Effects of 4-aminopyridine on combined nerve and muscle injury and bone loss

Abstract

Purpose:

Musculoskeletal injuries are common and peripheral nerve injury (PNI) causes significant muscle and bone loss within weeks. 4-aminopyridine (4-AP) improves functional recovery and muscle atrophy after PNI. However, it is unknown whether 4-AP has any effect in isolated traumatic muscle injury and PNI-induced bone loss.

Methods:

Standardized crush injury was performed on sciatic nerve and muscles in mice, and animals were assigned to daily normal saline or 4-AP treatment for 21 days. Post-injury motor and sensory functional recovery was assessed, injured muscles were processed for histomorphometry, and the tibial bone was scanned for bone density.

Results:

4-AP significantly accelerated post-injury motor and sensory function recovery, improved muscle histomorphometry, increased muscle satellite cell numbers, and shifted muscle fiber types after combined nerve and muscle injury. Importantly, 4-AP treatment significantly reduced PNI-induced bone loss. In contrast, in isolated muscle injury, 4-AP had no effect on functional recovery and bone density but it improved muscle-specific histomorphometry to a limited extent.

Conclusions:

These findings demonstrate the potential beneficial effects of 4-AP on the recovery of muscle morphology and bone density after combined muscle and nerve injury.

Clinical Relevance:

Nerve injuries frequently involve muscle and results in rapid muscle and bone atrophy. In this scenario, 4-AP, in addition to accelerating nerve functional recovery, may work as an adjunctive agent to improve the recovery of injured muscle and attenuate PNI-induced bone loss.

Introduction

Peripheral nerve injury (PNI) occurs in ~3% of all trauma patients and is frequently occurs concurrently with muscle and bone injury along with soft tissue contusion, crush, and stretching.¹ Muscle injuries account for 10-55% of all injuries.² Hamstring and quadriceps muscle injuries are most common in sports medicine.^3,4 Injury to a muscle may either be isolated or partially dependent on innervation by a damaged nerve. Since decreased activity level, prolonged rehabilitation time, and fear of reinjury are important sources of morbidity associated with musculoskeletal injury,^5,6 we sought to modify a PNI model to include a more realistic direct traumatic injury to muscle in the vicinity of the nerve injury, as is seen typically in nerve injury patients.^1,7 In choosing the PNI model to modify, we selected the standard sciatic nerve crush injury model that we and others commonly use,^8-13 and added a direct local muscle injury. To injure both the hamstring (innervated by branches of the sciatic nerve)¹⁴ and quadriceps (innervated by the femoral nerve)⁴ muscles, we employed a modified PNI model with some features of real-world traumatic limb injuries, where muscle is rarely wholly spared in the setting of PNI. We could find no examples of a carefully performed mixed injury model like this in the literature.

Muscle and bone are complementary and essential for musculoskeletal function.¹⁵ It is well documented that a mechanical and biochemical coupling exists between muscle and bone that is essential in their anabolism, catabolism, and function.^16-18 Nerve fibers innervate periosteum, bone marrow and mineralized bone, and recent studies demonstrate a significant role of the nervous system in bone structure, function, mineralization, and remodeling.^19-21 Intact neuromuscular signaling is important for bone homeostasis and PNI results in substantial bone loss with increased fracture risk.^22-24 Although the extent and severity of both nerve and muscle injury directly affect bone recovery and volume, the integrity of surrounding soft tissues plays a vital role in bone regeneration and remodeling.^17,25 Despite increased understanding of PNI-induced pathophysiology,^26-29 no single therapy significantly limits the pathophysiological changes in nerve, muscle and bone after traumatic injury.

4-aminopyridine (4-AP) is a potent potassium channel blocker and a Food and Drug Administration (FDA)-approved drug for neurodegenerative disorders such as multiple sclerosis, where it restores impaired nerve impulse propagation and improves walking performance.^30,31 Several studies have shown a potential repurposing effect of 4-AP on myelination, nerve conduction, neurogenic muscle atrophy, muscle quality and functional recovery after PNI in rodents.^8-13 However, it is unknown whether the beneficial effect of 4-AP is present in the presence of a concomitant traumatic muscle injury or muscle disuse-induced bone loss, especially in the setting of combined nerve and muscle crush injury. Therefore, the goal of this pilot study was to establish a novel combined nerve-muscle injury model and evaluate the potential repurposing effect of 4-AP on traumatic muscle recovery and PNI-induced bone loss.

Materials and Methods

The experimental design and animal protocols were approved by the institutional animal care and use committee (IACUC) at The Pennsylvania State University College of Medicine. Ten-week-old male C57BL/6J mice (Jackson Laboratories, Bar Harbor, Maine) weighing 25 ± 3g were used in this study. The experimental animals were handled according to the guidelines of IACUC and the National Institutes of Health (NIH) for the care and use of laboratory animals.

Mouse model of combined sciatic nerve and muscle crush injury

Sciatic nerve crush injury was performed as previously described.^32,33 Briefly, under anesthesia and aseptic conditions, the right sciatic nerve was explored by splitting the biceps femoris (BF) and quadriceps femoris (QF) muscles along the length of the femur (Fig. 1A-B). In combined nerve and muscle injury, a 30-second crush injury to the sciatic nerve was performed approximately 5 mm proximal to its trifurcation using a calibrated forceps (3.3 mm tip width; Miltex Premium, BL-1901, Miltex Instruments, Switzerland) coupled with a jig to ensure an injury pressure of ~4.4MPa (Fig. 1C). The BF and QF muscles were then crushed individually for 60 seconds at the level of mid-femur using the same forceps but with a different jig producing an injury pressure of ~10.8MPa (Fig. 1D). This muscle crush injury pressure was selected based on our preliminary pressure-injury screening tests where we found that a low pressure (~2.2MPa) for 60 seconds with forceps caused a subtle injury, whereas a high pressure (~20.3MPa) with a needle driver caused a near complete laceration in 60 seconds (Supplemental Fig. S1). In isolated muscle injury alone, the BF and QF muscles were crushed individually for 60 seconds as described above. After injury, muscles were apposed together covering the sciatic nerve, the skin was closed with surgical staples. Slow release buprenorphine (0.05 mg/kg) was given subcutaneously as an analgesic post-operatively.

FIGURE 1:

Dissection of the lateral hindlimb showing representative gross images of BF and QF muscles and sciatic nerve before and after injury. A, Uninjured QF and BF muscles and B, uninjured sciatic nerve. C, Damage to the sciatic nerve after 30 seconds of 4.4MPa crush injury. D, Damage to the QF and BF muscles after 60 seconds of 10.8MPa crush injury. From Fig. 1D it is clearly visible that forceps pressure-induced a strong muscle contusion producing a mixed Grade I and Grade II type of muscle injuries with bruises and partial tearing at the injury site. The arrows indicate the crushed injury sites at sciatic nerve and muscles.

The experimental animals (5-6 animals/group) were randomized immediately after surgery to receive intraperitoneal normal saline (0.1 ml/mouse) or 4-AP (40μg/20g mouse) treatment daily for 21 days. This body mass-adjusted dose of 4-AP (2 mg/kg) is equivalent to the lowest available starting dose in humans (10 mg daily).

Post-injury functional recovery was assessed by walking track analysis (WTA), sensory nerve test (Von Frey test, VFT), and grip strength test, and analyzed against baseline (before surgery, day 0) values. On post-injury day 22, muscles (BF, QF) and tibial bone were processed for histomorphometric analysis and imaging, respectively.

Sciatic functional index (SFI)

To evaluate in vivo global motor function recovery, the sciatic function index was determined by WTA before crush injury (day 0) and at post-injury day 1, 3, 7, 14, and 21, as previously described.^8,33

Hindlimb grip strength test

Grip strength is a reliable predictor of muscle strength.³⁴ The grip strength meter (BIO-GS3; Bioseb-In Vivo Research Instruments) was used to measure hindlimb grip strength as described previously.³³ Briefly, the mice were allowed to grasp the angled grid and then gently pulled along the length of the sensor grid until the grip was released, and the value in the "g" force was recorded.

Von Frey test (VFT)

For sensation, VFT was performed as previously described using the Von Frey filament unit (NC12775-08, Touch Test Sensory Evaluators).³³ Briefly, the filament pressure (1g force) was applied to the mid-plantar surface of the hind limb via the mesh floor, until the filament bends (maximum 3s) or the animal withdraws their paw. The withdrawal reflex of the hind limb was recorded and each animal tested three times to calculate the percent response of sensory nerve. Of note, in a pilot study, we checked sensory response to different filament forces (0.25, 0.75, 1, 2 g) following sciatic nerve crush injury (vs. control) and found that only 1 g force was able to elicit a reproducible response.

Muscle processing for hematoxylin and eosin (H&E) staining and immunofluorescence (IF) analysis

Muscle processing, cryosectioning, staining and analysis was performed as previously described.⁹ Briefly, on post-injury day 22, BF and QF muscles at the injury site were harvested from the injured and uninjured contralateral hindlimbs, embedded with optimal cutting temperature medium, and cross-sectioned at 7μm for H&E staining and IF analysis.

Cross sectional area (CSA) of H&E-stained muscle fibers was quantified using Image J software. Three random microscopic fields were chosen from each muscle and 3 animals were analyzed per group.

Immunofluorescence analysis of BF and QF muscle cryosections for paired box transcription factor 7 (Pax7)-expressing resident stem cells [satellite cells (SCs)], cell proliferation marker Ki67, and muscle fiber types (Type I myosin heavy chain positive) was performed, as described previously with slight modification.^9,35 Primary antibody staining was performed with anti-laminin (1:100, #L9393, Sigma), monoclonal anti-myosin (skeletal, slow) (1:500, #M8421, Sigma), anti-Pax7 (1:100, #ARP32742_P050, Aviva Systems Biology), and anti-Ki67 (1:400, #9129S, Cell Signaling Technology) followed by appropriate secondary antibody incubation [Alexa Fluor 488 (1:500, #A11008, Invitrogen) or Alexa Fluor 594 (1:500, #A11032, Invitrogen)]. Staining without primary antibodies was used as a control for non-specific fluorescence. Nuclei were counter-stained using ProLong Gold anti-fade reagent with DAPI (#P36935, Invitrogen) and sections were examined under a fluorescent microscope (ZEISS Apotome 2). Stained muscle tissue images were analyzed for Pax7⁺ and Ki67⁺ cells, and muscle fiber types.

Dual energy X-ray absorptiometry (DXA) analysis

Bone mineral density (BMD, mg/cm²) and bone mineral content (BMC, mg) were measured ex vivo using the UltraFocus DXA system (Faxitron Bioptics). Briefly, the injured and uninjured (contralateral) hindlimbs of mice were harvested on post-injury day 22 and were preserved in 10% Neutral buffered formalin (NBF, #SF100-20) until DXA scanning. The instrument was calibrated using a phantom with known BMD. The scan included both right and left hind limbs, however, the region of interest was limited to the entire tibia and measured in triplicate.

Statistical analyses

The sample size required to reach the 5% significance level was calculated through computer simulation and preliminary experiments. The power analysis showed that at least 5-7 mice are required for nerve crush injury and fewer animals (n=3-4) are needed for muscle histomorphometry and immunohistochemistry.^8,11,12 All values are presented as mean ± SEM. The data were analyzed using one-way or two-way analysis of variance (ANOVA) with either Bonferroni or Tukey’s correction. Probability (P) values of < 0.05 were considered statistically significant.

Results

4-AP treatment accelerates functional recovery after combined nerve and muscle injury

Fig. 2A shows that 4-AP treatment modestly accelerated post-injury SFI as compared with the saline group and it was significant at post-injury day 7 (*P<0.05, n = 6/group). SFI returned to normal level at post-injury day 21 in 4-AP group compared with low SFI in the saline group. 4-AP treatment also significantly improved hindlimb muscle grip strength (Fig. 2B) at post-injury day 7 (*P<0.05, n = 6/group). Similar to motor functional recovery, 4-AP treatment significantly improved post-injury sensory functional recovery (Fig. 2C) compared with the saline group (*P<0.05, n = 6/group).

FIGURE 2:

Effect of daily systemic 4-AP treatment on post-injury motor and sensory function recovery after combined nerve and muscle crush injury. A, Systemic 4-AP treatment starting from the surgery day significantly enhanced SFI and hind limb grip strength (g) on day 7 (A, B), and cutaneous sensation (VFT) of the nerve on day 7, 14, and 21 (C). Data are expressed as the mean ± SEM; *P< 0.05 vs. saline group; n = 6/group.

Effect of 4-AP treatment on muscle histomorphometry, cellular markers of muscle regeneration, and muscle fiber types after combined nerve and muscle injury

Representative uninjured and injured muscle images of BF (Fig. 3A) and QF (Fig. 3C) muscles in both saline and 4-AP groups show distinct differences in muscle architecture as described in the figure legends. Quantitative analysis (Figs. 3B and 3D) revealed significantly reduced CSA in the injured limbs (*P<0.05 vs. uninjured, n = 3/group); however, 4-AP significantly improved CSA in the injured muscles compared to the injured-saline group ^$P<0.05).

FIGURE 3:

Effect of daily 4-AP treatment on skeletal muscle morphology after 21 days of combined nerve and muscle crush injury. (A, C) Representative images of BF and QF cross sections stained with H&E. Uninjured BF (Figure 3A) and QF (Figure 3C) muscles in both saline and 4-AP groups show normal muscle architecture with a tightly packed homogeneous polygonal shaped muscle fiber distribution, peripherally placed nuclei, minimal intramyofiber spacing, and little cellular infiltration. Compared to uninjured muscle, cross sections of injured muscles in both groups show muscle fibers with increased intermyofiber spacing and cellular infiltration. Each representative image is drawn from 9 images from 3 different mice. Scale bar, 50μm; magnification, 20×. Quantitative bar graphs for CSA (μm²) from BF (B) and QF (D) muscles. Data are expressed as mean ± SEM; *P<0.05 vs. uninjured group; ^#P<0.05 vs. uninjured saline group; ^$P<0.05 vs. injured saline group; n = 3/group.

Fig. 4 shows representative images (Figs. 4A, 4C) and quantitative analysis (Figs. 4B, 4D) of Pax7⁺ and Ki67⁺ cells in the muscles. There was no significant difference in the number of Pax7⁺ and Ki67⁺ cells between uninjured saline and injured saline groups. 4-AP significantly increased the number of Pax7⁺ and Ki67⁺ cells in the injured BF muscles compared with the uninjured 4-AP (*P<0.05) and injured saline (^$P<0.05) group. In QF muscles, 4-AP treatment was able to reverse the decreased number of Pax7⁺ and Ki67⁺ cells in the injured muscles towards the normal level.

FIGURE 4:

Effect of 4-AP on muscle proliferating cells 21 days after combined nerve and muscle crush injury. Representative Pax7⁺ (red), Ki67⁺ (green), DAPI (blue) and merged immunofluorescence images in the injured BF (A) and QF (C) cross sections from both saline and 4-AP groups at post-injury day 22. Each representative image is drawn from 18 images from 3 different mice. Scale bar, 20μm; magnification, 40×. Quantitative bar graphs for Ki67⁺ and Pax7⁺ cells from BF (B) and QF (D) muscle sections. Data are expressed as mean ± SEM and presented as % of total cells; *P< 0.05 vs. uninjured 4-AP group; ^$P<0.05 vs. injured-saline group; n = 3/group.

Effect of 4-AP on BMD and BMC after combined nerve and muscle injury

Fig. 5A shows DXA scanned images of bones from both hindlimbs. Qualitative analysis of DXA images revealed a significant loss of BMD (Fig. 5B; *P<0.05) and BMC (Fig. 5C; *P<0.05, n = 6/group) in nerve injury-associated tibial bone (injured limb) compared with the contralateral tibial bone. Further analysis revealed that 4-AP significantly improved both BMD (^$<0.05) and BMC (^$<0.05) in the tibial bone of injured limb compared with injured-saline group.

FIGURE 5:

Effect of 4-AP on bone composition after 21 days of combined nerve and muscle crush injury. (A) Representative ex vivo images of mouse hind limb bones acquired from the uninjured and injured limbs of saline and 4-AP groups at post-injury day 22. Quantitative bar graphs for tibial BMD (B) and BMC (C). Data are expressed as mean ± SEM; *P< 0.05 vs. uninjured group; ^$P<0.05 vs. injured-saline group; n = 6/group.

Effect of 4-AP treatment on functional recovery, muscle histomorphometry, and bone density after isolated muscle injury alone

Supplemental Fig. S2 shows that SFI, grip strength and von Frey testing remained unchanged throughout the post-injury period as compared with baseline (day 0) and 4-AP treatment had no effect on any of these parameters. Qualitative analysis of DXA images in Supplemental Fig. S3 revealed non-significant borderline changes in tibial BMD and BMC, without any effect of 4-AP treatment. For the muscle, we looked for the effect of 4-AP treatment on BF and QF muscles CSA. Supplemental Fig. S4 shows that 4-AP treatment had different effects on injured muscle’s CSA where it significantly improved BF CSA (P<0.05, n = 3/group) without any effect on QF CSA.

Discussion

Our findings in this study provide insights into the effects of 4-AP in conditions where both muscle and nerve are traumatically injured. We demonstrate that 4-AP treatment can overcome the pathophysiological changes after combined nerve and muscle crush injuries, and is associated with augmented motor-sensory functional recovery, improved muscle fiber morphology, increased numbers of muscle regenerating stem cells, and it attenuated bone loss. Importantly, the functional recovery is consistent with our earlier studies with 4-AP on nerve crush injury alone^8-12 and demonstrates that 4-AP can also effectively promote functional recovery after combined nerve and muscle crush injury in a murine model. Of note, 4-AP treatment in isolated muscle injury caused a muscle-specific limited improvement of CSA without any effect on nerve function and bone density.

Functional recovery after PNI is poor,^36-38 and combined nerve and skeletal muscle injuries are major cause of morbidity with profound long-term functional impairment and permanent disability.^39,40 Several studies demonstrated the potential beneficial effects of novel therapeutic agents in diverse animal models of PNI,^27,29,41 but there is only limited data on combined nerve-muscle injury model.⁴² We chose BF muscle because of its proximity to and dependence on the sciatic nerve,¹⁴ and QF muscle because of its proximity to the same nerve and separate innervation.⁴ There is no study to date on concurrent sciatic nerve and BF and QF muscles injury. We aimed to explore a novel animal model of combined nerve-muscle injury to mimic clinically-relevant PNI conditions where both nerve and adjacent muscles are involved, and we were able to use the model (Fig. 1 and Supplemental Fig. S1) to show a beneficial effect of 4-AP on post-injury neuromuscular function (Fig. 2) and muscle morphologic recovery (Fig. 3) in comparison with isolated muscle injury (Supplemental Figs. S2 and S4).

Denervation causes rapid muscle atrophy within weeks,^36-38 but skeletal muscle has a remarkable regenerative capacity because of its myogenic resident satellite (stem) cells.^43,44 Importantly, muscle repair capacity is closely associated with satellite cell number because these cells can proliferate and form new tissue.⁴⁵ Neuromuscular electrical stimulation is clinically used to restore skeletal muscle mass and function, increase range of motion, and prevent muscle atrophy.⁴⁶ We have recently shown an entirely new approach to cellular excitation by pharmacological modulation of membrane K⁺ channels using 4-AP to improve muscle atrophy and muscle contraction with increased expression of Pax7⁺ satellite cells and increased muscle fiber diameter after isolated sciatic nerve crush injury in mice.⁹ In this study, even after exaggerated insult to the muscle with combined nerve-muscle injury, we observed that 4-AP treatment significantly improved muscle fiber CSA in the injured BF and QF muscles and it was associated with increased number of Pax7⁺ satellite cells and proliferating cells in BF muscles. Interestingly, in isolated muscle injury, 4-AP treatment improved CSA in injured BF muscle without any effect on injured QF muscle. Taken together these findings demonstrate a muscle protective and regenerative property of 4-AP after traumatic injury, perhaps because of its neuromuscular stimulatory properties, activation of resident satellite cells, and muscle-specific direct effects.

Muscle fiber-type composition is an important indicator of global muscle health.⁴⁷ Mammalian skeletal muscle is broadly characterized into two distinct categories of muscle fibers: Type I (slow twitch, oxidative, resistant to fatigue) and type II (fast twitch, glycolytic, fatigable).⁴⁸ To determine the effect of 4-AP treatment on muscle fiber types, we performed immunohistochemical analysis on muscle cryosections. Our findings show that most of the muscle fibers (>97%) in the saline uninjured BF and QF muscles were type II fibers, while type I fibers accounted for a minor fraction (<3%). While combined nerve-muscle injury did not cause a significant shift in muscle fibers types in BF or QF muscles in the saline group, 4-AP treatment itself induced a significant shift of BF muscle fiber populations toward type I fibers (fast-to-slow transitions) in both uninjured and injured limbs compared with the saline group. While the mechanism of this muscle fiber changes is unknown, it has been reported that low-frequency electrical stimulation causes slow-to-fast fiber shifts⁴⁹ and increased neuromuscular activity and mechanical loading induces fast-to-slow fiber transitions.⁵⁰ As discussed above, it is possible that 4-AP-induced pharmacological stimulation of neuromuscular electrical properties might be involved in muscle fiber type shifts.

Nerve and muscle injuries effectively downregulate metabolic and osteogenic functions of bone, which eventually results in bone volume loss.^19-21 Sciatic nerve crush injury, chronic constriction injury or partial sciatic nerve ligation has been reported to cause significant early bone loss in rodents.^22-24 In this context, we explored the potential beneficial effect of 4-AP on tibial bone loss and found that combined nerve-muscle injury-induced trabecular bone loss was significantly attenuated by 4-AP treatment. While the increase in bone density is multifactorial and improved mechanical loading is considered a potent stimulator of osteogenesis, biochemical coupling between muscle and bone is also reported play an important role in bone metabolism.^16,17 In addition to the secondary effects of improved mobility and strength of the injured limb on bone health, neuromuscular electrical signaling might be involved in trabecular bone homeostasis in our study, as reported by others.^22-24 This novel finding of the effects of 4-AP on bone loss prevention beyond nerve and muscle injury recovery opens a new window for future research in bone trauma or disorders.

Our pilot study has some limitations. First, we did not directly test strength in BF or QF muscles for the fear of variably exacerbating the muscle injury. Second, we did not investigate the cellular, molecular and biochemical mechanism(s) involved in 4-AP treatment-induced improvements in muscle morphology and bone density. Finally, we studied the most common nerve crush injury and not the permanent denervation model or bone fracture model.

In conclusion, these findings demonstrate a potential therapeutic effect of 4-AP on the recovery of traumatic muscle injury and bone loss after combined muscle and nerve injury. Future studies in nerve transection, isografting, and bone injury model with advanced imaging and bio-mechanical properties should determine important mechanistic insights in 4-AP-induced neuro-musculoskeletal protection.