, the rabbit received 2000 times impulses at an experimental dose of 2.5 bar, 6 Hz to the same area.
Measurement of myogenic contracture
The rabbits were euthanized with an excess amount of sodium pentobarbital. The left hindlimb was dislocated at the left hip joint and completely removed. As in our previous study, the range of motion (ROM) of the left knee joint before and after myotomy was measured, respectively [16,17]. In accordance with the method described by Trudel and Uhthoff , we evaluated the myogenic contracture caused by the muscular structures. The following formulas were used to calculate the degree of contracture.
Decrement in ROM as a result of total contracture = ROM before myotomy (of the control knee) - ROM before myotomy (of the contracted knee)
Decrement in ROM as a result of arthrogenic contracture = ROM after myotomy (of the control knee) - ROM after myotomy (of the contracted knee)
Decrement in ROM as a result of myogenic contracture = Decrement in ROM as a result of total contracture - Decrement in ROM as a result of arthrogenic contracture
Measurement of muscle fiber cross-sectional area
After measuring ROM, two muscle tissue samples (1 cm × 1 cm × 0.5 cm in size) were cut from the middle of the rectus femoris. One sample was stained with hematoxylin and eosin, while the other was frozen in liquid nitrogen at -80°C until histological analysis.
In the rectus femoris sample stained with hematoxylin and eosin, the cross-sectional area (CSA) of individual myofibers was photographed using a TE2000-U inverted microscope (Nikon Corporation, Tokyo, Japan) and measured using Image J software (National Institutes of Health, USA). Six randomly selected fields of view were analyzed in each group.
Western Blot analysis
The skeletal muscle samples were ground into powder with liquid nitrogen using a grinder and homogenized in RIPA buffer (Beyotime, China) containing protease inhibitors at four degrees Celsius. Homogenates were centrifuged at 12,000 × g for 30 min three times at four degrees Celsius, and the resulting supernatants were collected. The protein concentrations were determined using the bicinchoninic acid method. Protein lysates were separated on a 10% sodium dodecyl sulfate-polyacrylamide electrophoresis gel and transferred on to polyvinylidene fluoride membranes (Millipore, USA). After being blocked with five-percent non-fat dry milk in Tris-buffered saline Tween-20 (TBST) at room temperature for two hours, the membranes were incubated with rabbit anti-phospho-NF-κB p65 antibody (dilution 1:1000; Cell Signaling, USA), rabbit anti-HIF-1α antibody (dilution 1:1000; Cell Signaling, USA) and rabbit anti-MyoD mAb (dilution 1:1000; Cell Signaling, USA) at four degrees Celsius overnight. On the second day, after being washed in TBST solution three times for 10 min per wash, the membranes were incubated with peroxidase-conjugated affinipure goat anti- rabbit IgG-HRP (dilution 1:3000; Cell Signaling, USA) as the secondary antibody for two hours at room temperature. After being washed three times with TBST for 10 min per wash, the membranes were evaluated with the enhanced chemiluminescence system in accordance with the manufacturer's instructions. The band densities were quantified using Image J software (National Institutes of Health, USA). The relative protein levels were calculated by comparison with the amount of GAPDH (dilution 1:1000; Cell Signaling, USA) as a loading control.
All data are expressed as the mean ± standard error of the mean. The assumptions of normality of data and homogeneity of variances between the groups were analyzed by SPSS 21.0 (Chicago, IL, USA). Differences in the myogenic contracture, CSA, average protein levels for HIF-1α and MyoD between group SR and group rESWT at each recovery timepoint were assessed using the Student’s t-test. A significant difference between group SR and group rESWT at the same timepoint was considered to be present when the 95% confidence interval did not overlap with zero. One-way analysis of variance (ANOVA) and the Tukey-Kramer test were performed to examine differences between the timepoints in the myogenic contracture, CSA, average protein levels for phospho-NF-κB p65, HIF-α and MyoD.