Table of Contents Author Guidelines Submit a Manuscript
Neural Plasticity
Volume 2018, Article ID 7975013, 14 pages
https://doi.org/10.1155/2018/7975013
Research Article

Botulinum Neurotoxin Application to the Severed Femoral Nerve Modulates Spinal Synaptic Responses to Axotomy and Enhances Motor Recovery in Rats

Department of Otorhinolaryngology, Jena University Hospital, Am Klinikum 1, 07747 Jena, Germany

Correspondence should be addressed to Andrey Irintchev; ed.anej-inu.dem@vehctniri.yerdna

Received 6 March 2018; Revised 15 July 2018; Accepted 5 August 2018; Published 5 September 2018

Academic Editor: Laura Baroncelli

Copyright © 2018 Marcel Irintchev et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Botulinum neurotoxin A (BoNT) and brain-derived neurotrophic factor (BDNF) are known for their ability to influence synaptic inputs to neurons. Here, we tested if these drugs can modulate the deafferentation of motoneurons following nerve section/suture and, as a consequence, modify the outcome of peripheral nerve regeneration. We applied drug solutions to the proximal stump of the freshly cut femoral nerve of adult rats to achieve drug uptake and transport to the neuronal perikarya. The most marked effect of this application was a significant reduction of the axotomy-induced loss of perisomatic cholinergic terminals by BoNT at one week and two months post injury. The attenuation of the synaptic deficit was associated with enhanced motor recovery of the rats 2–20 weeks after injury. Although BDNF also reduced cholinergic terminal loss at 1 week, it had no effect on this parameter at two months and no effect on functional recovery. These findings strengthen the idea that persistent partial deafferentation of axotomized motoneurons may have a significant negative impact on functional outcome after nerve injury. Intraneural application of drugs may be a promising way to modify deafferentation and, thus, elucidate relationships between synaptic plasticity and restoration of function.

1. Introduction

Injury to peripheral nerves in adult mammals causes deafferentation of the axotomized motoneurons, a phenomenon known as “synaptic stripping” [1]. Synaptic terminals are removed from cell bodies and dendrites of motoneurons by activated microglial and astroglial cells [16]. The overall posttraumatic loss is reversed to a large extent if muscles become reinnervated [3, 6, 7], but restoration of some synaptic inputs is incomplete [811]. Such deficits, for example, in cholinergic and glutamatergic innervation, may contribute to functional deficits after muscle reinnervation as they are well correlated with functional performance after long-term reinnervation [9, 12].

Here, we pursued to influence synaptic responses after peripheral nerve injury and, thus, eventually alter the outcome by using botulinum neurotoxin A (BoNT) or brain-derived neurotrophic factor (BDNF). When applied intramuscularly, BoNT blocks synaptic transmission at the neuromuscular junction and, in addition, is transported retrogradely to the motoneuron cell body and possibly also transcytosed to afferent synaptic terminals [1316]. BoNT causes progressive synaptic stripping detectable at 4 days after intramuscular injection and abolishes excitatory and inhibitory synaptic transmission on motoneurons at 1-2 weeks after application [17]. Rather than intramuscularly, we applied BoNT to the proximal nerve stump immediately after nerve transection similar to the application of retrograde tracers assuming that this type of application will enhance synaptic stripping similar to intramuscular BoNT application. In other animals, we applied BDNF to the proximal stump of the freshly cut nerve hoping to achieve an effect opposite to that of BoNT, that is, attenuation of synaptic loss. When administered to cut proximal axons immediately after transection, BNDF reduces synaptic stripping and enhances recovery of tonic firing of regenerating motoneurons [18]. Synaptotrophic effects of exogenous BDNF have also been reported after ventral root avulsion [19]. Finally, a single session of brief electrical stimulation (20 Hz, 1 hour) of the proximal stump of the freshly transected femoral nerve in rats leads to enhanced nerve regeneration over weeks and this effect is apparently associated with an upregulation of BDNF and its cognitive receptor TrkB in the motoneuron cell body [20, 21]. It is possible, though not proven, that this enhanced BDNF signaling leads to, among other mechanisms, better regeneration via synaptotrophic effects. We measured the effects of BoNT or BDNF application using stereological estimates of chemically defined nerve terminal densities in motor nuclei, a motor recovery test, and retrograde labeling of motoneurons. For this first experiment using intraneural drug application, we selected the femoral nerve model in rats for a practical reason: the anatomy in this model allows work with a longer proximal trunk after nerve transection as compared with, for example, the facial nerve and, thus, easier application of BoNT or BDNF solutions to the severed nerve using plastic mini cups. The well-established femoral nerve model is a valuable alternative to other spinal nerve models like the sciatic one offering the possibility to analyze precision of target reinnervation, reliable functional assessments, and a straightforward search of anatomical deficits and structure-function correlations [22]. Helpful for this study was also previous data on long-term functional recovery, precision of motor reinnervation, and correlations between these measures after section/suture of the femoral nerve in adult rats [23].

2. Materials and Methods

2.1. Animals and Experimental Design

Ten-week-old female Wistar Unilever rats () from Charles River Laboratories (Sulzfeld, Germany) were used. To monitor short-term numerical changes in synaptic terminal populations, retrograde neuronal tracer (Fluoro-Gold, FG) was injected unilaterally into the quadriceps muscles of 20 animals (experiment I). Four days later, the femoral nerve on the injected side was cut and solutions containing bovine serum albumin (BSA), BoNT, or BDNF were applied to the proximal nerve stump (5 rats per group, see details on application below). Synaptic populations in the quadriceps motor nucleus, defined by the retrograde labeling, were studied one week after nerve transection. The rest five rats served as an “intact” control, that is, they were similarly treated and analyzed with the exception of nerve injury. To analyze long-term synaptic alterations, the rats in experiment II were subjected to nerve lesion and application of BSA (), BoNT type A (), or BDNF (). Intramuscular (i.m.) injections of FG were performed two months after injury followed by, one week later, video recordings for single-frame motion analysis (SFMA) and tissue sampling for synaptic terminal analyses. Analysis of long-term functional effects was done in experiment III. After nerve injury and application of BSA (), BoNT (), or BDNF (), the animals were repeatedly video recorded over a 20-week observation period and then subjected to retrograde labeling of motoneurons regenerated beyond the injury site to analyze “preferential motor reinnervation” [24]. The animals were housed under standard conditions and received food and water ad libitum. Visual examinations for complications like BoNT-induced muscle paralysis, abnormal grooming, or self-mutilations were performed regularly (once daily in the first week, once or twice weekly at later time periods). Such complications were not observed. Experiments were performed according to the animal protection laws of Germany and the European Community. Experiments were blinded.

2.2. Surgery and Drug Application

Rats were anesthetized with fentanyl (Fentanyl Janssen, Janssen, Neuss, Germany, 0.005 mg/kg i.m.), midazolam (Dormicum-R, Roche, Basel, Switzerland, 2 mg/kg i.m.), and medetomidine (Domitor-R, Orion Pharma, Espoo, Finland, 0.15 mg/kg i.m.). The trunk of the right nerve was exposed under an operation microscope and cut at approximately 7 mm proximal to the bifurcation of the saphenous and quadriceps muscle branches (Figure 1(a)). The proximal nerve stump was inserted for 30 min into a cup containing 0.1% BSA (Sigma, Taufkirchen, Germany) in saline, 100 U/ml BoNT (Xeomin, Merz Pharma, Frankfurt, Germany), or 20 μg/ml human recombinant BDNF (Biomol, Hamburg, Germany) in 0.1% BSA saline (Figure 1(b)). As a rough orientation for the drug concentrations served previous in vivo studies on synaptic effects using BoNT [13, 17] and BDNF [18]. The cups were cut from standard yellow pipette tips after their distal ends were heat-sealed using a lighter (Figure 1(b), capacity ~10 μl). After drug treatment, the nerve trunks and their surroundings were thoroughly rinsed with saline and the nerve ends were aligned using two epineural 10–0 sutures (Ethicon, Norderstedt, Germany). Finally, the skin was closed with 4–0 sutures (Ethicon) and the rats received subcutaneously an antidote cocktail consisting of atipamezole (Antisedan, Orion Pharma, 0.75 mg/kg), flumazenil (Anexate, Roche, 0.2 mg/kg), and naloxone (Naloxon, CuraMed Pharma, Karlsruhe, Germany, 0.12 mg/kg).

Figure 1: Drug application to the severed nerve. (a) The right femoral nerve trunk (arrow) prior to nerve injury. Proximally, the nerve is fixed by an epineural suture (short arrow) to the nearby muscle aponeurosis to prevent withdrawal of the proximal stump after nerve cut. Seen are also the 10–0 thread (upper arrowhead) used to fix the nerve and its needle (lower arrowhead), as well as the femoral vein (V). (b) The femoral nerve is transected, and the proximal stump is inserted in a self-made cup (T, see Materials and Methods) filled with drug solution. The distal nerve stump is marked by an asterisk.
2.3. Single-Frame Motion Analysis (SFMA)

SFMA was performed as described previously [23]. Briefly, the rats (experiments II and III) were video recorded prior to nerve injury from behind and from the left and right side during walking along a wooden plate (1500 mm long, 120 mm wide, and 20 mm thick) using a video camera (100 frames per second, Pike F-032, Allied Vision Technologies, Stadtroda, Germany). The video recordings were repeated 8 weeks (experiment II) or at 1, 2, 4, 8, 12, 16, and 20 weeks (experiment III) after injury. At least three walking trials were recorded per rear, left and right side view of each animal per time point. Analyses were performed using noncommercial software packages: VirtualDub 1.6.19 (http://www.virtualdub.org) and Image Tool 3.0 (University of Texas Health Science Center at San Antonio, TX, USA, http://compdent.uthscsa.edu/imagetool.asp). Two parameters were measured: the foot-base angle (FBA) and the step length ratio (SLR). The FBA is measured at toe-off position on the side ipsilateral to injury as an angle between the line dividing the sole surface into two halves and the horizontal line (minimum of 3 measurements per animal and time point). The SLR is calculated as ratio of the lengths of two successive steps (minimum of 6 SLR values per animal and time point). Using the FBA and SLR values, two additional parameters were calculated: (1) the product FBA × SLR and (2) the FBA × SLR recovery index [23].

2.4. Retrograde Labeling of Motoneurons

To label the quadriceps motor nucleus (experiments I and II), 125 μl of 1% Fluoro-Gold (Fluorochrome, Denver, CO, USA) in saline was injected into the right quadriceps muscle without anesthesia of the rats (Figure 2(a)). For analysis of “preferential motor reinnervation” [23], 20 weeks after injury, the rats in experiment III were anesthetized as described above. The quadriceps and the saphenous branches were cut approximately 5 mm distal to the bifurcation. Fluoro-Ruby (tetramethylrhodamine dextran, MW 10,500, Molecular Probes/Life Technologies, Darmstadt, Germany) and Fluoro-Emerald (fluorescein dextran, MW 10,000, Molecular Probes) crystals were applied for 30 min to the proximal stumps of the quadriceps and the saphenous branch, respectively. Labeling was considered successful if no leakage of dye beyond the parafilm sheaths underlying the nerve ends was noticed after the 30 min application period. Six days later, the rats were anaesthetized and perfused with 4% formaldehyde in 0.1 M sodium cacodylate buffer, pH 7.3. The lumbar spinal cords were removed, postfixed overnight, and cut transversely (serial sections of 40 μm thickness) on a cryostat (CM1850, Leica Microsystems, Wetzlar, Germany). The sections were collected on SuperFrost Plus glass slides (Carl Roth, Karlsruhe, Germany) and coverslipped using Fluoromount G (Southern Biotechnology Associates/Biozol, Eching, Germany). Counting was based on stereological principles and done on an Axiophot 2 fluorescence microscope [25].

Figure 2: Images of synaptic terminals and Iba1+ cells in the quadriceps motor nucleus. (a-b) A section containing back-labeled cell bodies of femoral motoneurons (a, arrows) is additionally stained for nuclei (a) and VGAT (b). The boundary of the quadriceps motor nucleus is indicated by a dotted line. Scale bar = 100 μm for (a-b). (c–e) VGAT+ and VGLUT 2+ axonal terminals (c, e) and VGLUT1+ varicosities (arrows, d). Scale bar = 10 μm for (c–e). (f) ChAT staining of two motoneuron cell bodies (MN) surrounded by cholinergic terminals (arrows). Counted were terminals around the MN soma with a visible nucleus (pale area in the center of the MN on the right hand side) which were in focus (thick arrows). Terminals out of focus or only partially seen in the focus plane (thin arrows) were not counted. No quantification was undertaken for the second MN profile (on the left hand side) since it had no visible nucleus. The arrowhead points to a ChAT+ cross-sectional profile of a dendrite close to the MN cell body. Such “perisomatic” dendritic profiles could be traced for long distances throughout the section thickness in contrast to the limited extent of the perisomatic terminals in the -axis. (g) Iba1+ cells (arrows) some of which surround a motoneuron cell body (MN). Scale bar indicates 25 μm and 50 μm for panels (f) and (g), respectively. (a–g) Shown are representative images from tissue sections after different treatments to illustrate the quality of each staining which was similar in all experimental groups and time points.
2.5. Immunofluorescence

Tissue processing and staining were performed as previously described [26]. Under anesthesia (see above), the rats were perfused with 4% formaldehyde in 0.1 M cacodylate buffer, pH 7.3, for 15 min at room temperature (RT). The lumbar spinal cords were then postfixed in the same fixative overnight at 4°C and cryoprotected by infiltration with 15% sucrose in cacodylate buffer for 2 days at 4°C. The samples were frozen in precooled 2-methyl-butane (isopentane, −80°C) for 2 min and stored in liquid nitrogen until sectioned. Transverse sections of 25 μm thickness were obtained using a cryostat (CM1850, Leica Microsystems, Wetzlar, Germany) such that 6 spaced serial sections 250 μm apart were present on each slide. Immunofluorescence staining was performed after antigen retrieval (30 min at 80°C in 10 mM sodium citrate solution, pH 9.0). Nonspecific binding was blocked for 1 hour at RT with phosphate-buffered saline (PBS, pH 7.3) containing 0.2% Triton X-100 (Sigma), 0.02% sodium azide (Sigma), and 5% normal serum (Jackson ImmunoResearch Europe, Suffolk, UK) from the species in which the secondary antibody was raised (Table 1). The primary antibodies were diluted in PBS containing 0.5% lambda-carrageenan (Sigma) and 0.2% sodium azide and applied to the sections for 3 days at 4°C (Table 1). Cy3-conjugated secondary antibodies, diluted in PBS containing 0.5% lambda-carrageenan and 0.2% sodium azide, were applied for 2 hours at RT (Table 1). Cell nuclei were stained for 10 min at RT with bis-benzimide solution (Hoechst 33258 dye, 5 μg ml−1 in PBS, Sigma). For each antigen, all sections were stained in the same primary and secondary antibody solutions stabilized by the nongelling vegetable gelatin lambda-carrageenan and kept in screw-capped staining plastic jars (capacity 35 ml, 10 slides, Carl Roth). This method enables repeated long-term usage and high reproducibility of the immunohistochemical staining [2628]. Staining controls included omitting the first antibody or replacing it by normal serum or IgG. These controls were negative. Examples of immunohistochemical stainings are shown in Figures 2(b)2(g).

Table 1: Antibodies used for immunohistochemistry.
2.6. Quantitative Immunohistochemical Analyses

Quantitative analyses were performed using the Stereo Investigator 8.1 software (MicroBrightField Europe, Magdeburg, Germany) and a fluorescence microscope (Axioskop 2 mot plus, Zeiss, Oberkochen, Germany) equipped with a motorized stage (Zeiss) and a CX 9000 digital camera (MicroBrightField) as described [9, 12]. Cell and synaptic terminal densities were estimated using the optical disector in every 10th spaced serial section (250 μm apart) in which back-labeled femoral motoneurons were visible (Figure 2(a)). The boundaries of the quadriceps motor nucleus were outlined (Plan Neofluar 5x objective, Zeiss, Figure 2(a)), and cell or synaptic terminal densities () were estimated using randomly placed disectors. For VGAT+ (Figures 2(b) and 2(c)), VGLUT1+ (Figure 2(d)), and VGLUT2+ terminals (Figure 2(e)), the disectors had a 100 μm2 base and a 5 μm height with an interdisector spacing of 100 μm. Individually discernible immunopositive puncta were counted using a Plan Neofluar 100x oil objective (Zeiss). For Iba1+ cells (Figure 2(g)), the size of the disectors was 3600 μm2 base and 10 μm height and the spacing between disectors was 100 μm.

Analyses of cholinergic perisomatic terminals were performed on ChAT-immunostained sections using the Stereo Investigator (Figure 2(f), [9]). All motoneuron profiles with discernible nucleus in a quadriceps motor column transect were analyzed. Each motoneuron, visualized at 100x magnification, was focused at the level of its largest cell body cross-sectional area, and its cell body perimeter and number of perisomatic terminals were determined (Figure 2(f)). Frequency of perisomatic ChAT+ terminals was calculated as number of perisomatic terminals per unit perimeter length. Mean values of individual animals were used to calculate group mean values.

2.7. Statistical Analyses

Data were analyzed using one-way analysis of variance (ANOVA) or two-way ANOVA for repeated measures followed by Holm-Sidak multiple comparison tests (SigmaPlot 12, SPSS, Chicago, IL, USA). Regression analyses were performed using SigmaPlot. The threshold value for acceptance of differences was 5%.

3. Results and Discussion

3.1. Short-Term Effects on Synaptic Terminal Numbers

We initially tested whether intraneural drug applications alter short-term synaptic responses to nerve injury in the spinal motor nucleus (experiment I). We estimated the effects of nerve injury and application of BSA as compared to rats without nerve lesions (“BSA” versus “Uninj.” in Figure 3) using antibodies against synaptic terminal markers (Table 1). Numbers of microglial cells were also analyzed since these cells are activated after injury and are involved in synaptic remodeling [2931]. The observed effects included reduced density of excitatory VGLUT2+ terminals (−20%, Figure 3(a)), increased density of Iba1+ microglia (+267%, Figure 3(b)), and decrease in modulatory perisomatic ChAT+ terminals (−36%, Figure 3(b)). Inhibitory VGAT+ and excitatory VGLUT1+ Ia boutons were not significantly affected (+2% and +13%, resp., Figures 3(a) and 3(b)). Assuming that BSA has no measurable influence on these variables, the differences found between the two groups represent axotomy-related responses. In line with this notion is the finding of similar changes in the rat facial nucleus 1 week after axotomy [9]. Compared with BSA, BDNF had only one effect: attenuation of injury-induced ChAT+ terminal loss (Figure 3(b)). A similar protective effect on ChAT+ terminals had also BoNT (Figure 3(b)). In addition, BoNT application resulted, again as compared with BSA, in increased density of VGAT+ terminals (+35%) and reduced density of VGLUT1+ boutons (−46%), while VGLUT2+ terminals and Iba1+ cells were not significantly affected (−9% and 0%, resp., Figure 3(a)).

Figure 3: Analysis of synaptic terminals and microglia in the quadriceps motor nucleus 1 week after femoral nerve injury and drug application. Included are also values from control rats without nerve injury and drug treatment (“Uninj.”). Shown are numerical densities (number per unit volume) of VGAT+, VGLUT1+, and VGLUT2+ terminals and Iba1+ microglial cells, as well as frequency (number per unit length) of ChAT+ perisomatic terminals (mean values + SEM). Asterisks indicate mean values significantly different from all other groups (one-way ANOVA, 3,16 = 4.87–44.8, ) with Holm-Sidak post hoc tests (). per group.

To test if the BoNT effects could be related to its retrograde transport into the spinal cord, we performed immunohistochemistry for BoNT-cleaved SNAP-25 (SNAP-25197) which labels sites of BoNT proteolytic activity [32]. One week after nerve injury and BoNT application, immunofluorescence labeling was present around back-labeled somata and in the neuropil of the femoral motor nucleus (Figure 4). This pattern of labeling is similar to that previously observed by other groups [13, 14] and suggests that BoNT action has been transported into the spinal cord and could possibly be active in afferent terminals.

Figure 4: Cleaved SNAP-25 staining of a spinal cord section one week after injury and BoNT application. Immunostaining (a) is seen around the somata of back-labeled motoneurons and in the neuropil among them (b, c). Scale bar = 50 μm.

Overall, these findings show that the drug applications altered some synaptic responses to axotomy. Our working hypothesis was (see Introduction) that BDNF would have synaptotrophic effects and, indeed, injury-related loss of ChAT+ perisomatic boutons was prevented. At the same time, however, other major inputs, excitatory VGLUT2+ and inhibitory VGAT+ terminals, were not affected as initially hypothesized. It is possible that the intracellular concentration of active exogenous BDNF achieved in our experiment has not been optimal to produce pronounced, long-term effects. BDNF appears to have a dose-dependent influence on nerve regeneration, that is, facilitation at low doses and inhibition at higher ones [33]. Therefore, we do not assume that BDNF is inefficient in our model unless this proves true in a future dose-dependence study.

In contrast to BDNF, we expected that BoNT would enhance loss of terminals after axotomy with a more pronounced effect on excitatory (VGLUT1+ and VGLUT2+) than on inhibitory (VGAT+) terminals [34, 35]. This appeared true for VGLUT1+ terminals, but the effects on VGAT+ and ChAT+ terminals were, on the opposite, synaptotrophic (Figure 3). This heterogeneity of effects suggests also other mechanisms of action in addition to inhibition of synaptic vesicle exocytosis by cleaving SNAP-25 [35]. It is possible, for example, that the increase in inhibitory VGAT+ terminals results from inhibition of some of these heterogeneous in origin terminals [36] and subsequent sprouting of unaffected inhibitory axons. Partial inhibition and reactive sprouting could also affect the cholinergic input to motoneurons. Alternatively or in addition, it is possible that BoNT has neurotrophic effects achieved via colocalization and signaling through the p75 receptor [15, 37]. This notion is not necessarily in disagreement with the limited effects of BDNF described above since different receptors (p75 versus TrkB) and neurotrophins may be involved.

3.2. Long-Term Synaptic Effects and Recovery of Function

We further investigated whether drug-related synaptic alterations persist after a longer reinnervation period, two months after injury (experiment II). We found, again compared with a BSA control group, that the BDNF effect on ChAT+ terminals at 1-week post injury has disappeared while a previously nonexisting deficit in VGLUT1+ terminals was now present (Figures 5(a) and 5(b)). BoNT-related differences in VGAT+ and VGLUT1+ terminal numbers had also disappeared at two months after injury, but the ChAT+ terminal frequency was still higher similar to 1 week after lesion (Figures 5(a) and 5(b)). Immunohistochemistry for cleaved SNAP-25 in the spinal cord at two months after injury showed labeling similar to the one observed at 1 week (data not shown). This observation suggests that BoNT enzymatic activity is present for a long period of time after application.

Figure 5: Analysis of synaptic terminals and microglia in the quadriceps motor nucleus two months after femoral nerve injury and drug application. Asterisks indicate mean values significantly different from all other groups (one-way ANOVA, 2,16 = 11.4 and 30.4, and 0.001 for VGLUT1 and ChAT, resp.) with Holm-Sidak post hoc tests (). per group. Note that numbers of Iba1+ cells and ChAT+ terminals (b) and numbers of VGLUT1+ terminals (a) in BSA-treated animals are much lower than these at 1 week after injury (Figures 3(a) and 3(b)). This is consistent with previous findings [9, 10].

Functional analysis performed in the same animal groups revealed significantly lower foot-base angle (FBA) and step length ratio (SLR) in the BoNT group as compared to BSA- and BDNF-treated rats (Figure 6(a)). This finding indicates better functional recovery as both parameters increase after injury and decrease as reinnervation and recovery proceed (see Figures 7(a) and 7(b)). Regression analysis did not indicate any significant statistical relationship between individual structural parameters (Figure 5) and functional measures (Figure 6(a)) with the exception of ChAT+ terminal densities (Figures 6(b)6(d)). Higher frequencies of cholinergic perisomatic terminals appeared to be associated with lower (“better”) functional values. The coefficients of determination (, values shown in Figures 6(b)6(d)) indicate that some 70% of the variability in functional parameters may be explained, in statistical terms, by variability in numbers of ChAT+ terminals. Previous work using facial nerve or spinal cord injury models has also shown strong statistical relationships between degree of functional recovery, on one side, and degree of preservation/recovery of ChAT+ terminal frequency on facial [9, 12] or spinal motoneurons [3840], on the other side. These large cholinergic terminals form C-type synapses on motoneuronal perikarya and proximal dendrites and utilize M2 muscarinic receptors for acetylcholine in the postsynaptic membrane [4146]. Although not that numerous, these synapses strongly influence motoneuron function by regulating action potential after hyperpolarization in a way that, under normal conditions, ensures sufficient motoneuron output to drive motor behavior [47, 48]. We can, therefore, assume that partial loss of perisomatic cholinergic terminals, associated with a reduced expression of postsynaptic receptors [49, 50], may significantly impair motor behaviors such as walking, whisking, and blinking [51].

Figure 6: Motor recovery and correlations between functional parameters and ChAT terminal frequency two months after femoral nerve lesion and drug application. (a) Shown are mean values + SEM of foot-base angle (FBA) on the operated side and step length ratio (SLR). , 7, and 7 for BSA, BDNF, and BoNT, respectively. For both parameters, one-way ANOVA showed effects of treatment (2,17 = 18.4 and 38.0 for FBA and SLR, respectively, for both parameters). The BoNT group mean values were significantly different from the values of the BSA and BDNF groups (asterisks, , Holm-Sidak test). (b–d) Individual values of functional parameters plotted against numbers of ChAT terminals. Shown are regression lines, coefficients of determination (), and probability values ().
Figure 7: Time course and degree of motor recovery after femoral nerve lesion and drug application. Shown are mean values ± SEM of foot-base angle on the operated side (FBA, a), step length ratio (SLR, b), product FBA × SLR (c), and recovery index for the product FBA × SLR (d) prior to injury (0 week) and 1–20 weeks p.o. The dashed horizontal line in (d) is drawn at 100%, a value indicating full degree of recovery. , 8, and 9 for BSA, BDNF, and BoNT, respectively. For all parameters shown, two-way ANOVA for repeated measures showed effects of time (7,147 = 52.4–209, ) and treatment (2,21 = 9.51–15.6, ). Indicated by symbols are group mean values significantly different from the corresponding postoperative values of the BSA and BDNF groups and # the corresponding value of the BSA group (, Holm-Sidak post hoc procedure).
3.3. Long-Term Functional Effects

Finally, we were interested whether functional effects of drug application could appear later or earlier than the analyzed postinjury time point (two months), a time period when reinnervation and recovery are well advanced but not completed. We performed experiment III in which rats were treated similarly to experiment II but monitored functionally between the first and the 20th week after injury. Time course and degree of recovery were very similar between BSA- and BDNF-treated animals (Figures 7(a)7(d)) and in agreement with previous observations after transection and suture of the femoral nerve in adult rats [23]. In contrast, recovery after BoNT application was accelerated between the 2nd and 12th week (Figures 7(a)7(d)) and advantages of this treatment were even present at the final time point studied, 20 weeks (Figure 7(a)).

After the 20-week observation period, the animals in experiment III were subjected to retrograde labeling to assess precision of reinnervation (Figures 8(a)8(c)), a factor that can influence the functional outcome after femoral nerve injury and regeneration in rats [23]. The numbers of motoneurons projecting into the appropriate quadriceps nerve only, into the inappropriate saphenous nerve, or into both nerves (“Muscle,” “Skin,” and “Both” in Figure 8(d), resp.) were similar in the three groups of rats. This finding suggests that the functional improvements seen in the BoNT group are not related to an enhanced preferential reinnervation of the muscle. This notion is supported by the lack of significant covariations between numbers of back-labeled motoneurons and functional parameters.

Figure 8: Retrograde labeling of motoneurons 20 weeks after lesion. (a–c) Representative images of motoneurons back-labeled through the muscle (quadriceps) and the skin (saphenous) branch of the femoral nerve (“Muscle” and “Skin”) using Fluoro-Ruby and Fluoro-Emerald (red and green fluorescence), (a) and (b), respectively, overlay in (c). Scale bar = 100 μm. (d) Quantitative analysis of retrogradely labeled cells including double-labeled motoneurons (“Both”). Shown are mean values + SEM. One-way ANOVA showed no effect of treatment on any of the motoneuron categories (2,18 = 0.95–1.14, –0.533). animals per group.
3.4. Possible Mechanisms of Drug Effects

We applied BoNT only once using the time frame between axonal membrane damage and sealing to load the proximal axon and cell body with toxin similar to retrograde tracers (Figures 8(a)8(c)). Our expectation was that this uptake will be sufficient to “prime” the initial responses of motoneurons to injury, in particular their deafferentation, and, thus, eventually achieve long-term effects on regeneration without need of repeated drug delivery to the injury site. As estimated by gait analysis, our experiment was successful as functional regeneration was enhanced already at two weeks after injury and recovery remained accelerated for months thereafter. Enhancement of axonal regrowth in the crushed sciatic nerve of mice by a single low-dose intraneural application of BoNT has been just reported, but the underlying mechanisms for these effects have remained unclear [37]. Here, we propose that the improvement of regeneration in our model is a consequence of attenuated loss of cholinergic modulatory input to femoral motoneurons (Table 2). In addition, it is possible that BoNT has an additional neuroprotective effect. At one week after injury, we found, compared with control rats, an increase in VGAT+ inhibitory afferents in the quadriceps motor nucleus, reduced numbers of excitatory VGLUT1+ Ia afferents, and no change in excitatory VGLUT2+ terminals (Figure 3, Table 2). We can speculate that this constellation attenuates the increased excitability of the axotomized motoneurons and, thus, allows better recovery of the motoneuron and its better regeneration [18, 52]. It is also thinkable that BoNT-related modulations of reflexes and/or pain-related transmission may have also positive functional consequences [5356]. A major unresolved issue in this study is why BoNT had synaptotrophic effects on some types of synapses. The unexpected observation, which is unrelated to the main goal and achievement of this work, has to be explained by future experiments.

Table 2: Summary of effects of drug application on VGLUT1+, VGLUT2+, ChAT+, and VGAT+ synaptic terminals and Iba1+ cells one week and two months after injury. Arrows indicate increase (↑), decrease (↓), or no difference (=) compared to BSA treatment.

Similar to BoNT, BDNF is retrogradely transported from the periphery to the cell body of motoneurons and then transcytosed to afferent presynaptic terminals [57]. Exogenous BDNF has already shown synaptotrophic properties in injury models [18, 19, 58], and exogenous BDNF can improve axonal regeneration [59, 60]. We indeed found a BDNF effect at one week after injury—prevention of injury-induced ChAT+ terminal loss (Figure 3, Table 2), but no functional effects were seen (Figures 6(a) and 7). This may be related to lack of a prolonged protective effect on ChAT+ terminals as observed two months after BoNT application (Figure 5, Table 2).

4. Conclusions

The results of this study provide further support to the notion that insufficient recovery of synaptic inputs to motoneurons, in particular, perisomatic cholinergic terminals, may be an essential factor limiting recovery after peripheral nerve injury and regeneration. In addition, it appears encouraging that single intraoperative application of drugs to the severed nerve can be a useful way to modify neuronal responses to axotomy and, thus, modulate regeneration and eventually improve functional outcome of nerve injury. The list of candidates for such applications may be long, ranging from other neurotrophins or combinations of neurotrophins (e.g., BDNF and neurotrophin-3 [18], NGF [61]) or growth factors (e.g., vascular endothelial growth factor (VEGF) [62]) to small bioactive molecules [63].

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that there is no conflict of interest regarding the publication of this paper.

Acknowledgments

The authors are grateful to Frau Heike Thieme for excellent technical assistance.

References

  1. K. Blinzinger and G. Kreutzberg, “Displacement of synaptic terminals from regenerating motoneurons by microglial cells,” Zeitschrift für Zellforschung und Mikroskopische Anatomie, vol. 85, no. 2, pp. 145–157, 1968. View at Publisher · View at Google Scholar · View at Scopus
  2. R. C. Borke, R. S. Bridwell, and M. E. Nau, “The progression of deafferentation as a retrograde reaction to hypoglossal nerve injury,” Journal of Neurocytology, vol. 24, no. 10, pp. 763–774, 1995. View at Publisher · View at Google Scholar · View at Scopus
  3. D. H. Chen, “Qualitative and quantitative study of synaptic displacement in chromatolyzed spinal motoneurons of the cat,” The Journal of Comparative Neurology, vol. 177, no. 4, pp. 635–663, 1978. View at Publisher · View at Google Scholar · View at Scopus
  4. J. M. Kerns and E. J. Hinsman, “Neuroglial response to sciatic neurectomy. I. Light microscopy and autoradiography,” Journal of Comparative Neurology, vol. 151, no. 3, pp. 237–253, 1973. View at Publisher · View at Google Scholar · View at Scopus
  5. G. W. Kreutzberg, “Dynamic changes in motoneurons during regeneration,” Restorative Neurology and Neuroscience, vol. 5, no. 1, pp. 59-60, 1993. View at Publisher · View at Google Scholar · View at Scopus
  6. B. E. H. Sumner and F. I. Sutherland, “Quantitative electron microscopy on the injured hypoglossal nucleus in the rat,” Journal of Neurocytology, vol. 2, no. 3, pp. 315–328, 1973. View at Publisher · View at Google Scholar · View at Scopus
  7. B. E. H. Sumner, “A quantitative analysis of boutons with different types of synapse in normal and injured hypoglossal nuclei,” Experimental Neurology, vol. 49, no. 2, pp. 406–417, 1975. View at Publisher · View at Google Scholar · View at Scopus
  8. T. Brännström and J. O. Kellerth, “Recovery of synapses in axotomized adult cat spinal motoneurons after reinnervation into muscle,” Experimental Brain Research, vol. 125, no. 1, pp. 19–27, 1999. View at Google Scholar
  9. A. Raslan, P. Ernst, M. Werle et al., “Reduced cholinergic and glutamatergic synaptic input to regenerated motoneurons after facial nerve repair in rats: potential implications for recovery of motor function,” Brain Structure & Function, vol. 219, no. 3, pp. 891–909, 2014. View at Publisher · View at Google Scholar · View at Scopus
  10. T. M. Rotterman, P. Nardelli, T. C. Cope, and F. J. Alvarez, “Normal distribution of VGLUT1 synapses on spinal motoneuron dendrites and their reorganization after nerve injury,” The Journal of Neuroscience, vol. 34, no. 10, pp. 3475–3492, 2014. View at Publisher · View at Google Scholar · View at Scopus
  11. A. J. Schultz, T. M. Rotterman, A. Dwarakanath, and F. J. Alvarez, “VGLUT1 synapses and P-boutons on regenerating motoneurons after nerve crush,” The Journal of Comparative Neurology, vol. 525, no. 13, pp. 2876–2889, 2017. View at Publisher · View at Google Scholar · View at Scopus
  12. G. Hundeshagen, K. Szameit, H. Thieme et al., “Deficient functional recovery after facial nerve crush in rats is associated with restricted rearrangements of synaptic terminals in the facial nucleus,” Neuroscience, vol. 248, pp. 307–318, 2013. View at Publisher · View at Google Scholar · View at Scopus
  13. F. Antonucci, C. Rossi, L. Gianfranceschi, O. Rossetto, and M. Caleo, “Long-distance retrograde effects of botulinum neurotoxin A,” The Journal of Neuroscience, vol. 28, no. 14, pp. 3689–3696, 2008. View at Publisher · View at Google Scholar · View at Scopus
  14. I. Matak, P. Riederer, and Z. Lackovic, “Botulinum toxin’s axonal transport from periphery to the spinal cord,” Neurochemistry International, vol. 61, no. 2, pp. 236–239, 2012. View at Publisher · View at Google Scholar · View at Scopus
  15. L. Restani, F. Giribaldi, M. Manich et al., “Botulinum neurotoxins A and E undergo retrograde axonal transport in primary motor neurons,” PLoS Pathogens, vol. 8, no. 12, article e1003087, 2012. View at Publisher · View at Google Scholar · View at Scopus
  16. L. Restani, E. Novelli, D. Bottari et al., “Botulinum neurotoxin A impairs neurotransmission following retrograde transynaptic transport,” Traffic, vol. 13, no. 8, pp. 1083–1089, 2012. View at Publisher · View at Google Scholar · View at Scopus
  17. A. M. Pastor, B. Moreno-Lopez, R. R. De La Cruz, and J. M. Delgado-Garcia, “Effects of botulinum neurotoxin type A on abducens motoneurons in the cat: ultrastructural and synaptic alterations,” Neuroscience, vol. 81, no. 2, pp. 457–478, 1997. View at Publisher · View at Google Scholar · View at Scopus
  18. M. A. Davis-López de Carrizosa, C. J. Morado-Diaz, J. J. Tena et al., “Complementary actions of BDNF and neurotrophin-3 on the firing patterns and synaptic composition of motoneurons,” The Journal of Neuroscience, vol. 29, no. 2, pp. 575–587, 2009. View at Publisher · View at Google Scholar · View at Scopus
  19. L. N. Novikov, L. N. Novikova, P. Holmberg, and J.-O. Kellerth, “Exogenous brain-derived neurotrophic factor regulates the synaptic composition of axonally lesioned and normal adult rat motoneurons,” Neuroscience, vol. 100, no. 1, pp. 171–181, 2000. View at Publisher · View at Google Scholar · View at Scopus
  20. A. Al-Majed, T. Brushart, and T. Gordon, “Electrical stimulation accelerates and increases expression of BDNF and trkB mRNA in regenerating rat femoral motoneurons,” The European Journal of Neuroscience, vol. 12, no. 12, pp. 4381–4390, 2000. View at Google Scholar
  21. A. A. al-Majed, C. M. Neumann, T. M. Brushart, and T. Gordon, “Brief electrical stimulation promotes the speed and accuracy of motor axonal regeneration,” The Journal of Neuroscience, vol. 20, no. 7, pp. 2602–2608, 2000. View at Publisher · View at Google Scholar
  22. A. Irintchev, “Potentials and limitations of peripheral nerve injury models in rodents with particular reference to the femoral nerve,” Annals of Anatomy, vol. 193, no. 4, pp. 276–285, 2011. View at Publisher · View at Google Scholar · View at Scopus
  23. M. Kruspe, H. Thieme, O. Guntinas-Lichius, and A. Irintchev, “Motoneuron regeneration accuracy and recovery of gait after femoral nerve injuries in rats,” Neuroscience, vol. 280, pp. 73–87, 2014. View at Publisher · View at Google Scholar · View at Scopus
  24. T. Brushart, “Preferential reinnervation of motor nerves by regenerating motor axons,” The Journal of Neuroscience, vol. 8, no. 3, pp. 1026–1031, 1988. View at Publisher · View at Google Scholar
  25. O. Simova, A. Irintchev, A. Mehanna et al., “Carbohydrate mimics promote functional recovery after peripheral nerve repair,” Annals of Neurology, vol. 60, no. 4, pp. 430–437, 2006. View at Publisher · View at Google Scholar · View at Scopus
  26. A. Irintchev, A. Rollenhagen, E. Troncoso, J. Z. Kiss, and M. Schachner, “Structural and functional aberrations in the cerebral cortex of tenascin-C deficient mice,” Cerebral Cortex, vol. 15, no. 7, pp. 950–962, 2005. View at Publisher · View at Google Scholar · View at Scopus
  27. A. Irintchev, M. Zeschnigk, A. Starzinski-Powitz, and A. Wernig, “Expression pattern of M-cadherin in normal, denervated, and regenerating mouse muscles,” Developmental Dynamics, vol. 199, no. 4, pp. 326–337, 1994. View at Publisher · View at Google Scholar · View at Scopus
  28. M. V. Sofroniew and U. Schrell, “Long-term storage and regular repeated use of diluted antisera in glass staining jars for increased sensitivity, reproducibility, and convenience of single- and two-color light microscopic immunocytochemistry,” The Journal of Histochemistry and Cytochemistry, vol. 30, no. 6, pp. 504–511, 1982. View at Publisher · View at Google Scholar · View at Scopus
  29. S. Cullheim and S. Thams, “The microglial networks of the brain and their role in neuronal network plasticity after lesion,” Brain Research Reviews, vol. 55, no. 1, pp. 89–96, 2007. View at Publisher · View at Google Scholar · View at Scopus
  30. D. Gomez-Nicola and V. H. Perry, “Microglial dynamics and role in the healthy and diseased brain: a paradigm of functional plasticity,” The Neuroscientist, vol. 21, no. 2, pp. 169–184, 2015. View at Publisher · View at Google Scholar · View at Scopus
  31. G. W. Kreutzberg, “Microglia: a sensor for pathological events in the CNS,” Trends in Neurosciences, vol. 19, no. 8, pp. 312–318, 1996. View at Publisher · View at Google Scholar · View at Scopus
  32. C. Rheaume, B. B. Cai, J. Wang et al., “A highly specific monoclonal antibody for botulinum neurotoxin type A-cleaved SNAP25,” Toxins, vol. 7, no. 7, pp. 2354–2370, 2015. View at Publisher · View at Google Scholar · View at Scopus
  33. M. Richner, M. Ulrichsen, S. L. Elmegaard, R. Dieu, L. T. Pallesen, and C. B. Vaegter, “Peripheral nerve injury modulates neurotrophin signaling in the peripheral and central nervous system,” Molecular Neurobiology, vol. 50, no. 3, pp. 945–970, 2014. View at Publisher · View at Google Scholar · View at Scopus
  34. C. Verderio, C. Grumelli, L. Raiteri et al., “Traffic of botulinum toxins A and E in excitatory and inhibitory neurons,” Traffic, vol. 8, no. 2, pp. 142–153, 2007. View at Publisher · View at Google Scholar · View at Scopus
  35. C. Verderio, D. Pozzi, E. Pravettoni et al., “SNAP-25 modulation of calcium dynamics underlies differences in GABAergic and glutamatergic responsiveness to depolarization,” Neuron, vol. 41, no. 4, pp. 599–610, 2004. View at Publisher · View at Google Scholar · View at Scopus
  36. P. H. Beske, S. M. Scheeler, M. Adler, and P. M. McNutt, “Accelerated intoxication of GABAergic synapses by botulinum neurotoxin A disinhibits stem cell-derived neuron networks prior to network silencing,” Frontiers in Cellular Neuroscience, vol. 9, p. 159, 2015. View at Publisher · View at Google Scholar · View at Scopus
  37. S. Cobianchi, J. Jaramillo, S. Luvisetto, F. Pavone, and X. Navarro, “Botulinum neurotoxin A promotes functional recovery after peripheral nerve injury by increasing regeneration of myelinated fibers,” Neuroscience, vol. 359, pp. 82–91, 2017. View at Publisher · View at Google Scholar · View at Scopus
  38. I. Apostolova, A. Irintchev, and M. Schachner, “Tenascin-R restricts posttraumatic remodeling of motoneuron innervation and functional recovery after spinal cord injury in adult mice,” The Journal of Neuroscience, vol. 26, no. 30, pp. 7849–7859, 2006. View at Publisher · View at Google Scholar · View at Scopus
  39. I. Jakovcevski, J. F. Wu, N. Karl et al., “Glial scar expression of CHL1, the close homolog of the adhesion molecule L1, limits recovery after spinal cord injury,” The Journal of Neuroscience, vol. 27, no. 27, pp. 7222–7233, 2007. View at Publisher · View at Google Scholar · View at Scopus
  40. H. J. Lee, I. Jakovcevski, N. Radonjic, L. Hoelters, M. Schachner, and A. Irintchev, “Better functional outcome of compression spinal cord injury in mice is associated with enhanced H-reflex responses,” Experimental Neurology, vol. 216, no. 2, pp. 365–374, 2009. View at Publisher · View at Google Scholar · View at Scopus
  41. Z. Csaba, E. Krejci, and V. Bernard, “Postsynaptic muscarinic m2 receptors at cholinergic and glutamatergic synapses of mouse brainstem motoneurons,” The Journal of Comparative Neurology, vol. 521, no. 9, pp. 2008–2024, 2013. View at Publisher · View at Google Scholar · View at Scopus
  42. M. S. Davidoff and A. P. Irintchev, “Acetylcholinesterase activity and type C synapses in the hypoglossal, facial and spinal-cord motor nuclei of rats. An electron-microscope study,” Histochemistry, vol. 84, no. 4-6, pp. 515–524, 1986. View at Publisher · View at Google Scholar · View at Scopus
  43. J. Hellström, U. Arvidsson, R. Elde, S. Cullheim, and B. Meister, “Differential expression of nerve terminal protein isoforms in VAChT-containing varicosities of the spinal cord ventral horn,” The Journal of Comparative Neurology, vol. 411, no. 4, pp. 578–590, 1999. View at Publisher · View at Google Scholar
  44. J. Hellström, A. L. R. Oliveira, B. Meister, and S. Cullheim, “Large cholinergic nerve terminals on subsets of motoneurons and their relation to muscarinic receptor type 2,” The Journal of Comparative Neurology, vol. 460, no. 4, pp. 476–486, 2003. View at Publisher · View at Google Scholar · View at Scopus
  45. J. I. Nagy, T. Yamamoto, and L. M. Jordan, “Evidence for the cholinergic nature of C-terminals associated with subsurface cisterns in alpha-motoneurons of rat,” Synapse, vol. 15, no. 1, pp. 17–32, 1993. View at Publisher · View at Google Scholar · View at Scopus
  46. J. M. Wilson, J. Rempel, and R. M. Brownstone, “Postnatal development of cholinergic synapses on mouse spinal motoneurons,” The Journal of Comparative Neurology, vol. 474, no. 1, pp. 13–23, 2004. View at Publisher · View at Google Scholar · View at Scopus
  47. G. B. Miles, R. Hartley, A. J. Todd, and R. M. Brownstone, “Spinal cholinergic interneurons regulate the excitability of motoneurons during locomotion,” Proceedings of the National Academy of Sciences of the United States of America, vol. 104, no. 7, pp. 2448–2453, 2007. View at Publisher · View at Google Scholar · View at Scopus
  48. L. Zagoraiou, T. Akay, J. F. Martin, R. M. Brownstone, T. M. Jessell, and G. B. Miles, “A cluster of cholinergic premotor interneurons modulates mouse locomotor activity,” Neuron, vol. 64, no. 5, pp. 645–662, 2009. View at Publisher · View at Google Scholar · View at Scopus
  49. D. B. Hoover, R. H. Baisden, and J. V. Lewis, “Axotomy-induced loss of m2 muscarinic receptor mRNA in the rat facial motor nucleus precedes a decrease in concentration of muscarinic receptors,” The Histochemical Journal, vol. 28, no. 11, pp. 771–778, 1996. View at Publisher · View at Google Scholar
  50. D. B. Hoover and J. C. Hancock, “Effect of facial nerve transection on acetylcholinesterase, choline acetyltransferase and [3H] quinuclidinyl benzilate binding in rat facial nuclei,” Neuroscience, vol. 15, no. 2, pp. 481–487, 1985. View at Publisher · View at Google Scholar · View at Scopus
  51. E. C. Witts, L. Zagoraiou, and G. B. Miles, “Anatomy and function of cholinergic C bouton inputs to motor neurons,” Journal of Anatomy, vol. 224, no. 1, pp. 52–60, 2014. View at Publisher · View at Google Scholar · View at Scopus
  52. D. Gonzalez-Forero and B. Moreno-Lopez, “Retrograde response in axotomized motoneurons: nitric oxide as a key player in triggering reversion toward a dedifferentiated phenotype,” Neuroscience, vol. 283, pp. 138–165, 2014. View at Publisher · View at Google Scholar · View at Scopus
  53. M. Kerzoncuf, L. Bensoussan, A. Delarque, J. Durand, J. M. Viton, and C. Rossi-Durand, “Plastic changes in spinal synaptic transmission following botulinum toxin A in patients with post-stroke spasticity,” Journal of Rehabilitation Medicine, vol. 47, no. 10, pp. 910–916, 2015. View at Publisher · View at Google Scholar · View at Scopus
  54. R. Kumar, H. P. Dhaliwal, R. V. Kukreja, and B. R. Singh, “The botulinum toxin as a therapeutic agent: molecular structure and mechanism of action in motor and sensory systems,” Seminars in Neurology, vol. 36, no. 01, pp. 010–019, 2016. View at Publisher · View at Google Scholar · View at Scopus
  55. R. Mazzocchio and M. Caleo, “More than at the neuromuscular synapse: actions of botulinum neurotoxin A in the central nervous system,” The Neuroscientist, vol. 21, no. 1, pp. 44–61, 2015. View at Publisher · View at Google Scholar · View at Scopus
  56. G. Sandrini, R. De Icco, C. Tassorelli, N. Smania, and S. Tamburin, “Botulinum neurotoxin type A for the treatment of pain: not just in migraine and trigeminal neuralgia,” The Journal of Headache and Pain, vol. 18, no. 1, p. 38, 2017. View at Publisher · View at Google Scholar · View at Scopus
  57. H. B. Rind, R. Butowt, and C. S. von Bartheld, “Synaptic targeting of retrogradely transported trophic factors in motoneurons: comparison of glial cell line-derived neurotrophic factor, brain-derived neurotrophic factor, and cardiotrophin-1 with tetanus toxin,” The Journal of Neuroscience, vol. 25, no. 3, pp. 539–549, 2005. View at Publisher · View at Google Scholar · View at Scopus
  58. J. Krakowiak, C. Liu, C. Papudesu, P. J. Ward, J. C. Wilhelm, and A. W. English, “Neuronal BDNF signaling is necessary for the effects of treadmill exercise on synaptic stripping of axotomized motoneurons,” Neural Plasticity, vol. 2015, Article ID 392591, 11 pages, 2015. View at Publisher · View at Google Scholar · View at Scopus
  59. M. Gao, P. Lu, D. Lynam et al., “BDNF gene delivery within and beyond templated agarose multi-channel guidance scaffolds enhances peripheral nerve regeneration,” Journal of Neural Engineering, vol. 13, no. 6, article 066011, 2016. View at Publisher · View at Google Scholar · View at Scopus
  60. D. Santos, F. Gonzalez-Perez, X. Navarro, and J. Del Valle, “Dose-dependent differential effect of neurotrophic factors on in vitro and in vivo regeneration of motor and sensory neurons,” Neural Plasticity, vol. 2016, Article ID 4969523, 13 pages, 2016. View at Publisher · View at Google Scholar · View at Scopus
  61. M. A. Davis-Lopez de Carrizosa, C. J. Morado-Diaz, S. Morcuende, R. R. de la Cruz, and A. M. Pastor, “Nerve growth factor regulates the firing patterns and synaptic composition of motoneurons,” The Journal of Neuroscience, vol. 30, no. 24, pp. 8308–8319, 2010. View at Publisher · View at Google Scholar · View at Scopus
  62. P. M. Calvo, R. R. de la Cruz, and A. M. Pastor, “Synaptic loss and firing alterations in axotomized motoneurons are restored by vascular endothelial growth factor (VEGF) and VEGF-B,” Experimental Neurology, vol. 304, pp. 67–81, 2018. View at Publisher · View at Google Scholar · View at Scopus
  63. K. W.-H. Lo, T. Jiang, K. A. Gagnon, C. Nelson, and C. T. Laurencin, “Small-molecule based musculoskeletal regenerative engineering,” Trends in Biotechnology, vol. 32, no. 2, pp. 74–81, 2014. View at Publisher · View at Google Scholar · View at Scopus
  64. Y. Imai, I. Ibata, D. Ito, K. Ohsawa, and S. Kohsaka, “A novel gene iba1 in the major histocompatibility complex class III region encoding an EF hand protein expressed in a monocytic lineage,” Biochemical and Biophysical Research Communications, vol. 224, no. 3, pp. 855–862, 1996. View at Publisher · View at Google Scholar · View at Scopus
  65. D. Ito, Y. Imai, K. Ohsawa, K. Nakajima, Y. Fukuuchi, and S. Kohsaka, “Microglia-specific localisation of a novel calcium binding protein, Iba1,” Molecular Brain Research, vol. 57, no. 1, pp. 1–9, 1998. View at Publisher · View at Google Scholar · View at Scopus
  66. F. A. Chaudhry, R. J. Reimer, E. E. Bellocchio et al., “The vesicular GABA transporter, VGAT, localizes to synaptic vesicles in sets of glycinergic as well as GABAergic neurons,” The Journal of Neuroscience, vol. 18, no. 23, pp. 9733–9750, 1998. View at Publisher · View at Google Scholar
  67. S. L. McIntire, R. J. Reimer, K. Schuske, R. H. Edwards, and E. M. Jorgensen, “Identification and characterization of the vesicular GABA transporter,” Nature, vol. 389, no. 6653, pp. 870–876, 1997. View at Publisher · View at Google Scholar · View at Scopus
  68. S. M. Wojcik, S. Katsurabayashi, I. Guillemin et al., “A shared vesicular carrier allows synaptic corelease of GABA and glycine,” Neuron, vol. 50, no. 4, pp. 575–587, 2006. View at Publisher · View at Google Scholar · View at Scopus
  69. F. J. Alvarez, R. M. Villalba, R. Zerda, and S. P. Schneider, “Vesicular glutamate transporters in the spinal cord, with special reference to sensory primary afferent synapses,” The Journal of Comparative Neurology, vol. 472, no. 3, pp. 257–280, 2004. View at Publisher · View at Google Scholar · View at Scopus
  70. A. L. R. Oliveira, F. Hydling, E. Olsson et al., “Cellular localization of three vesicular glutamate transporter mRNAs and proteins in rat spinal cord and dorsal root ganglia,” Synapse, vol. 50, no. 2, pp. 117–129, 2003. View at Publisher · View at Google Scholar · View at Scopus