Comparing electrical stimulation and tacrolimus (FK506) to enhance treating nerve injuries

Abstract

Introduction

Neuroenhancing therapies are desired because repair of nerve injuries can fail to achieve recovery. We compared two neuroenhancing therapies, electrical stimulation (ES) and systemic tacrolimus (FK506), for their capabilities to enhance regeneration in the context of a rat model.

Methods

Rats were randomized to four groups: ES 0.5 mA, ES 2.0 mA, FK506, and repair alone. All groups underwent tibial nerve transection and repair, and outcomes were assessed by using twice per week walking track analysis, cold allodynia response, relative muscle mass, and nerve histology.

Results

Electrical stimulation and FK506 groups demonstrated improved functional recovery and myelinated axon counts distal to the repair compared with repair alone. Electrical stimulation provided improvements in nerve regeneration that were not different from optimized FK506 systemic administration.

Discussion

Providing ES after nerve repair improved regeneration and recovery in rats, with minimal differences in therapeutic efficacy to FK506, further demonstrating its clinical potential to improve management of nerve injuries.

1 |. INTRODUCTION

Peripheral nerve injuries occur in an estimated 2% to 3% of over 3 million upper extremity traumas in the United States annually.^1–3 An additional 11% to 17% of nerve injuries are iatrogenic, and more than 50 000 nerve repair surgeries are performed in the United States each year.^4–6 These patients with nerve injuries often fail to gain meaningful functional recovery even after undergoing surgery. These suboptimal outcomes may be attributed to factors intrinsic to nerve regeneration. Specifically, the slow growth of regenerating axons, potential long distances for axon growth, axonal misdirection from their appropriate targets, and a general decline in the capacity to promote axon growth over time are all evidence-supported theories that effect recovery and are thought to play pivotal roles in clinical outcomes.^7–12 Thus, therapies that can complement nerve repair to improve regeneration and recovery are greatly needed.

Systemic tacrolimus (FK506) administration is a well-known, neuroenhancing therapy that can accelerate axon growth and improve functional recovery.^13–20 However, the drug is a potent immunosuppressant used to prevent the rejection of transplanted tissue. Its systemic administration has been associated with a host of undesirable side effects. Although systemic FK506 administration has been used with positive effect in patients with peripheral nerve injuries reconstructed with transplanted nerve allografts,²¹ the detrimental side effects profile of FK506 has discouraged its general use in patients with repaired nerve injuries.

Electrical stimulation (ES) has emerged as an alternative neuroenhancing therapy to treat a variety of nerve injuries. Previous studies have shown in vivo and through in vitro cell culture models that brief ES accelerates and increases axonal outgrowth^22–26 and preferential reinnervation of motor and sensory pathways.^{22–24,27–29} Indeed, in nerve injury with repair animal models, ES results in a greater number of neurons regenerating their axons across the repair site in less time.^28,29 This improved regenerative axonal response leads to earlier motor and sensory recovery.^30–35 Furthermore, animal studies have demonstrated that ES has the capability to enhance regeneration even when nerve repair is delayed after initial injury,³⁶ improving clinical utility.

However, despite its potential, there are few studies that have compared its neuroenhancement to other therapies, such as FK506, for context. In this study, we evaluated the similarities and differences between two neuroenhancing therapies, ES and FK506. We hypothesized that supplementing nerve repair with either ES or FK506 would improve regeneration and recovery compared with repair alone in a rat model of tibial nerve transection and repair.

2 |. MATERIALS AND METHODS

2.1 |. Animal care

All animal procedures were approved by the Washington University in St Louis Institutional Animal Care and Use Committee and performed in accordance with the guidelines of the National Society for Medical Research and the National Institutes of Health. Adult male Lewis rats weighing between 220 and 250 g were used (Charles River Laboratories, Wilmington, Massachusetts). All rats were housed in a central animal care facility and fed rat chow (No. 5001; Purina Mills, St Louis, Missouri) and water ad libitum.

2.2 |. Experimental design

The study was carried out by using three arms (Table S1). Arm A measured whether 1 hour of ES provided by the ES device (Checkpoint stimulator/locator; Checkpoint Surgical, Cleveland, Ohio) had an impact on healthy nerve. While this ES device delivers ES with a pulse width of 0 to 200 μs and frequency of 16 HZ similarly to previous ES protocols known to promote neuroenhancement,^23,31,32,37 it has not been used to provide therapeutic ES requiring 1 hour stimulation of nerve. In arm A, 12 rats were used to evaluate histological changes to nerve after ES. Animals were randomized to two experimental groups with two endpoints (n = 3/group/endpoint), ES 0.5 mA and ES 2.0 mA. The contralateral, untreated nerve served as control. In the ES 0.5-mA and ES 2.0-mA groups, the ES device was used to deliver 1 hour of intraoperative ES at two different current amplitudes, either 0.5 or 2.0 mA, at a pulse width of 100 μs and a frequency of 16 HZ. After ES, nerve was either immediately harvested for analysis (day 0) or harvested 3 days later (day 3). Wallerian degeneration and morphological changes to axons after damage would be observable by day three.⁷

The other two arms (arms B and C) evaluated the degree of therapy provided by ES or FK506 to nerve injury after repair. Arm B measured functional recovery after treatments, while arm C measured early indications of nerve regeneration with histology. These rats received a tibial nerve transection followed by immediate epineural repair, which is described under surgical procedures. For arm B, 36 rats were used for the evaluation of recovery, which consisted of walking track analysis, cold allodynia response, and assessment of relative muscle mass. Animals were randomized to four experimental groups (n = 9/group): ES 0.5 mA, ES 2.0 mA, FK506 as a positive control, and repair alone as a negative control. In the ES 0.5-mA and ES 2.0-mA groups, the ES device was again used as in arm A. The FK506 positive control group received daily subcutaneous administrations of FK506 (2 mg/kg subcutaneously; cat. No. F-4900; LC Laboratories, Woburn, Massachusetts) starting 3 days preoperatively and continuing daily for the duration of the experiment. Previous studies have shown this dosing regimen of FK506 to result in substantial improvement of axonal regeneration.²⁰ Fresh aliquots of FK506 were made every week by using ethanol and polyethoxylated castor oil (Sigma-Aldrich, St Louis, Missouri) as diluents and stored at −20°C until use. The repair alone group did not receive any intraoperative or postsurgical interventions after tibial nerve transection and repair (negative control group).

In arm C, 15 rats were used for histomorphometric analysis of the nerve distal to the repair site. The purpose of this arm was to evaluate early indications of nerve regeneration. Histological analysis was performed 21 days after tibial transection and repair in the ES 0.5-mA, FK506, and repair alone (n = 5/group) groups. Twenty one days was chosen because previous studies have shown that this time point maximizes the sensitivity to measure differences in nerve morphometry among groups when using the rat tibial nerve transection and repair model.³⁸ The experimental treatments were carried out just as described for arms A and B of the study.

2.3 |. Surgical procedures

Surgeries were performed by using aseptic technique and with the aid of an operating microscope. Anesthesia was delivered with a cocktail of ketamine (75 mg/kg; Zoetis, Kalamazoo, Michigan) and dexmedetomidine (0.5 mg/kg; Zoetis). During surgery, animals were placed on a warming pad for body temperature maintenance and given 1 mL of normal saline subcutaneously for hydration. The right tibial nerve was exposed by using a gluteal muscle-splitting approach. For arms B and C, the tibial nerve was transected 5 mm distal to the sciatic trifurcation and repaired immediately with 9–0 nylon suture (Sharpoint, Wyomissing, Pennsylvania). For the ES groups, a stainless steel 304 wire electrode (Component Supply Company, Sparta, Tennessee) was hooked around the tibial nerve 2 mm proximal to the cut and repair site. The nerve was hooked by fashioning the wire into a half circle and applying gentle upward tension to the wire to secure it to the nerve without compressing the nerve. A return current electrode was placed securely in musculocutaneous fascia in the dorsum of the animal. Electrical stimulation was delivered for 1 hour at either 0.5 mA or 2.0 mA of current by using a monopolar stimulation paradigm, depending on the experimental group. The device used provides feedback via an LED light indicating whether current has been provided, and the circuit is complete. Furthermore, for both the 0.5- and 2.0-mA currents, evidence of excitation based on the contractions of the foot, even after transecting the nerve, were used to demonstrate that current was provided to the nerve. After the completion of ES, the electrodes were gently removed from the nerve and fascia. Wounds were closed in a layered fashion with 4–0 vicryl suture for muscle and 4–0 nylon suture for skin (Ethicon, Somerville, New Jersey).

Animals were recovered with a subcutaneous injection of atipamezole HCl (Antisedan 1 mg/kg; Orion Corporation, Finland) and placed on a warming pad postoperatively. Postoperative pain and hypersensitivity were managed with a single dose of buprenorphine (Buprenorphine SR 1 mg/kg, ZooPharm, Windsor, Colorado) given intraoperatively. Animals were returned to the central housing facility and closely monitored for infection, distress, and other morbidities.

2.4 |. Walking track functional assessment

Walking track analysis was used to track the timeline of recovery after tibial nerve transection and repair. Rat hind feet were glazed with water-soluble, nontoxic paint before the animal was placed on a 10.5 × 61-cm plexiglass track lined with construction paper. Each animal walked down the track until at least three pairs of footprints with clear markings were obtained.

On each footprint, the following three parameters were measured with a digital caliper: external toe spread, measuring the distance between the first and fifth toes on the experimental (ETS) and contralateral sides (NTS); intermediary toe spread, measuring the distance between the second and fourth toes on the experimental (EIT) and contralateral sides (NIT); and print length, measuring the distance of the print from the distal end of the third toe to the heel on the experimental (EPL) and contralateral sides (NPL). These six variables were used to calculate the tibial functional index (TFI) with the following formula developed by De Medinaceli et al³⁹ and modified by Bain et al⁴⁰:

$$TFI=-37.2\frac{EPL-NPL}{NPL}+104.4\frac{ETS-NTS}{NTS}+45.6\frac{EIT-NIT}{NIT}-8.8$$ (1)

A more negative TFI indicates worse motor function. Smaller ETS and EIT and greater EPL result in a more negative TFI. More equivalent values on the experimental and normal sides produce a TFI that is closer in value to zero, indicating greater motor recovery. Animals were tested twice per week for 6 weeks. It is normal for this model to take up to 12 weeks to gain ~50% recovery of function based on walking track.^40,41

2.5 |. Cold allodynia testing

At 6 weeks postoperatively, animals were tested for cold allodynia response, similarly to previous methods.⁴² Each animal was placed on top of a finely meshed metal grid and allowed to acclimate for at least 5 minutes. A drop of acetone was released onto the bottom pad of the right foot by using a syringe without its needle. The animal’s behavior was monitored for 1 minute. Any paw-licking or paw-shaking behavior was deemed to be a response to the acetone. The total amount of time that the animal exhibited this behavioral response was measured.

2.6 |. Relative muscle mass

After the completion of walking track analysis and cold allodynia testing, animals were humanely killed at postoperative week 6. The gastrocnemius muscle was harvested from the experimental and contralateral sides. Wet muscle weight was recorded on each side, and the ratio of the ipsilateral to contralateral muscle weight was calculated.

2.7 |. Nerve histomorphometry

At 21 days postoperatively, animals in the third arm of the study were reanesthetized and prepared for exposure of the right tibial nerve. After tissue harvest, animals were humanely killed. En bloc specimens of the tibial nerve 5 mm distal to the repair site underwent histomorphometric analysis as previously described.⁴³ Nerve was harvested and stored in 3% glutaraldehyde (Polysciences, Warrington, Pennsylvania). The nerves were postfixed in 1% osmium tetroxide, serially dehydrated in ethanol and toluene, embedded in epoxy (Polysciences), and sectioned on an ultramicrotome into 1-μm cross-sections. Slides were counterstained with 1% toluidine blue dye. The slides were then analyzed at ×1000 on a Leitz Laborlux S microscope. The Leco IA32 Image Analysis System (Leco, St Joseph, Michigan) was used to quantitate nerve fiber counts, percentage neural tissue, fiber sizes, and myelin thickness. All analyses were performed by an observer blinded to the experimental groups.

2.8 |. Statistical analysis

All statistical analyses were performed in Prism 7 (GraphPad Software, LaJolla, California). Data are presented as mean ± SD. Data for walking track analysis were analyzed by using two-way analysis of variance (ANOVA), followed by Tukey’s multiple-comparisons test for post hoc analysis between groups at each time point. All other data were analyzed by using one-way ANOVA with Tukey’s comparisons for post hoc analysis between groups for each variable. The study was adequately powered (β = .80), and significance (α) was set at P < .05.

3 |. RESULTS

3.1 |. Healthy nerve morphometry was unchanged after ES delivery

During the intraoperative stimulation, the 0.5-mA constant current provided by the device was sufficient stimulus to provide muscle contractions for the entire 1-hour period. Additionally, 2.0 mA of current provided similar yet stronger muscle contractions. In the stimulated tibial nerve harvested immediately and 3 days after ES, there were no qualitative or quantitative differences in nerve structure or morphometric parameters compared with healthy controls (Figure 1 and Table 1). This conclusion was true regardless of the ES dose, 0.5 mA or 2.0 mA.

FIGURE 1.

Representative histological sections obtained 3 days after 1-hour electrical stimulation (ES) of healthy nerve. Uninjured tibial nerve was harvested 3 days after ES to evaluate whether there was any damage to nerve due to the 1-hour ES protocol. A, ES 0.5 mA. B, ES 2.0 mA. C, No ES (uninjured). Toluidine blue counterstained nerve cross-sections show myelinated axons in all groups where no detectable changes to axons or myelin are observed after ES. Scale bars = 20 μm [Color figure can be viewed at wileyonlinelibrary.com]

TABLE 1.

Histomorphometric data after ES of healthy tibial nerve

	ES 0.5 mA		ES 2.0 mA		No ESa
Outcome metric	Day 0	Day 3	Day 0	Day 3	Not applicable
Myelinated axon count	5615 ± 246	6013 ± 63	5756 ± 842	6105 ± 1689	5769 ± 596
P value vs no ES	>.05	>.05	>.05	>.05
Percentage nerve	51 ± 3	50 ± 8	55 ± 2	51 ± 11	56 ± 3
P value vs no ES	>.05	>.05	>.05	>.05
Nerve Density	14 536 ± 1743	14 191 ± 1297	18 320 ± 1066	17 384 ± 4868	17 170 ± 2041
P value vs no ES	>.05	>.05	>.05	>.05

Abbreviations: ES, electrical stimulation; FK506, tacrolimus.

^aUninjured.

3.2 |. Electrical stimulation or FK506 promoted more rapid functional recovery

All animals showed impaired function in the afflicted hind limb after the nerve injury. In addition, some motor recovery, while incomplete, was appreciated in all animals during the experimental period of 6 weeks after surgery. Baseline values of TFI on day 0 were not significantly different among groups. After injury, all groups demonstrated a significant change in TFI value, and no groups showed improvement in TFI values within the first 3 weeks after injury with repair. However, for both ES (0.5 mA and 2.0 mA) groups and the FK506 group, recovery as indicated by TFI values was significantly improved by day 25 postinjury compared with repair alone (Figure 2A; ES 0.5 mA, P < .05; ES 2.0 mA, P < .05; FK506, P < .05). For both ES groups and the FK506 group, this improved level of motor recovery compared with repair alone was maintained at study completion and day 42 (Figure 2B; ES 0.5 mA, P < .001; ES 2.0 mA, P < .05; FK506, P < .01) and reached nearly 50% recovery. The ES and FK506 groups’ TFI values did not differ from one another throughout the study course. Overall, both ES and FK506 therapies promoted more rapid functional recovery after tibial nerve transection with repair.

FIGURE 2.

Recovery after tibial nerve injury and repair measured by using walking track analysis. From walking tracks, the tibial functional index (TFI) was calculated to compare the motor recovery in each group. A more negative TFI value indicates worse function. A, Both electrical stimulation (ES) groups and tacrolimus (FK506) had increased TFI values compared with repair alone by day 25. B, Scatter plot of data at the day 42 end point. All data are mean ± SD, n = 9 per group. *P < .05 between ES 0.5 mA and repair alone, ^P < .05 between ES 2.0 mA and repair alone, ^#P < .05 between FK506 and repair alone [Color figure can be viewed at wileyonlinelibrary.com]

3.3 |. Cold allodynia response was unchanged with ES or FK506

In uninjured rat hind limbs, there is very limited or no response to cold acetone applied to the foot pad. However, after nerve injury, both ES groups and the FK506 group demonstrated increased responses to cold acetone compared with uninjured rats, indicating a degree of increased cold sensitivity (Figure 3A; P < .05). Neither the ES groups nor the FK506 group demonstrated any differences in cold allodynia responses compared with the repair alone group (Figure 3A; P > .05). However, while cold allodynia responses among experimental groups were not statistically significant from one another, their duration demonstrates a trend reflected from the individual data points whereby response is increased in ES and FK506 groups compared with repair alone.

FIGURE 3.

Outcomes 6 weeks after tibial nerve injury and repair. A, Both electrical stimulation (ES) groups and the tacrolimus (FK506) group demonstrated increased cold allodynia response time compared with uninjured nerve. No significant differences in cold allodynia response time were observed between the experimental groups. Uninjured nerve data are represented by dotted lines (mean in black and SD in grey) *P < .05. B, Only the FK506 group showed increased relative muscle mass compared with repair alone group. *P < .05 vs repair alone group. All data are mean ± SD, n = 9 per group [Color figure can be viewed at wileyonlinelibrary.com]

3.4 |. Relative muscle mass was unchanged by ES

The FK506 group demonstrated a larger relative muscle mass compared with the repair alone group (Figure 3B; P < .05). No significant difference in relative muscle mass was seen in either ES group compared with repair alone.

3.5 |. Electrical stimulation or FK506 increased the number of regenerated myelinated axons distal to the repair site

For ES treatments, because the 0.5-mA ES group demonstrated that this level of ES can provide neural stimulation and still improve functional recovery, histological evaluation was considered for only the 0.5-mA ES group. Improvements in nerve regeneration were observed in ES 0.5-mA and FK506 treatment groups. The total nerve fiber counts (number of myelinated axons) were increased in both FK506 and ES 0.5-mA groups compared with the repair alone group (Table 2). No significant differences between groups were observed in percentage neural tissue and nerve density (Table 2) or in myelin or axon thickness (data not shown). Representative histological sections from each group are presented in Figure 4.

TABLE 2.

Histomorphometric data 3 weeks posttibial nerve injury and repair

Outcome metric	ES 0.5 mA	FK506	Repair alone
Myelinated axon count	3671 ± 192	4022 ± 967	2173 ± 683
P value vs repair alone	<.05	<.005
Percentage nerve	6.28 ± 2.52	4.57 ± 2.52	3.21 ± 2.23
P value vs repair alone	>.05	>.05
Nerve density	6790 ± 2372	5043 ± 2676	3572 ± 2580
P value vs repair alone	>.05	>.05

Abbreviations: ES, electrical stimulation; FK506, tacrolimus.

FIGURE 4.

Representative histological sections obtained 3 weeks after surgery. Distal nerve was assessed 5 mm from the tibial nerve repair site from the following groups: A, Repair alone. B, Tacrolimus (FK506). C, Electrical stimulation (ES) at 0.5 mA. Toluidine blue counterstained nerve cross-sections show myelinated axons in all groups, but the repair alone group contained the fewest myelinated axons. Scale bars = 20 μm [Color figure can be viewed at wileyonlinelibrary.com]

4 |. DISCUSSION

Our hypothesis that supplementing nerve repair with either ES or FK506 would improve regeneration and recovery compared with repair alone was not rejected. This finding was not intuitive because the mechanisms by which these neuroenhancing therapies promote their effects are slightly different.^44,45 FK506 can bind immunophilin receptors contained on neurons, leading to activation of heat-shock proteins and, in turn, increased expression of growth-associated genes, such as growth-associated protein (GAP)-43, and activation of the mitogen-activated protein kinase (originally called extracellular signal-regulated kinase) pathway.^14,46–48 Electrical stimulation alternatively affects neurons by stimulating rapid calcium influx, leading to increased cyclic adenosine monophosphate and, in turn, increased expression of growth-associated genes, such as GAP-43, and growth factors and their receptors, such as brain-derived neurotrophic factor and tropomyosin receptor kinase B (also known as tyrosine receptor kinase B).^30,32 In addition, while FK506 increases the rate of axonal growth, ES instead increases the rate at which axons cross a suture site during nerve repair.^30,48 While the mechanisms differ, a common theme to both treatments is an anticipated outcome whereby a greater number of axons reach their end-organ target more rapidly. Therefore, the recovery outcomes from our comparison of the treatments seem rational. However, because FK506 is regarded as a potent neuroregenerative drug with serious systemic side effects,⁴⁹ that ES has few differences compared with FK506 in its neuroregenerative effects has significant implications for the translational impact of ES as a therapy to augment nerve surgery because no known side effects due to ES therapy have been discussed in the literature to date.

Clinically significant outcomes, such as muscle function and recovery, are perhaps most meaningful when translational impact is considered. The rapid improvement in motor function as measured by walking track analysis occurred without a corresponding increase in relative muscle mass in the ES groups compared with repair alone. This finding that ES improves functional recovery without causing a significant increase in relative muscle mass is in line with results from previous literature on ES.^35,50 However, the improvement of outcome in ES groups based on walking track analysis are corroborated by the results of histological analysis. We found an increase in total nerve fiber counts 5 mm distal to the repair site in the ES group compared with repair alone by 21 days, which suggests a rationale for the observed improvements in recovery. Again, our findings suggest that, as a larger number of axons reach the muscle in a shorter time, functional recovery is achieved more rapidly. Our histological findings also correspond well to the literature regarding how ES can potentially confer benefits to regeneration and recovery.^10,51 Although previous studies have shown increased numbers of retrograde labeled neurons regenerating axons after ES,²³ this is the first study to show an increase in myelinated axon number with ES by using a morphometric outcome.

While ES holds great promise, data regarding its clinical use are limited, and standardized guidelines to facilitate an effective translation of this technology to the clinic have not been developed.^37,51 While we found in a rat model that both 0.5- and 2.0-mA current improved functional recovery after nerve injury repair, the findings regarding the use of 0.5-mA current have unique applicability for translation. Specifically, 0.5 mA of current provided to a rat sciatic nerve was sufficient to produce mild muscle contractions, while 2.0 mA of current elicited strong muscle contractions. This distinction may be important because previous therapeutic ES protocols applied to rodent or human nerves usually provide ES at supramaximal levels sufficient to produce nerve stimulation and/or mild muscle contractions.^23,51 Therefore, the 0.5-mA current protocol in a rat represents an ES therapy protocol design whereby the lowest dose of ES that elicited neural stimulation still provided neuroenhancement. This concept is important because ES protocols developed for the clinic may not entirely mirror ES doses provided to rats. However, our data suggest that ES protocols sufficient to provide some nerve stimulation will yield therapeutic benefits. The strong performance of ES in our study provides an impetus for further assessing therapeutic ES in other animal or nerve injury models to measure the potential for more rapid or greater recovery.

In conclusion, we found that relatively low-dose (0.5–2.0 mA) intraoperative ES for 1 hour increases nerve regeneration and functional recovery in a rat model of tibial nerve transection and repair. Functional and histological outcomes were not different between the ES 0.5 mA and FK506 groups. Because of this comparison to FK506, the use of ES as a therapy to augment nerve repair may have considerable benefits.

Abbreviations

ANOVA

analysis of variance

EIT

distance between the second and fourth toes on the experimental side

EPL

distance of the print from the distal end of the third toe to the heel on the experimental side

ES

electrical stimulation

ETS

distance between the first and fifth toes on the experimental side

FK506

tacrolimus

GAP-43

growth-associated protein 43

NIT

distance between the second and fourth toes on the contralateral side

NPL

distance of the print from the distal end of the third toe to the heel on the contralateral side

NTS

distance between the first and fifth toes on the contralateral side

TFI

tibial functional index