Porcine Acellular Nerve-Derived Hydrogel Improves Outcomes of Direct Muscle Neurotization in Rats

Abstract

Background:

The ability to reinnervate a muscle in the absence of a viable nerve stump is a challenging clinical scenario. Direct muscle neurotization (DMN) is an approach to overcome this obstacle; however, success depends on the formation of new muscle endplates, a process, which is often limited due to lack of appropriate axonal pathfinding cues.

Objective:

This study explored the use of a porcine nerve extracellular matrix hydrogel as a neuroinductive interface between nerve and muscle in a rat DMN model. The goal of the study was to establish whether such hydrogel can be used to improve neuromuscular function in this model.

Materials and Methods:

A common peroneal nerve-to-gastrocnemius model of DMN was developed. Animals were survived for 2 or 8 weeks following DMN with or without the addition of the hydrogel at the site of neurotization. Longitudinal postural thrust, terminal electrophysiology, and muscle weight assessments were performed to qualify and quantify neuromuscular function. Histological assessments were made to qualify the host response at the DMN site, and to quantify neuromuscular junctions (NMJs) and muscle fiber diameter.

Results:

The hydrogel-treated group showed a 132% increase in postural thrust at 8 weeks compared with that of the DMN alone group. This was accompanied by an 80% increase in the number of NMJs at 2 weeks, and 26% increase in mean muscle fiber diameter at 8 weeks.

Conclusions:

These results suggest that a nerve-derived hydrogel may improve the neuromuscular outcome following DNM.

Impact statement

The development of a novel peripheral nerve–matrix hydrogel that can promote successful reinnervation of an injured muscle could be of benefit to both the clinical and preclinical research spheres. Clinically, this innovation would allow for a more biologically based therapeutic to focus on restoration of physiologic function. This advancement would create a platform for further research and development into other extracellular-based therapeutics to treat nerve, muscle, and immune-related prognoses.

Introduction

In normal healthy muscle, functional connection between a peripheral nerve and muscle occurs at the neuromuscular junction (NMJ), an interface responsible for control of muscle contraction. Injury to any component of the connection between muscle and nerve leads to loss of function and atrophy of the muscle. Depending on the nature of the injury, standard of care includes multiple approaches to restore neuromuscular connectivity, with the end goal of restoring function and preventing atrophy. In cases of simple nerve transection, primary repair (neurorrhaphy) is achieved by reapproximating and securing the nerve endings. In cases where primary repair is not achievable, nerve grafting through autograft, acellular allograft, or nerve conduit can be used to bridge a nerve gap.

Despite the ability of the peripheral nervous system to regenerate, fewer than 50% of patients experience satisfactory functional recovery associated with traditional methods of nerve repair.^1,2 This is due to slow and difficult growth across sites of injury due to inflammation, scarring, and axonal misdirection.^3,4 Injuries resulting in nerve gaps >1 cm are considered challenging, while gaps larger than 3 cm seldom achieve effective repair using any currently available surgical products and/or techniques.^3,5–9 Preclinical attempts to improve outcomes in nerve regeneration have included combinations of biomaterials, growth factors, and/or cellular constituents.

However, while significant success has been achieved preclinically,^10–12 few of these approaches have translated to the clinic due to limitations, including challenges in engineering complex systems, tailoring of timed release, regulatory concerns, and cost. Furthermore, proximal nerve injuries are often not amenable to these approaches due to muscle atrophy during the long time required for nerve regeneration. In these cases, nerve transfers can be used to shorten the distance between a functional nerve ending and the downstream muscle.

However, each of these approaches rely on the presence of an available segment of nerve connected to the muscle for grafting.^13–16

Successful reanimation of a muscle following significant traumatic neuromuscular tissue loss, such as that resulting from a gunshot or blast injury, poses a substantial clinical problem due to the loss of the distal nerve where it enters the muscle, which is commonly referred to the distal stump or pedicle.^16,17 In these cases, more complex approaches are required to restore movement, such as free-functioning muscle transfers, whereby expendable muscles are transferred to new locations. These procedures are complex, time consuming, pose substantial risks, and limited to large muscles and lack multiple donor options, meaning that it is uncommon to perform more than one free-functioning muscle transfer.^17,18

Another approach to reinnervation is direct muscle neurotization (DMN), which involves implantation of a nerve directly to a targeted tissue or muscle of interest without connection to an existing neuromuscular structure.¹⁹ While a number of studies have reported preclinical and clinical success with direct neurotization procedures, DMN is often viewed as experimental in nature and its therapeutic benefit is not well established. Success of DMN relies on the ability of the implanted nerve to reestablish connections with existing, or support newly formed, motor endplates within the denervated muscle. Much of the challenge in DMN is the ability of the transferred axons to pathfind to prior NMJs—as endoneurial tubes provide guidance to the NMJs within the muscle.²⁰ While reinnervation of existing muscle end plates and the formation of new, innervated muscle endplates has been demonstrated in animal models of DMN, methods that can enhance the recovery of neuromuscular connectivity may have an impact upon the ability to provide functional recovery following significant neuromuscular injuries.^20,21

The present study examined the use of an injectable porcine peripheral nerve extracellular matrix hydrogel (peripheral nerve-derived matrix [PNM]) as a neuroinductive interface between nerve and muscle in a rat DMN model.

A xenogeneic (porcine to rodent) approach was used based upon previous studies reporting the low immunogenicity and cytotoxic effects of porcine tissue-derived matrices as well as relative availability of high-quality tissue, long history of human use, and translatability to the clinic.^10–13,22 The study hypothesis was that PNM would accelerate and improve the formation of neuromuscular connections at the site of neurotization, leading to improved function. This hypothesis was based upon studies demonstrating that PNM promotes recovery in multiple peripheral nerve repair applications, including nerve crush, transection, and gap injury.^23–25 Due to these previous studies, PNM was solely applied to this model to test its impact on the functional outcomes associated with a peripheral nerve-specific matrix. A DMN model of the common peroneal nerve to the gastrocnemius was studied with and without application of PNM at the nerve–muscle interface.

One cohort of animals was evaluated longitudinally for functional recovery over 8 weeks and terminally for electrophysiological response. An additional cohort of animals was evaluated at 2 weeks along with the 8-week cohort for histologic evidence of reinnervation and muscle architecture.

Materials and Methods

Study overview

A xenogeneic peripheral nerve-derived matrix (PNM) was formulated from porcine sciatic nerves, which were frozen, decellularized, digested, lyophilized, and reconstituted at 10 mg/mL, as previously described. A total of 26 Lewis rats were randomized into two experimental groups as follows: (1) DMN (n = 5 at 2 weeks; n = 8 at 8 weeks), and (2) DMN+PNM (n = 5 at 2 weeks; n = 8 at 8 weeks). In the 2-week cohorts, the gastrocnemius with the site of neurotization was harvested at the time of euthanasia for histologic evaluation of the acute response. No electrophysiological or functional testing was performed in the 2-week cohort as no meaningful recovery is expected during this period. In the 8-week cohort, animals underwent weekly postural thrust testing of both the surgically manipulated hindlimb and the contralateral limb. Additionally, electrophysiological assessments were performed at the terminal time point. Tissues were collected after electrophysiology for measurement of muscle weight and histologic assessments of the chronic response.

Animal usage

Fourteen-week-old female Lewis rats were obtained from Charles River Laboratory (Wilmington, MA). Animal studies were performed in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals, the NIH guide for Care and Use of Laboratory Animals, federal and state regulations, and was approved by the University of Pittsburgh Institutional Animal Care and Use Committee (IACUC).

Surgical model

The sciatic nerve was exposed at the point of trifurcation through incision of the skin and biceps femoris using standard surgical technique. The tibial branch of the sciatic nerve was then transected midway between the trifurcation of the sciatic nerve and the distal insertion of the tibial nerve into the gastrocnemius.

A small segment (5 mm) of the tibial nerve was resected and the proximal and distal ends were then ligated using 9-0 nylon suture. The sural nerve was then transected and ligated using a similar approach. The common peroneal nerve was transected immediately proximal to its point of insertion into the tibialis anterior muscle. The distal end of the transected common peroneal nerve was then grafted into a superficial incision (∼2 mm width, ∼1 mm depth) made in the mid-belly of the gastrocnemius muscle using two 10-0 nylon sutures. The incision was made approximately one third the distance from the proximal to distal end of the muscle, approximating the location of the NMJs within the gastrocnemius muscle. The remaining distal stump of the common peroneal nerve was then ligated with a 9-0 nylon suture at the insertion to the tibialis anterior.

Following grafting of the common peroneal nerve, animals in the DMN+PNM group received ∼100 μL of PNM immediately subjacent to the nerve within the muscle incision using a 30G needle. The PNM was allowed to gel in situ for 5 min before removing excess PNM outside the engraftment site using a sterile cotton-tipped applicator and closing the surgical site using standard technique. Animals from the DMN group received the same peroneal nerve engraftment and were subject to a 5-min delay before closing the surgical site for consistency with the PNM group. The animals were survived to 2 or 8 weeks postimplantation. The DMN surgical model, procedure, and PNM deployment are depicted in Figure 1.

FIG. 1.

Rat DMN model. A carcass model (A, B) and intraoperative images (C, D) are shown for clarity. The sciatic nerve and its three branches are exposed (A, C). The tibial (black arrow), common peroneal (white arrow), and sural (blue arrow) branches of the sciatic nerve can be seen along with the proximal portion of the gastrocnemius (green arrow). The tibial and sural nerves are transected and ligated. The common peroneal nerve is then transected and only the distal portion of the nerve is ligated. DMN (red arrow) is then performed (B, D). In animals receiving PNM injection, the material was injected within the site of neurotization of the common peroneal nerve until excess material was observed within the site of implantation. After 5 min to allow gelation, excess PNM was removed from the surgical site. (E) Larger scale view showing DMN of the gastrocnemius following application of PNM (yellow arrows). Scale bar = 5 mm in (C, D), scale bar = 2.5 mm in (E). DMN, direct muscle neurotization; PNM, peripheral nerve-derived matrix. Color images are available online.

Postural thrust

A single operator blinded to treatment group recorded postural thrust measurements weekly, as previously described.²⁶ Animals were allowed to acclimate to the assessment for 10 days before surgery. Assessments were then performed every week until termination. For each measurement, rats were held upright in a towel to restrain both the upper extremities and the contralateral hindlimb. The experimental limb remained extended allowing contact of the distal metatarsus and toes with a digital balance. The animal was lowered until the heel contacted the balance platform. Postural thrust was measured as the peak force applied to resist contact of the heel with the balance while animal was lowered toward the platform. The postural thrust was recorded five times, and the average of the three middle measurements was recorded as the postural thrust value for each animal at each time point.

Electrophysiology

At the 8-week study endpoint, animals were anesthetized, and the main trunk of the sciatic nerve was surgically exposed. The sciatic nerve was then carefully dissected to expose a 1 cm segment proximal to the trifurcation of the nerve. A small, sterile rubber sheet (∼1.5 × 3 cm) was placed underneath the dissected nerve segment to electrically insulate the nerve from its surrounding tissues. A ground electrode was placed in the tail of the animal and a 7 mm subdermal needle was inserted through the skin into the caudal gastrocnemius surface with caudal–cranial direction. A second 7 mm subdermal needle was placed in the lower body of the gastrocnemius muscle through the skin at a distance of 1.5 cm from the proximal dermal needle.

A flexible arm was used to support a double hook nerve stimulator electrode underneath the nerve, keeping the nerve slightly elevated from the underlying rubber sheet ensuring contact with both electrodes. The positive end (distal electrode) of the stimulator was placed 2.5 cm from the proximal subdermal recording electrode. This electrode configuration was used to maximize measurement uniformity among animals.

Electrophysiological assessment was not performed in 2-week animals, as no recovery of nerve function was expected at that time point.

A series of current stimulations were then performed by increasing the amplitude by 0.01 mA per stimulation (5–10 s pause between stimulations) until 0.2 mA was reached. Thereafter, stimulation was increased by 0.1 mA until 0.9 mA was reached. Postprocessing was then performed to calculate the following electrophysiological values for each animal: (1) minimum stimulation current (mA) required to achieve compound motor action potential; (2) minimum stimulation current (mA) required to achieve supramaximal stimulation; and (3) supramaximal compound motor action potential (mV).

Termination, gross morphology, and tissue harvest

At 2 weeks postsurgery or following electrophysiological assessment at 8 weeks, animals were euthanized, and images were taken to assess the gross appearance of the surgical site. The left gastrocnemius was then explanted along with ∼5 mm of peroneal nerve. Muscle weights were recorded on a digital balance.

Tissue preparation, histology, and immunolabeling

Harvested muscle samples were placed onto a cork board and snap frozen in liquid nitrogen using an isopentane bath. The samples were then embedded in optimal cutting temperature medium and frozen as a block for sectioning on a cryostat. Samples were first sectioned in the longitudinal direction capturing 20 μm sections at two levels spaced 100 μm apart. The frozen block with the sample was then divided in the cross-section at the insertion point of the nerve, and at the distal third end of the gastrocnemius. The divided muscle was reoriented within the frozen block, and additional 10 μm sections were cut.

The orientation of the sectioning is shown below in Figure 2. Longitudinal and cross-sections were prepared and stained with hematoxylin and eosin for general histologic assessment. Quantitative analysis of muscle fiber diameter was performed by blinded evaluators using the cross-sectioned samples. Briefly, at each level of the cross-sectioned tissue, a region of interest with preset dimensions covering 30–40% of the total tissue area was randomly chosen. In each area 30–50 muscle fibers were chosen at random, and their diameter measured with ImageJ software.

FIG. 2.

Histological sectioning of the gastrocnemius. (A). Level I longitudinal sectioning from 0 to 120 μm in depth, and Level II longitudinal sectioning from 120 to 240 μm in depth. Cross-sectioning was completed after longitudinal cutting. Representative images of samples following longitudinal (B) and cross-sectioning (C) are shown. Insertion point of the nerve is depicted in the black box. Color images are available online.

The analysis was repeated on two consecutive sets of cross-sections for each sample (six total cross-sections evaluated per animal).

Additional slides were fixed with a 2% paraformaldehyde solution and underwent a series of 1 × phosphate-buffered saline washes before an overnight incubation with antibodies to neurofilament heavy chain (NF200; Ab8135, 1:300; Abcam). On the following day, the slides were washed and then incubated with both a conjugated antibody to acetylcholine receptors (α-Bungarotoxin, Alexa Fluor 488 conjugate dye, 1:250; B13422; Thermo Fisher) and a secondary antibody (donkey anti-rabbit 568, A10042, 1:300; Invitrogen). Each slide was coverslipped with an aqueous mounting media containing 4′,6-diamidino-2-phenylindole to allow visualization under fluorescent microscopy to allow general qualitative assessment of sample morphology and nerve outgrowth. Quantitative assessment of NMJ number was then performed in the longitudinal sections. Briefly, a preset region of interest encompassing the common peroneal insertion site was imaged on a Zeiss AxioObserver 7 inverted fluorescence microscope with Apotome 2.0. Tiled images were acquired and stitched using Zen 3.4 software.

QuPath, an open source bioimage analysis software, was then used to automatically identify α-Bungarotoxin positive NMJ using an optimized object classifier, trained using images from the data set.

Statistical methods

Statistical analysis was performed using Prism 9 (GraphPad, San Diego, CA). Continuous outcome measures (electrophysiological measurements, muscle fiber diameter, and NMJ number) were assessed using a Student's t-test. A mixed effects model with time and treatment was used to assess differences in longitudinal postural thrust data at individual time points in the 8-week cohort. For all tests, significance was determined using a p < 0.05 threshold.

Results

Postural thrust is improved in animals receiving PNM treatment

Significant increases in postural thrust were observed between DMN and DMN+PNM groups starting at 7 weeks postsurgery (Fig. 3A). By the 8-week time point there was a 2.3-fold improvement in postural thrust of the DMN+PNM group as compared with the DMN group (33.04 ± 7.20 g vs. 14.25 ± 8.35 g, p < 0.0001; Fig. 3B). These improvements represent recovery of 15.2% and 31.6% of the postural thrust measured in the contralateral limb at 8 weeks in the DMN and DMN+PNM groups, respectively.

FIG. 3.

Postural thrust assessment. (A). Longitudinal assessment. (B). Differences in postural thrust at the 8-week terminal endpoint. ***p = 0.0001. Data are presented as mean ± SD. SD, standard deviation. Color images are available online.

No statistically significant changes in electrophysiology were observed between groups

No differences in the minimum stimulation current (0.04 ± 0.02 mA vs. 0.04 ± 0.02 mA), the stimulation current at supramaximal Compound Motor Action Potential (0.09 ± 0.06 mA vs. 0.12 ± 0.09 mA), or the magnitude of the supramaximal CMAP (6.58 ± 2.44 mV vs. 7.48 ± 5.95 mV) were observed between the DMN and DMN+PNM groups (Table 1). No functional reinnervation was expected in the 2-week cohort; therefore, electrophysiological analysis was not performed on these animals.

Table 1.

No Statistically Significant Differences Were Observed Between Direct Muscle Neurotization and Direct Muscle Neurotization +Peripheral Nerve-Derived Matrix Groups at 8 Weeks Across Multiple Metrics, Including Minimum Stimulation Current (mA), Current at Supramaximal Compound Motor Action Potential (mA), and Magnitude of Supramaximal Compound Motor Action Potential (mV), All Comparisons Nonsignificant, p > 0.05

	Minimum stimulation current (mA)	Current at supramaximal CMAP (mA)	Magnitude of supramaximal CMAP (mV)
DMN	0.04 ± 0.02	0.09 ± 0.06	6.58 ± 2.44
DMN+PNM	0.04 ± 0.02	0.12 ± 0.09	7.48 ± 5.95

DMN, direct muscle neurotization.

Application of PNM does not disrupt the surgical site

In both groups, the site of DMN was found to be intact with the common peroneal nerve well integrated with the surface of the gastrocnemius. Varying degrees of adipose tissue formation were observed surrounding the sites of nerve ligation in all animals. Representative images of the gross morphologic appearance of the surgical site at 2 and 8 weeks are shown in Figure 4.

FIG. 4.

DMN gross morphologic assessment. Scale bar = 5 mm. Color images are available online.

Histologic assessment and immunolabeling showed larger muscle fiber diameters and earlier NMJ formation in animals receiving PNM treatment

The histologic appearance of the neurotization site in both DMN and DMN+PNM groups was characterized by a Wallerian degeneration response including loss of fascicular structure, fragmentation of myelin, and cellular infiltration within the site of neurotization at 2 weeks. By 8 weeks, the histologic appearance of both the DMN and DMN+PNM groups was characterized by an ongoing cellular response and evidence of sprouting and maturation of axons within the nerve and surrounding muscle (Fig. 5).

FIG. 5.

DMN histologic appearance. Representative hematoxylin and eosin-stained images of the neurotization site in DMN and DMN+PNM groups at 2 and 8 weeks are shown. Scale bar = 500 μm, inset scale bar = 100 μm. Proximal aspect of gastrocnemius at left, distal aspect at right. Color images are available online.

PNM was found to be fully degraded by the 2-week time point, without signs of inflammatory or adverse reactions. Quantitative analysis showed no differences in mean muscle fiber diameter at 2 weeks (43.73 ± 3.63 μm vs. 39.93 ± 12.32 μm), but there was a 1.3-fold increase in the DMN+PNM group at 8 weeks (40.98 ± 2.87 μm vs. 51.59 ± 7.62 μm, p = 0.0028; Fig. 6).

FIG. 6.

Muscle fiber diameter assessment. (A). Representative example of image assessed for muscle fiber diameter in DMN+PNM group at 8-week time point. Black box indicates the region of interest for assessment, red box is area occupied by the representative images in (B, C). Scale bar = 2 mm. (B). Representative images and quantitative assessment of muscle fiber diameters in DMN and DMN+PNM groups at 2 weeks. Scale bars = 50 μm. (C). Representative images and quantitative assessment of muscle fiber diameters in DMN and DMN+PNM groups at 8 weeks. Scale bars = 50 μm. All data are presented as mean ± SD. **p = 0.0028. Color images are available online.

Slides immunolabeled for NF200 and α-bungarotoxin were assessed for general signs of reinnervation, nerve outgrowth, and the presence of NMJs in the site of neurotization. At 2 weeks postimplantation, the appearance of both the DMN and DMN+PNM groups was characterized by the presence of NF200-positive labeling adjacent to the implanted nerve, with limited and diffuse sprouting of axons outside the area of neurotization. This was accompanied by scattered α-Bungarotoxin-positive NMJ, primarily away from the site of neurotization. By 8 weeks postimplantation, the appearance of both the DMN and DMN+PNM groups was characterized by more substantial outgrowth of axonal extensions, spreading into the muscle body from the neurotization site, and an increased number of innervated NMJ as compared with the 2-week time point.

Quantitative assessment of the presence of NMJs showed that there was a statistically significant 1.80-fold increase in the DMN+PNM group as compared with the DMN group (118.2 ± 67.4 vs. 212.8 ± 73.32, for the DMN and DMN+PNM groups, respectively p = 0.047); however, no differences were observed at the 8-week time point (161.5 ± 71.2 vs. 224.8 ± 170.2, for the DMN and DMN+PNM groups, respectively). Representative images of the NMJ observed in longitudinal sections from the DMN and DMN+PNM groups at 2 and 8 weeks are shown in Figure 7.

FIG. 7.

Immunofluorescent assessment of NMJ. (A). Representative example of tiled area assessed for NMJ content in DMN+PNM group at 8-week time point. Inset provides higher magnification view. Scale bar = 1 mm in tiled image, 50 μm in inset. (B). Representative images and quantitative assessment of NMJ in DMN and DMN+PNM groups at 2 weeks. Scale bars = 100 μm. (C). Representative images and quantitative assessment of NMJ in DMN and DMN+PNM groups at 8 weeks. Scale bars = 100 μm. All data are presented as mean ± SD. *p = 0.047. NMJ, neuromuscular junction. Color images are available online.

Discussion

DMN has been described in the literature in multiple preclinical models, some individual cases, and a few case series^16,17; however, it is considered an inefficient method for re-establishing neuromuscular connectivity when compared with other approaches.²⁰ This is in part due to the absence of guiding endoneurial structures, which are present in the more traditional nerve-to-nerve reconstruction techniques.²¹ Thus, DMN is often viewed as a salvage technique, when no other options for reanimation exist.¹⁷ Nerve functional recovery ultimately depends on the efficient reinnervation of the muscle fibers by the axons of the transferred nerve.

The ability of the transferred nerve to reinnervate the neurotized muscle fibers depends upon the condition of existing NMJ, the formation of new NMJ, and the efficiency upon which extending axons establish functional connections with muscle fibers.¹⁷ The rate of reinnervation by a single axon of multiple adjacent muscle fibers previously innervated by multiple distinct axons (“adoption”) is another factor that affects the degree of axon-to-muscle fiber connectivity.¹⁹ More efficient DMN approaches, able to improve upon these factors, could create promising new opportunities for reinnervation.

Factors affecting the success of DMN have been identified and include the extent of tissue damage, time of denervation, surgical technique, nerve used as a donor, and age of the patient, among others.^{16,17,21–25,27,28} A number of preclinical studies have demonstrated the ability of nerve regeneration techniques to enhance outcomes associated with DMN. Delivery of neurotrophic factors (fibroblast growth factor-2 and nerve growth factor) within the site of reinnervation in rats resulted in increased muscle mass and higher numbers of motor endplates compared with neurotization alone.^29,30 Transplantation of Schwann cells or bone marrow mesenchymal stem cells in rats has also resulted in improvements in nerve functional outcomes.^31,32 However, the mechanisms underlying these improvements in preclinical models have not been thoroughly investigated, and further work is necessary to determine their suitability and potential in the clinic.

The goal of the present study was to investigate the use of PNM as a neuroinductive interface to establish its potential to improve functional and histologic outcomes of DMN. The results of the study suggest that application of PNM following DMN may lead to early increases in the number of NMJ (1.8-fold at 2 weeks), increased muscle fiber diameter (1.3-fold at 8 weeks), and improved motor functional recovery (2.3-fold by 8 weeks) compared with controls. It is unclear whether the presence of NMJ was due to better maintenance of existing NMJ or newly formed NMJ following DMN. The preliminary results from this study suggest that a more mechanistic approach into the mechanism surrounding the NMJ formation following DMN is a necessary piece of understanding the way in which PNM is impacting the injury site. If the NMJ is found to be maintained from its preinjury state, this may suggest that the PNM plays a role in preventing muscle atrophy, which can translate into other therapeutic uses for the PNM. A recent study showed that injecting decellularized ECM into muscle following an injury prevented muscle atrophy after nerve injury.²⁷

To address this further, future studies have been proposed to use more terminal time points to look at the temporal denervation and muscle atrophy stages following DMN. Additional in vitro work has been planned to further characterize this phenomenon.

The exact mechanisms for the improvements observed in the present study are unclear. However, PNM has previously been shown to improve outcomes in nerve regeneration and functional connectivity in rodent models of nerve crush, transection, and gap injury.^33–37 These studies also showed that the placement of PNM at the site of nerve injury is associated with enhanced Schwann cell recruitment, increased expression of proangiogenic factors, and transition of macrophages from an M1-proinflammatory to an M2-prohealing phenotype.^35,37 While these factors were not investigated in the present study, they are known to have a positive impact upon nerve recovery following injury and may similarly have the potential to enhance outcomes following DMN.

Additionally, improvements in both histologic and functional outcomes were observed in the absence of improvements in electrophysiological outcomes. Previous studies have shown similar functional improvements lacking in the same CMAP improvement as seen in this study.³⁸ Future studies have begun to include a larger sample size to decrease the sensitivity of this exploratory study. Several factors, including differences in muscle fiber type grouping, collateral sprouting, and synkinetic reinnervation may explain the observed outcomes and should be investigated in follow-up studies to corroborate these findings. Additionally, no metrics of nociception were included in the present study. Differences in sensation and pain in the affected hind limb may have impacted postural thrust and toe spread. Further studies are needed to elucidate the interactions between nociception and improved function in this model. Future studies to include a repositioning of the tibial nerve in the DMN will also be considered, along with inclusion of measuring the tetanic force produced by the gastrocnemius muscle following DMN.

Conclusion

This study suggests that the use of PNM can promote neuromuscular connectivity and improve functional outcomes in applications in which DMN is a clinically viable option.

Authors' Contributions

M.N.B. and B.N.B. designed and performed the experiments, performed the analysis, and wrote the article. L.S. and M.A.M. helped design the experiments and assisted with drafting of the article. C.D.S. assisted with the experiments and animal husbandry. M.K., A.L.N., and C.T.F. designed and performed the histologic analyses and assisted with their associated analyses. M.K. also assisted with the article drafting. All participated in the review and approval of the final article.

Disclaimer

The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the Department of the Army, Department of Defense, or the U.S. Government.

Disclosure Statement

L.S. is an employee and owns shares in Renerva, LLC (Renerva). M.A.M. has received payments from and has an equity stake in Renerva. B.N.B. has received payments from and owns shares in Renerva.

Funding Information

This work was supported by the U.S. Army Medical Research and Development Command, Fort Detrick, Maryland, under Task Order W81XWH-18-9-0009 through Medical Technology Enterprise Consortium, Project Number MTEC-18-05-PeripheralNerve-0025.