Relaxin inhibits patellar tendon healing in rats: a histological and biochemical evaluation

A total of 36 female Sprague-Dawley rats, 6–8 weeks old and weighing from 250 to 300 g, were used in this study. The rats were purchased from the Experimental Animal Center of Soochow University. Also, they were kept in a ventilated environment with a 12 h:12 h light-dark cycle at a constant temperature of 21 °C.

To induce tendon defects, two stacked sharp blades, described in a previous study, was used to remove the center of the patellar tendon (1 mm in width) from the patella distal apex and insert to the tibia tuberosity without causing any damage to the fibrocartilage zone [15] (Fig. 1). The surgery was performed under sterile conditions. After surgery, nylon 5–0 suture were used to close the skin incision. The rats were placed on heating pads in the recovery cages until they recovered from anesthesia. For the first 24 h of recovery, the animals were housed in individual cages and were provided with softened rat chow. The operated rats were divided into two groups: the vehicle group and the relaxin group, with 12 rats in each group. The 12 rats that underwent sham surgery were used as the control group. The vehicle group received subcutaneous injection near the knees of saline (0.2 μg/kg/day) for two and four weeks. The relaxin group also received subcutaneous injection of relaxin (0.2 μg/kg/day) (Protech, Nevada, USA) for two and four weeks. The injection needle was a kind of micro syringe (Gaoge, Shanghai, China). Diameter of needle is 0.5 mm. When we did the injections, two operators performed at the same time, one operator was in fixing the neck and limbs of the rats, and the other operator was responsible for the injection. At the time of injection, the needle tip of the syringe was injecteded with an angle of 30°-50° and avoided damage to surrounding skin and subcutaneous tissue. The dose of relaxin used in this study was based on previous investigations [16]. The rats were sacrificed by intraperitoneal injection an over dose of sodium pentobarbital (150 mg/kg) after 2 and four weeks.

The right patellar tendons of rats, sacrificed 2 and 4 weeks post-operatively, were kept at room temperature in PBS-soaked gauze pads until testing within 1 h after collection. To carry out tensile testing of the tendons, the patella and the proximal tibia were fixed at each clamp (5 mm in between two clamps) of a testing machine (Instron, MA, USA). The samples were tested at a speed of 1 mm/s until failure.

For histological analysis, the left patellar tendon was fixed in 4% paraformaldehyde for one day. After embedding in paraffin, the samples were sectioned longitudinally at a thickness of 6 μm, then deparaffinized in xylene, washed, and hydrated with washing ethanol. The sections were processed for histological examination using hematoxylin-eosin (H&E) and Masson trichrome staining. To carry out the regenerating tissue quality evaluation, we adopted a histological scoring system [12]. The histological analysis was examined independently by two blinded observers. The scores of seven defined tissue regeneration criteria, which were described by Stoll (Additional file 1: Table S1), were then recorded [17]. A score of 12 was given to an intact Achilles tendon without any surgical operation.

The sections were de-waxed, hydrated and inactivated. Then the tissue were blocked with 5% normal goat serum after antigens repairing and incubated at 4 °C overnight using Col I (1:500), Col III (1:500) (Abcam, Shanghai, China), RXFP1 (1:200) (Biorbyt, California, USA), MMP1 (1:500), MMP9 (1:500) (Abcam, Shanghai, China) and MMP13 (1:500) (Biorbyt) monoclonal antibodies. Finally, they were incubated with HRP-conjugated secondary antibody (1:2,000) for 1 h. Sections were colored by incubating with the chromogen 3.3′-diaminobenzidine (DAB) tetrahydrochloride and counterstained with hematoxylin.

There was no loss of specimens during the tests. The results of mechanical testing showed that tissue repair after tendon removal at each postoperative time point in the vehicle and relaxin groups was worse than in the control group (Table 1). At two weeks after surgery, maximum load, stiffness, and Young’s modulus in both the vehicle and relaxin group were lower than in the control group (p < 0.05). This observation suggests that surgical operations change the mechanical properties of tendons. The results demonstrate that the vehicle and relaxin groups have similar mechanical properties during the earlier period of repair. However, the data obtained four weeks postoperatively indicated that the control group exhibited significantly higher maximum load and Young’s modulus (p < 0.05) than the relaxin group. These findings suggest that relaxin disrupts tendon healing, whereas the control group exhibited regeneration. No statistical difference in tendon length and cross-sectional areas between the three groups after two and four weeks was observed.

H&E staining revealed delayed healing in the relaxin group (Fig. 2a). Early collagen formation was seen two and four weeks after wounding in the vehicle and the relaxin groups, respectively. Two weeks after wounding, both vehicle and relaxin groups showed high cell density in the wound areas. However, four weeks after wounding, the vehicle group exhibited a more distinct cell arrangement than the vehicle group. Masson staining was utilized to evaluate collagen maturation levels. At week 2, there were more collagen found in vehicle group than in relaxin group. At week 4, there was further deposition observed. (Fig. 2a). At each time point, intact tendons showed significant differences in histological scores compared with both study groups (p < 0.05). However, the relaxin group presented lower histological scores (p < 0.05) than the vehicle group, indicating that the relaxin group exhibited poor tendon structural recovery (Fig. 2b). These findings indicate that relaxin impairs collagen deposition in the wound areas.

At week 2 and 4 after surgery, immunohistochemistry staining of Col I and Col III were applied in the new tendon tissues. The expressions of Col I and Col III in the neo-tendon tissues were stronger in the vehicle group and more uniform at both weeks 2 and 4. Compared to normal tendons, the expressions of Col I and III were more intense, suggesting that more ECM was deposited into the wound areas during tendon healing (Fig. 2c). The relative expression of Col I and Col III is shown in Fig. 2d, demonstrating a much higher abundance of Col III of vehicle group relative to relaxin group (p < 0.05).

To investigate how relaxin alters collagen deposition in wound areas, we first evaluated the expression of RXFP1, which is the receptor for relaxin. Immunohistochemical staining showed higher RXFP1 expression in the wound areas of the relaxin group compared to the vehicle group (Fig. 3a). An increase in RXFP1 expression in the wound areas was observed in the relaxin group (p < 0.05) (Fig. 3b). These results reveal that relaxin increases RXFP1 expression in injured tendons.

MMPs play an important role in collagen metabolism. We assessed the effects of relaxin on the expression of MMPs and TIMP1. Immunohistochemical staining indicated that MMP9 and MMP13 were highly expressed in the relaxin group (p < 0.05). However, no difference in MMP1 and TIMP1 expression between the two groups was observed. Relaxin treatment effectively increased the expression of MMP9 and MMP13. These findings indicate an alteration in the balance between MMPs and TIMP expression (Fig. 4), thereby resulting in worse healing in the relaxin group compared to the vehicle group.