Muscle-in-vein conduit is successfully employed for repairing nerve injuries: the vein prevents muscle fiber dispersion, while the muscle prevents the vein collapse and creates a favorable environment for Schwann cell migration and axon regrowth. However, it requires microsurgical skills. In this study we show a simple strategy to improve the performance of a chitosan hollow tube by the introduction of fresh skeletal muscle fibers. The hypothesis is to overcome the technical issue of the muscle-in-vein preparation and to take advantage of fiber muscle properties to create an easy and effective conduit for nerve regeneration. Rat median nerve gaps were repaired with chitosan tubes filled with skeletal muscle fibers (muscle-in-tube graft), hollow chitosan tubes, or autologous nerve grafts. Our results demonstrate that the fresh skeletal muscle inside the conduit is an endogenous source of soluble Neuregulin 1, a key factor for Schwann cell survival and dedifferentiation, absent in the hollow tube during the early phase of regeneration. However, nerve regeneration assessed at late time point was similar to that obtained with the hollow tube. To conclude, the muscle-in-tube graft is surgically easy to perform and we suggest that it might be a promising strategy to repair longer nerve gap or for secondary nerve repair, situations in which Schwann cell atrophy is a limiting factor for recovery.
Currently, the gold standard technique used to repair large peripheral nerve defects is the autologous nerve graft. However, this procedure has some well-known disadvantages: the need of an additional surgery to harvest the donor nerve, sensory deficits at the donor site, the possibility of neuroma-in-continuity formation with consequent neuropathic pain, and the limited availability of donor nerves in terms of number and diameter [
The use of conduits made by nonnervous materials (tubulization technique) to bridge nerve gap has been widely investigated and has shown promising results both experimentally and clinically [
Several materials, both of biological and of synthetic origin, have been used to build tubular conduits [
Chitosan is a natural polysaccharide that has been demonstrated to be a good biomaterial with a wide range of biomedical and tissue engineering applications [
Nevertheless, when more complex lesions occur, such as long gaps, hollow tubes not always lead to good results and conduit enrichment might create a favorable environment for nerve regrowth.
In this study we combined the use of a successful conduit (the chitosan tube) with the promising and simple intraluminal structure (fresh longitudinal skeletal muscle fibers) to evaluate peripheral nerve regeneration after rat median nerve reconstruction. The hypothesis is to (i) overcome the technical issue of the muscle-in-vein preparation by the use of a successful conduit, and to (ii) take advantage of fiber muscle properties to create a surgically easy and effective conduit for nerve regeneration.
Rat median nerve gaps were repaired with (i) chitosan tubes filled with skeletal muscle fibers (muscle-in-tube graft), (ii) hollow chitosan tubes, or (iii) autologous nerve grafts (surgical gold standard). Samples harvested at early (1, 7, 14, and 28 days after nerve repair) time points, together with
Longitudinal pieces of rat
At the same time points, degenerating muscles were collected and frozen for RNA extraction and quantitative real time PCR (qRT-PCR) analyses. As control for RNA analysis, fresh muscle fibers were also collected. Experiments were carried out in biological triplicate.
Total RNA was isolated according to the manufacturer’s instructions using TRIzol Reagent (Invitrogen). Reverse transcription and quantitative real time PCR were performed and analyzed as previously described [
As calibrator for relative quantification, the average of uninjured nerve ΔCt was used for nerve regeneration
As housekeeping gene to normalize data, ANKRD27 (Ankyrin repeat domain 27) was used for
For
During surgical procedures, animals were placed under general anaesthesia induced by IM injection of Tiletamine + Zolazepam (Zoletil, 3 mg/kg) and were positioned in supine position. Using an incision from the nipple to the elbow, the median nerve was isolated to establish a defect in the middle of the exposed part, immediately followed by nerve repair according to the experimental group:
(i)
(ii)
Preparation of the muscle-in-tube graft. A longitudinal piece of the
(iii)
Animals were sacrificed by anaesthetic overdose at different times points: 1 day (only autograft and muscle-in-tube graft) and 7, 14, and 28 days for early time point analysis; 1 day hollow chitosan tube graft was not analyzed, because it is colonized only by fluid and it was not possible to withdraw it. For late time point analysis (only hollow tube and muscle-in-tube), animals were sacrificed 12 weeks after the surgery. Data about autograft 12 weeks after surgery are already present in literature [
Healthy median nerve segments and healthy
The grasping test was performed to estimate the functional recovery after nerve reconstruction. The analysis was carried out before animal sacrifice (12 weeks after nerve repair) following the same procedure previously described [
Samples corresponding to the grafted region (the autologous nerve and the filler of the chitosan tubes) harvested 7, 14, and 28 days after repair and the distal portion of regenerated median nerve, harvested 12 weeks after repair, were processed for resin embedding as previously described and semithin and ultrathin sections were cut for light and electron microscopy analysis [
The quantification of myelinated nerve fibers was performed on electron microscopy micrographs. Briefly, on one randomly selected ultrathin section, 15-20 fields were chosen using a systematic random sampling protocol, as earlier described [
For statistical analysis IBM SPSS Statistics 22.0 software was used. Data were expressed as mean ± standard error of the mean (SEM). Data sets containing two groups were analyzed through two-tailed Student’s
To evaluate the expression level of the different Neuregulin 1 (NRG1) isoforms in degenerating muscles, longitudinal pieces of the
Degenerating muscle expresses NRG1. (a) qRT-PCR showing the relative quantification (-ΔΔCt = log2fold change) of the different NRG1 isoforms in
Conditioned medium was collected and analyzed by ELISA to quantify soluble NRG1 release. ELISA was performed on 50
To verify the effectiveness of the muscle-in-tube graft to repair rat median nerve gaps, we compared this method with hollow chitosan tubes and autologous nerve graft.
To observe the qualitative morphology of the nerve in the early stages of regeneration, high resolution light micrographs of toluidine-blue stained semithin sections were taken at short-term time points postsurgery (7, 14 and 28 days). Images were gathered inside the different grafts, both proximally and distally, to follow nerve regeneration alongside the conduits (Figure
Representative high resolution light images of toluidine-blue stained semithin transverse sections of regenerated median nerve repaired with autograft (a), hollow chitosan tube (b), or muscle-in-tube graft (c). Images were taken inside the grafts, both proximally (left columns, about 1,5 mm from the proximal suture point) and distally (right columns, about 1,5 mm from the distal suture point), at different time points (7, 14, 28 days after repair). Bar: 40
Nerves repaired with the autograft (positive control) showed an early regeneration, as expected (Figure
Regeneration inside the hollow chitosan tube is slower when compared to the autograft (Figure
In the muscle-in-tube graft, longitudinal muscle fibers were inserted inside the chitosan conduit to provide a physical and trophic scaffold for axon and cell growth (Figure
At different time points after nerve reconstruction (1, 7, 14, and 28 days), three Schwann cell markers (S100, p75, GFAP), ErbB receptors, and different NRG1 isoforms were examined by qRT-PCR in the grafts of the three experimental groups (Figures
Quantitative analysis of SC marker expression. Relative quantification (-ΔΔCt = log2fold change) of S100 (a), GFAP (b), and p75 (c) was evaluated by qRT-PCR. Both healthy nerves and healthy muscle fibers were analyzed and shown in the figures (green and red lines, respectively). ANKRD27 was used as housekeeping gene to normalize data. Values in the graphics are expressed as mean ± SEM. One-way ANOVA was carried out; asterisks (
Quantitative analysis of the NRG1/ErbB system expression. qRT-PCR showing the relative quantification (-ΔΔCt = log2fold change) of ErbB receptors (a-d) and the different NRG1 isoforms (e-j). Both healthy nerves and healthy muscle fibers were analyzed and shown in the figures (green and red lines, respectively). ANKRD27 was used as housekeeping gene to normalize data. Values in the graphics are expressed as mean ± SEM. One-way ANOVA was carried out; asterisks (
The expression of Schwann cell markers S100 and GFAP was significantly lower in hollow chitosan tube and muscle-in-tube compared to both the autograft and the healthy nerves. After 28 days muscle-in-tube graft still showed a lower expression of these two markers (Figures
p75 expression was significantly higher in the autograft samples at the different time points, while after 28 days no differences among the three repair experimental models were detectable (Figure
Then, ErbB receptors and NRG1 isoforms were analyzed. ErbB1 expression level in the muscle-in-tube samples was downregulated after injury and repair and was lower than in the hollow tube and in the autograft samples, where the expression was similar (Figure
ErbB2 expression in the muscle-in-tube samples was strongly and stably upregulated at 7 and 14 days after injury relatively both to the healthy nerve and muscle and to the autograft and the hollow chitosan groups. At day 14 the autograft showed an expression level higher than the empty chitosan group (Figure
ErbB3 expression in the chitosan tube and in the muscle-in-tube after injury was lower than in the autograft samples. 28 days after injury ErbB3 expression in the hollow chitosan samples was similar to the autograft and higher than the muscle-in-tube samples (Figure
ErbB4 expression is significantly lower in the muscle-in-tube samples—relatively to both healthy muscle and nerve—already 1 day after injury and remained really low until 28 days. In the hollow chitosan samples ErbB4 expression was similar to the muscle-in-tube samples, while in the autograft ErbB4 expression decreased after injury but was higher than the muscle-in-tube graft and the hollow chitosan samples. 28 days after injury only hollow chitosan samples showed an expression level similar to the autograft (Figure
NRG1 expression analysis is really complex, because each primer pair amplifies a single isoform, but soluble and transmembrane isoforms can be type
Autograft showed a strong early upregulation (1 day) of the different NRG1 isoforms (except for NRG1c) followed by a return to values similar to control nerves. On the contrary, NRG1c was downregulated from the 14th day (Figures
In the first days after implantation, the hollow chitosan tube is filled by cells which do not express NRG1 or that express it at low level; indeed, at day 7 mRNA expression of most NRG1 isoforms was similar or lower to control nerve values. At day 14 NRG1
Finally, muscle-in-tube graft showed an expression pattern very similar to that of the autograft: all soluble NRG1 isoforms showed a strong peak 1 day after implantation (except for NRG1c) and then the upregulation of most of them was still high at 7 and 14 days, returning to control values after 28 days (except for NRG1
Overall, the upregulation peak of NRG1 isoforms in the hollow chitosan tube started only 14 days after injury and is lower than muscle-in-tube graft. Intriguingly, the upregulation at day 1 of soluble NRG1 (
Two-way ANOVA was carried out for all analyzed genes to evaluate the effect on their expression of the different repair experimental models and of the time after repair (see Figure
To determine whether the presence of fresh muscle fibers inside the chitosan tube influences nerve regeneration, we performed a functional test (grasping test) and morphometrical analysis on the nerve segment distal to the graft 12 weeks after nerve reconstruction (Figure
Functional, morphological, and morphometrical analyses of long-term regenerated nerves after hollow tube or muscle-in-tube (MIT) graft primary repair. (a) Histograms showing the result of the grasping test, performed 12 weeks after nerve repair. (b-f) Histograms of morphometrical analysis of the regenerated myelinated fibers. Analyses were performed 3 mm distal to the graft: (b) cross-sectional area of the whole nerve section, (c) myelinated fiber density, (d) total number of myelinated fibers, (e) size parameters (axon and fiber diameter, myelin thickness), and (f) g
The functional test performed at week 12 after repair did not show statistical differences between the hollow chitosan tube and the muscle-in-tube groups (Figure
Results of the quantitative analysis performed on regenerated nerve fibers show no significant differences in all the analyzed parameters between the two experimental groups (Figures
The use of hollow conduits to repair nerve defects is a valid alternative technique to autograft because of their well-demonstrated advantages, and in the last years several kinds of conduits have been proposed for their use in clinic.
However, the efficiency of hollow conduits is still insufficient, especially for large nerve gap, probably due to an inadequate formation of the extracellular matrix, thus limiting cell migration and axonal regrowth [
In 1993, Brunelli et al. [
To overcome this technical issue, the muscle tissue can be introduced inside a tubular conduit. However, so far, the placement of denatured skeletal muscle tissue [
In this study we used a conduit made of chitosan (Reaxon Nerve Guide), that has been shown to promote successful nerve regeneration [
We used the fresh muscle tissue because it has been demonstrated that muscle cells produce and release factors that contribute to the survival of motoneurons
Here we focused our attention on Neuregulin 1 (NRG1), because it is known to be one of the most important factors regulating Schwann cell activity (survival, proliferation, dedifferentiation, and migration) and, therefore, it is a key factor for peripheral nerve regeneration.
Taken together, these results show that the fresh skeletal muscle can be an endogenous source of the gliotrophic factor NRG1. For this reason, we moved to
Our
Our results indicate that muscle-in-tube graft promotes nerve regeneration. With respect to functional recovery and quantitative morphometry, no significant differences were observed between the two experimental groups at 12 weeks after surgery, suggesting that both conduits are effective for repairing peripheral nerve defects in this experimental model (short gap primary repair). The comparison of morphometrical data obtained in this study with our previous data obtained using autograft [
The mRNA analysis of these samples is really complex. Indeed, muscle-in-tube samples at early time points (1 day and 7 days after injury and repair) are in fact mostly muscles and for this reason we have to compare them also with the muscle, not only with the nerve. Then, step by step, the muscle mRNA percentage decreases, while the nerve mRNA percentage increases. Consequently, although healthy nerves were used as a common calibrator for the three experimental groups, both healthy nerves and healthy muscle fibers were analyzed and shown in the figures. Also hollow chitosan tubes represent a complex model, although in a different way: they need to be filled by migrating cell populations and for this reason the possible gene downregulation in the first time point (7 days) is only apparent, because at time zero they are just empty tubes. Therefore, 7 days should be considered the expression starting point for these samples.
mRNA analysis, performed inside the conduits at shorter time points, demonstrates that Schwann cell markers S100, p75, and GFAP are less expressed both in the hollow chitosan tube and in the muscle-in-tube graft when compared with the autograft. This is an expected result because the chitosan tube (empty or filled with muscle fibers) needs to be colonized by Schwann cells, whereas the autograft already contains Schwann cells that start to dedifferentiate immediately after repair. The slight (and insignificant) different expression between hollow chitosan tube and muscle-in-tube graft might be explained by the fact that in the muscle-in-tube graft the nerve RNA is “diluted” by the presence of muscle RNA.
Interestingly, hollow chitosan tube and muscle-in-tube graft differed greatly in terms of NRG1 expression: different isoforms of soluble NRG1 are highly expressed in the muscle-in-tube early after nerve repair, whereas no NRG1 expression is seen in the hollow chitosan tube. The NRG1 expression pattern observed in muscle-in-tube samples is very similar to that observed in autograft samples, in which NRG1 is upregulated by Schwann cells that colonize the graft. A similar expression pattern has also been described in other experimental models, such as crush injury and end-to-end repair [
Our results show that the muscle-in-tube graft promotes nerve regeneration as efficiently as the hollow chitosan tube and that the fresh skeletal muscle inserted inside the chitosan conduit may be an endogenous source of soluble NRG1, a source that is absent in the hollow tube. We recently showed that, after prolonged degeneration of the median nerve distal stump, Schwann cells undergo atrophy and downregulate the expression of soluble NRG1. Even after the cross-suture with the freshly axotomized ulnar nerve proximal stump, NRG1 remains at very low expression level and nerve regeneration results are impaired [
The raw data will be provided upon request.
Giulia Ronchi and Benedetta Elena Fornasari share co-first authorship. Stefania Raimondo and Giovanna Gambarotta share senior authorship.
The authors report no conflicts of interest.
Medical grade chitosan for chitosan nerve guides manufacturing was supplied by Altakitin SA (Lisbon, Portugal). Chitosan was supplied by Medovent GmbH (Mainz, Germany). This study was supported by the European Community’s Seventh Framework Programme (FP7-HEALTH-2011) (Grant no. 278612; BIOHYBRID) and by Compagnia di San Paolo (InTheCure Project).