Effect of surgically guided axonal regrowth into a 3-way-conduit (isogeneic trifurcated aorta) on functional recovery after facial-nerve reconstruction: Experimental study in rats

Abstract

Background:

The “post-paralytic syndrome” after facial nerve reconstruction has been attributed to (i) malfunctioning axonal guidance at the fascicular (branches) level, (ii) collateral branching of the transected axons at the lesion site, and (iii) intensive intramuscular terminal sprouting of regenerating axons which causes poly-innervation of the neuromuscular junctions (NMJ).

Objective:

The first two reasons were approached by an innovative technique which should provide the re-growing axons optimal conditions to elongate and selectively re-innervate their original muscle groups.

Methods:

The transected facial nerve trunk was inserted into a 3-way-conduit (from isogeneic rat abdominal aorta) which should “guide” the re-growing facial axons to the three main branches of the facial nerve (zygomatic, buccal and marginal mandibular). The effect of this method was tested also on hypoglossal axons after hypoglossal-facial anastomosis (HFA). Coaptational (classic) FFA (facial-facial anastomosis) and HFA served as controls.

Results:

When compared to their coaptation (classic) alternatives, both types of 3-way-conduit operations (FFA and HFA) promoted a trend for reduction in the collateral axonal branching (the proportion of double- or triple-labelled perikarya after retrograde tracing was slightly reduced). In contrast, poly-innervation of NMJ in the levator labii superioris muscle was increased and vibrissal (whisking) function was worsened.

Conclusions:

The use of 3-way-conduit provides no advantages to classic coaptation. Should the latter be impossible (too large interstump defects requiring too long interpositional nerve grafts), this type of reconstruction may be applied. (230 words)

Keywords

Facial nerve hypoglossal nerve facial-facial anastomosis (FFA)hypoglossal-facial anastomosis (HFA)aortic 3-way-conduit

1 Introduction

Facial nerve injury may occur during trauma, parotidectomy, vestibular schwannoma surgery, or petrous bone surgery. Following transection of the nerve, both stumps retract and hinder a reliable primary suture This is why, nerve grafts are usually used to bridge the gap (Lundborg et al., 1994).

Despite ongoing advances in microsurgical techniques and the indigenous regeneration capability of the peripheral nervous system, recovery of functions following repair of transected nerves remains usually suboptimal (Kerrebijn & Freeman, 1998). For example, following facial nerve injury, there occurs an “aberrant” reinnervation of the original targets, caused by inaccurate axonal regrowth within a given nerve branch (poor pathfinding), extensive collateral branching of axons at the lesion site, and vigorous intramuscular (terminal) sprouting with poly-innervation at the neuromuscular junctions (NMJs) (Ozsoy et al., 2011; Skouras et al., 2011; Tomita et al., 2007).

Hypoglossal-facial anastomosis (HFA) is a common technique in cases with an extensive intracranial or intratemporal destruction of the facial nerve. It is also used in cases where the integrity of the facial nerve cannot be restored (Hwang et al., 2006). The proximal stump of a transected normal hypoglossal nerve is sutured to the distal stump of a transected facial nerve resulting in the regrowth of hypoglossal axons into the facial parotid plexus. Earlier studies in a rat model have also shown that the hypoglossal motoneurons regrow into the facial muscles without difficulties. It has been demonstrated, that the transected axons not only manage to approach and reinnervate the whiskerpad muscles, but also that their number increases rapidly with time. Within 2 months after surgery, the vibrissal muscles are almost completely reinnervated (Ozsoy et al., 2011; Guntinas-Lichius et al., 2007).

It might be thus assumed, that regenerating axons should be supported or even guided to make the correct choice when entering distal nerve branches; this, in turn, should ensure better quality of re-innervation of their original muscle targets (Hadlock et al., 1998; Kitahara et al., 1999; Spector et al., 1995). In our earlier work we tried to improve axonal navigation using surgical reconstruction with an isogeneic aortic Y-tube: the proximal portion of the transected facial nerve was inserted into the long arm of the Y-tube and both short arms were inserted into the distal zygomatic and buccal branches. Our results showed that the Y-tube technique reduced collateral axonal branching by 50%. However, the proportion of poly-innervated NMJs remained high (∼30%) and recovery of the vibrissal motor function (whisking) was poor (Ozsoy et al., 2011; Hizay et al., 2012).

Various types of conduits including collagen, silicon and poly-L-lactide have been used studies on cats and rabbits to regenerate the main trunk of the facial nerve. Biological conduits have been also used for bridging peripheral nerve defects. A lot of attention has been given to arterial (Büngner, 1891) and venous (Fansa et al., 1999, 2001, 2002; Hudson et al., 2000; Johnson & Soucacos, 2008) grafts, which seem to facilitate healing in a variety of ways. Their cut ends promote directional axonal regeneration, minimize infiltration by fibrous tissue (Hudson et al., 2000) and provide a conduit in which neurotropic and neurotrophic, factors secreted by the damaged nerve stumps, diffused. In addition to this, regeneration-promoting factors appear to be produced by the aortic tissue itself, which is obviously an advantage over the silicon tubes (Williams, 1987).

As a possible improvement of the Y-tube guiding, in the present study we used a 3-way-conduit (isogeneic abdominal aorta with three well expressed branches: both common iliac arteries and a. sacralis mediana) to ascertain whether it would further reduce collateral branching and improve the pattern of NMJ-reinnervation. Despite the fact that the use of the abdominal aorta with the iliac bifurcation as a “double-choice” Y-tube-conduit is not novel (Weiss, 1944; Weiss & Hoag, 1946; Weiss & Taylor, 1943; Weiss & Taylor, 1944; Politis et al., 1982), this is the first study ever reporting whether an isogeneic three-way-conduit would promote selective regrowth of facial or hypoglossal axons towards the zygomatic, buccal and marginal mandibular branches of the facial nerve.

2 Materials and methods

2.1 Number, strain and sex of animals

Ninety-six adult female inbred rats (Harlan Laboratories Israel, strain “Wistar”, F20 generation) were obtained from the Laboratory Animal Unit of the Akdeniz University. Animals were housed in cages under standard environmental conditions (light between 06.00 and 18.00 h, temperature 22°C, free access to chow and water). All experimental protocols were approved by the Animal Welfare Committee of the Akdeniz University (Protocol Number: Hay. Den. Etik-384/8076) and conducted in accordance with the International Law on the Protection of Animals.

2.2 Animal groups and experimental design

The abdominal aorta and its 3 branches (both common iliac arteries and a. sacralis mediana) were harvested from 24 animals. To avoid tissue rejection 24 inbred rats of F20 generation were used for harvesting the aortal trifurcation, which we used as a 3-way-conduit in groups 4 and 6.

The remaining 72 animals were divided into 6 groups, each of 12 rats. Group 1 comprised intact controls, rats in Group 2 were sham-operated, rats in Group 3 were subjected to transection and subsequent end-to-end nerve suture of the facial nerve trunk (facial-facial-anastomosis, FFA), and rats in Group 4 underwent FFA by means of aortic 3-way-conduit (see below). Rats from Group 5 were subjected to transection of the hypoglossal nerve and subsequent end-to-end suture of its proximal stump to the distal fragment of the transected facial nerve (hypoglossal-facial-anastomosis, HFA) Rats in Group 6 underwent HFA with an aortic 3-way-conduit (see below).

After 4 months, all animals were subjected to video-based motion analysis of vibrissal motor performance. At the conclusion of the behavioral testing, half of the animals (randomly selected) were used to quantify collateral axonal branching at the lesion site using double retrograde labeling and neuronal counts in the facial nucleus. The other half were used to quantify the proportion of mono-, poly-, or non-innervated NMJ using immunocytochemical staining of axons with neuronal class III beta tubulin and histochemical staining of the post-synaptic N-acetylcholine receptor using alpha-bungarotoxin (see below).

2.3 Preparation of the aortic 3-way-conduit

Standardized 3-way-tubes with the necessary dimensions for facial nerve reconstructive surgery are not commercially available. This is why isogeneic animals had to be used as donors for the 3-way-conduits.

Animals were sacrificed by intra-peritoneal injection of urethane (1.5 g/kg) (Sigma, St. Louis, USA). Following midline laparotomy and retraction of the intestines, the abdominal aorta was approached by retraction of the fatty tissue surrounding the aorta and inferior vena cava. One hemostatic clamp was applied to the aorta just below the diaphragm to avoid any excessive bleeding. The aorta was cut from below this level and dissected downwards. Care was taken not to puncture the aorta. Approximately 1.5–2 cm aorta including common iliac arteries and the a. sacralis mediana was harvested (Fig. 1A, B). The aorta was irrigated with phosphate buffer to remove the blood clots. Small branches (lumbar arteries) originating from the aorta were cauterized to prevent regeneration of nerve fibers out of the conduit. (Fig. 1C, D). Harvested 3-way-conduits were kept in 0.1 M phosphate buffered saline (PBS, pH 7.4). Two surgical teams worked separately to avoid time delay between 3-way-conduit harvesting and facial nerve reconstruction.

Fig.1

Preparation of the three-way conduit. A: The abdominal aorta with its three branches; ivc – inferior vena cava; sm – a. sacralis mediana; lcia – left common iliac artery; rcia – right common iliac artery. B: Appearance of the abdominal aorta with its trifurcation after harvesting; abbreviations like in A. C: Appearance of the abdominal aorta after cauterization of lumbar artery D: Dimensions of the three-way-conduit. Numbers placed on the short arms of the conduit show the entubulation sites for the zygomatic, buccal and marginal mandibular branches, respectively.

2.4 Surgical procedures

All animals were anesthetized with a mixture of Ketamine (100 mg/kg) and Xylazine HCl (5 mg/kg) injected intraperitoneally. All surgical procedures were performed unilaterally (right side) under an operating microscope with fiberoptic illumination (Olympus SZ61).

2.4.1 Approach to the facial nerve and its branches

Animals were positioned with a suture retractor applied to the teeth. The surgical area was shaved with a razor blade. A skin incision approximately 3 cm in length was made and the wound kept open using four retractors made from silk sutures and enabling easy access. The visible lacrimal gland was retracted anteriorly. The main trunk of the facial nerve and its branches (parotid plexus) were approached by light retraction of the surrounding soft tissue. For the sham-operated animals (Group 2) the surgical intervention ended at this stage. A subsequent wound closure followed. In all other experimental groups (3–6) the retromandibular vein crossing over the zygomatic and buccal branches was cauterized.

2.4.2 Facial-facial-anastomosis (FFA)

Transection and end-to-end suture of the proximal and distal stumps of the facial nerve was performed unilaterally under an operating microscope in 12 rats (Group 3). Animals were anesthetized and surgery was undertaken on the right side of the face. The exposed facial nerve trunk was transected and immediately sutured end-to-end with 10.0 atraumatic suture material (drawing not shown, see however Fig. 1A in Guntinas-Lichius et al. (2007).

2.4.3 Hypoglossal-facial anastomosis (HFA)

Transection and end-to-end suture of the proximal stump of hypoglossal and distal stump of facial nerve was performed unilaterally (on the right side of the face) under an operating microscope in 12 rats (Group 5). The parotid plexus of the facial nerve and the ipsilateral hypoglossal nerve were exposed. The hypoglossal nerve was transected at the level of its bifurcation into medial and lateral branches. The facial nerve was cut after emerging from the stylomastoid foramen and its proximal stump was ligated to prevent regeneration. The distal facial nerve stump was sutured to the proximal hypoglossal fragment with 10.0 atraumatic sutures (drawing not shown, see however Fig. 1B in Guntinas-Lichius et al. (2007).

2.4.4 Facial-facial nerve repair using a 3-way-conduit

The 3-way-conduit was placed over the exposed facial trifurcation and shortened to match the 5 mm long interstump distance between the facial nerve trunk and the distal stump of each branch. Then the facial nerve trunk, the zygomatic, buccal and marginal mandibular branches were exposed and transected (Figs. 2B and 3). The distal portion of the facial nerve trunk and the proximal fragments of its branches were removed. The proximal stump of the facial nerve was inserted into the long arm of the 3-way-conduit and the distal stumps of the major branches were into the its 3 short arms. All nerves were inserted into the conduit and affixed to its walls with 10.0 silk sutures. A nerve gap of 5 mm remained between the nerve stumps (Figs. 2C and 3).

Fig.2

Schematic drawing of the facial-facial anastomosis and hypoglossal-facial anastomosis with the 3-way-conduit. A: Anatomy of the intact facial and hypoglossal nerves; fn – facial nerve; zb – zygomatic branch; bb – buccal branch; mm – marginal mandibular branch; hn – hypoglossal nerve. B: Transection and excision of the facial nerve trunk trunk together with its zygomatic, buccal, marginal mandibular branches. C: Facial-facial nerve repair (FFA) with 3-way-conduit. D: Hypoglossal-facial nerve repair (HFA) with 3-way-conduit.

Fig.3

Facial-facial nerve repair with 3-way-conduit. A: Identification of facial branches (zb, bb and mm) within the parotid gland (pg) after the skin incision; eolg – extraorbital lacrimal gland. B: Appearance of the facial nerve and its branches after elevation of external auditory canal cartilage and peripheral venous structures. C: Better exposure of the nerve and its branches after cauterization of the retromandibular vein crossing over the zygomatic and buccal branches; cb – cervical branch. D: Entubulation of the zygomatic, buccal and marginal mandibular branches into the first, second, third rays of the 3-way-conduit, respectively. The cervical branch has been always ligated to avoid sprouting.

2.4.5 Hypoglossal-facial nerve repair using a 3-way-conduit

This reconstruction technique was used in 12 rats (Group 6). The proximal stump of the transected hypoglossal nerve was entubulated into the long arm of the 3-way-conduit tube and secured with two sutures. The distal stumps of the zygomatic, buccal and marginal mandibular branches were entubulated and sutured to the short arms of the 3-way-conduit. A nerve gap of 5 mm remained between the nerve stumps (Figs. 2D and 4).

Fig.4

Hypoglossal-facial anastomosis by 3-way-conduit. A, B: Entubulation of the zygomatic, buccal and marginalis mandibulae branches into the first, second, third rays of the three-way-conduit, respectively. The proximal stump of the facial nerve trunk has been ligated to avoid regenerative sprouting.

2.5 Analysis of motor performance

Video-based motion analysis of vibrissal motor performance is well established by our group (Angelov et al., 2005; Bendella et al., 2016; Bischoff et al., 2009; Grosheva et al., 2018).

Under light anesthesia, all vibrissal hairs were clipped with the exception of two large vibrissae in row C (i.e. the third row from dorsal) on each side of the face (Fig. 5). All animals were video-taped for 3–5 min during active exploration using a digital videocamera (Sony Handycam DCR-SR70 HDD Camcorder). Selected sequences containing pronounced whisks (vibrissal bouts) on the intact side contralateral and the side ipsilateral to surgery were captured by a 2DÚManual Advanced Video System (WinAnalyze, Mikromak, Berlin, Germany).

Fig.5

Schematical drawings illustrating the procedure of retrograde labeling with fluorescent tracers after 3-way conduit FFA (A) and HFA (B). Representative pictures from the facial (C) and hypoglossal (D) nucleus show red, green, and blue stained neurons whose neurons project into the zygomatic, buccal, and marginal mandibular branches, respectively. Double- (FR + FB, FR + FE, and FB + FE) or triple-(FR+FB+FE) labeled neurons (arrows, also in the insets) are consequence of collateral axonal branching at the lesion site.

The geometrical model which we used consisted of three reference points: (i) a point in the medial sagittal line close to the end of the nose and points corresponding to the medial angle of the left (ii) and right (iii) orbita (see Fig. 4 in Hizay et al., 2012). The model was used to evaluate (i) whisking frequency (cycles of protraction and retraction per second), (ii) angle at maximal protraction (the rostrally open angle between the midsagittal plane and the hair shaft in degrees), (iii) amplitude (the difference between maximal retraction and maximal protraction in degrees), (iv) angular velocity during protraction in degrees per second and (v) angular acceleration during protraction in degrees per second (Ozsoy et al., 2011).

2.6 Estimation of axonal branching by fluorescence labelling

2.6.1 Application of fluorescent tracers

Half of the animals in each group (6 rats) were used for this purpose. After sufficient anesthesia, the zygomatic, buccal, and marginalis mandibulae branches of the facial nerve were instilled with crystals of Fluoro-Ruby (FR, Molecular Probes, Cat. No. D-1817), Fast Blue (FB; EMS-Chemie GmbH, Groß-Umstadt, Germany) and Fluoro-Emerald (FE; Molecular Probes, Cat. No. D-1820) respectively. Crystals were left in situ (60 min). Then the application sites were rinsed, dried and the wound closed.

2.6.2 Animal sacrifice and tissue harvesting

Fifteen days after instillation of the fluorescent tracers, animals were transcardially perfused with 4% paraformaldehyde in 0.1 M phosphate buffered saline (PBS), pH 7.4. Brainstems as well as the levator labii superioris (LLS) muscles were harvested. All specimens were kept in 4% paraformaldehyde in 0.1 M phosphate buffered saline, pH 7.4 (in dark at 4°C).

2.6.3 Analysis of collateral axonal branching at the lesion site

Brainstems were sectioned coronally at 50μm using a vibratome. Sections were observed using an epifluorescence microscope (Axioplan, Zeiss, Oberkochen, Germany) through an UV-excitation filter (Carl Zeiss, Filter Set 01: Excitation BP 365/12, Emission LP 397), allowing identification of FB-labeled neurons (blue). A Filter Set 15 (Carl Zeiss, Excitation BP 546/12, Beamsplitter FT 580, Emission LP 590) and Filter Set 10 (Carl Zeiss, Excitation BP 450–490, Beamsplitter FT 510, Emission BP 515–565) allowed identification of FR- (in red) and FE- (in green) labelled motoneurons respectively (Fig. 5).

The fractionator principle (Gundersen, 1986) was employed to include all retrogradely labelled motoneurons with a visible cell nucleus in every third section through the facial nucleus. Details have been described previously (Dohm et al., 2000; Tomov et al., 2002). All separate color images were captured by a CCD Video Camera System (Optronics DEI-470, Goleta, CA, USA) and combined by means of image analysis software (ImagePro Plus 6.2; Media Cybernetics, Baltimore, MD). ”FB-masks” of all FB-labeled motoneurons were then superimposed over the FR- and FE-labeled perikarya. The technique allowed us to readily identify cells stained by FB^only, or FR^only, or by FE^only as well as all those double labelled (FB + FR, FB + FE, or FR + FE) and triple labeled (FR + FB + FE) by which were then counted manually on the computer screen (Fig. 5). All counts were performed in a blind fashion.

Although time-consuming, the procedure allows quantification of the degree (index) of collateral axonal branching which represents the ratio of motoneurons projecting branched axons into the zygomatic, buccal and marginal mandibular branches to all motoneurons sending axons through both branches, i.e. the percentage of double- and triple-labeled motoneurons. Rats with an intact facial nerve trunk subjected only to surgery for tracer application had an index of axonal branching of 0%. The proportion of double-labeled motoneurons determines the ratio of collateral axonal branching, which is an indirect sign for the quality of axonal regrowth: the more double- or triple-labelled perikarya, the stronger the collateral branching of axons at the lesion site, the poorer the navigation.

2.7 Analysis of target muscle reinnervation

Also one day after videotaping, the remaining six rats from each group were used to establish the quality of two target muscles reinnervation. Following perfusion fixation the right levator labii superioris (LLS) muscle was dissected free, postfixed in 4% paraformaldehyde overnight and cryoprotected in 20% sucrose in PBS. Longitudinal sections (30μm thick) were cut on a cryostat and mounted on SuperFrost Plus slides (Art. No. J1800 AMNZ, Menzel-Gläser, Germany). The number of sections obtained from individual muscles was within the range of 32–37. The tissue slices were stored at –80°C until immunostaining.

Using the fractionator principle (Gundersen, 1986), every third section was processed as follows. Sections were air-dried and rinsed (3×10 minutes in 0.1 M PBS, pH 7.4). Nonspecific binding was blocked using PBS containing 0.2% Triton X-100 (Sigma, St. Louis, USA; CAS Number 9002-93-1; facilitates antibody penetration), 0.02% sodium azide (Sigma, St. Louis, USA; CAS Number 26628-22-8) and 5% normal sheep serum (NSS; Sigma, St. Louis, USA, Cat. Nr. S3772) for 30 minutes at room temperature. Sections were incubated with primary antibody (rabbit polyclonal against neuronal class III ß-tubulin, Covance, No. PRB- 435P) diluted 1:2000 in PBS containing 0.5% lambda-carrageenan (Sigma, St. Louis, USA, CAS Number 9064-57-7; a non-gelling vegetable gelatin which reduces unspecific binding and stabilizes the antibody solution) and 0.02% sodium azide, overnight at 4°C. All sections were stained in the same solution kept in glass-capped staining glass cuvettes (capacity 250 ml, 20 slides, Roth). The next morning, sections were brought to room temperature, rinsed in PBS and incubated with secondary antibody (Cy3-conjugated sheep anti-rabbit IgG, Sigma, St. Louis, USA; Cat Nr. C2306) diluted 1:200 in PBS-carrageenan solution for 2 hours at room temperature. After rinsing as indicated above, sections were incubated with Alexa Fluor 488-conjugated alpha-bungarotoxin (Molecular Probes, Cat. Nr. B-13422) at a dilution of 1:500 in PBS-sodium azide solution. Slides were rinsed, air-dried, coverslipped and observed with a Zeiss Axioskop 50 epifluorescence microscope through “rhodamine” (No. 15 of Carl Zeiss) and “fluorescein” (No. 10 of Carl Zeiss) filters.

The quality of end-plate reinnervation was evaluated by a simple and straightforward criterion: the number of axonal branches (identified by beta-tubulin staining) that enter or, in some cases, possibly leave, the boundaries of individual end-plates (identified by acetylcholine receptor staining with alpha-bungarotoxin). Entries by preterminal branches of one axon were counted as single events. According to this criterion, end-plates were identified as ‘monoinnervated’ (one axon), ‘polyinnervated’ (two or more axons) or denervated (no visible axonal association with the receptor staining; see Fig. 6 in Hizay et al., 2012). Counts of endplates were performed directly under the microscope (using×40 magnification) in a blind fashion.

2.8 Statistics

For any given parameter, data from all six experimental groups were tested using one-way analysis of variance (one-way ANOVA) or Kruskal-Wallis test for overall experimental effects. The Mann-Whitney U test was used to compare numbers of single- (FB^only, FR^only and FE_only), double- (FB + FR,FB + FE,FR + FE), and triple-(FR + FB + FE) labeled neurons in the facial nucleus (or hypoglossal nucleus in Groups 5 and 6). If significant effects were detected (p < 0.05), each surgical group was compared with the intact group using Tukey’s or Dunn post-hoc test at a significance level of 0.05. For analysis, GraphPad Prism version 5.0 (GraphPad Software, Inc, San Diego, Calf) was used.

3 Results

3.1 Reconstruction of the facial nerve with a 3-way-conduit does not improve functional recovery of vibrissal whisking

3.1.1 Qualitative observations on the motor performance of the mystacial vibrissae

Under normal physiological conditions, rat mystacial vibrissae are erect and oriented anteriorly. Their simultaneous sweeps, known as “whisking” or “sniffing” (Semba et al., 1980; Welker, 1964) occur 5–11 times per second (Bermejo et al., 1996; Carvell & Simons, 1990; Komisaruk, 1970). The key movements of this motor activity are the protraction and retraction of the vibrissal hairs by the piloerector (follicular) muscles. The striated muscle fibers mediating protraction form a sling around the rostral aspect of each hair follicle: contraction of these muscles pulls the base of the follicle caudally, moving the distal aspects of the whisker hair forward. By contrast, retraction of the vibrissae depends primarily upon passive elastic properties of the deep connective tissue (Dorfl, 1985; Wineski, 1985). All vibrissal muscles are innervated by the buccal and marginal mandibular branches of the facial nerve (Dorfl, 1985).

Following facial nerve transection and surgical reconstruction (classic FFA, FFA + 3-way-conduit, classic HFA, or HFA+3-way-conduit) the vibrissae initially dropped down and became motionless. Thereafter they gradually “rose” to the level of the mouth and acquired a posterior orientation. Overall, restoration of function was poor in all 4 surgical groups (Groups 3–6) with whiskers failing to show normal rhythmic sweeps. In contrast, rats of Group 2 (sham-operated) recovered whisking movements on day 1 after surgery.

3.1.2 Biometric analysis of whisking behavior

Intact rats. During exploration, mystacial vibrissae swept back and forth with a frequency of about 6 Hz. The mean amplitude of whisking (the difference between maximal retraction and maximal protraction in degrees) measured about 60û. These movements were performed at a sagittal angular velocity of about 1300°/sec and a sagittal angular acceleration of 30000°/sec² (Table 1). These results are consistent with previous observations (Guntinas-Lichius et al., 2001; Guntinas-Lichius et al., 2002; Tomov et al., 2002).

Table 1
Biometric analysis of the vibrissal hair movements. Small numbers in the superscripts indicate the group with significantly different values. (ANOVA and post-hoc Tukey’s test, p < 0.05)

Group Amplitude (in degrees) Frequency (in Hz) Angular velocity during protraction (degrees per sec) Angular velocity during protraction (degrees/sec²)

Group 1, Intact 59±7^3,4,5,6 6.8±0.7^3,4,5,6 1378±402^3,4,5,6 28438±5086^3,4,5,6

Group 2: Sham operated 56±5^3,4,5,6 6.4±0.8^3,4,5,6 1298±237^3,4,5,6 31982±7017^3,4,5,6

Group 3: FFA 20±5^1,2,4 3.1±1.3^1,2 271±123^1,2 7654±1015^1,2

Group 4: FFA+3 way conduit 14±4^1,2,3 3.4±1.1^1,2 203±98^1,2 5089±991^1,2

Group 5: HFA 22±6^1,2,6 4.0±1.8^1,2 285±95^1,2 6813±2021^1,2

Group 6: HFA+3 way conduit 17±8^1,2,5 2.9±0.6 ^1,2 161±157^1,2 5359±2817^1,2

Group	Amplitude (in degrees)	Frequency (in Hz)	Angular velocity during protraction (degrees per sec)	Angular velocity during protraction (degrees/sec²)
Group 1, Intact	59±7^3,4,5,6	6.8±0.7^3,4,5,6	1378±402^3,4,5,6	28438±5086^3,4,5,6
Group 2: Sham operated	56±5^3,4,5,6	6.4±0.8^3,4,5,6	1298±237^3,4,5,6	31982±7017^3,4,5,6
Group 3: FFA	20±5^1,2,4	3.1±1.3^1,2	271±123^1,2	7654±1015^1,2
Group 4: FFA+3 way conduit	14±4^1,2,3	3.4±1.1^1,2	203±98^1,2	5089±991^1,2
Group 5: HFA	22±6^1,2,6	4.0±1.8^1,2	285±95^1,2	6813±2021^1,2
Group 6: HFA+3 way conduit	17±8^1,2,5	2.9±0.6 ^1,2	161±157^1,2	5359±2817^1,2

Operated rats. With the exception of the sham-operated rats (Group 2) large functional deficiencies were evidenced by the significantly reduced frequency of whisking, the smaller amplitude of vibrissae movement as well as the low angular velocity and acceleration during protraction (Table 1). These post-operative changes were due to inadequate muscle function during the active protraction phase. The mean amplitude (the most indicative parameter) that we measured after the classic coaption operations FFA and HFA (20°±5° and 22°±6° respectively) was - though very far away from those in intact rats (about 60°) - better than after their 3-way-conduit equivalents (14°±4° and 17°±8°). Taken together, these findings show that, regardless of the type of reconstructive surgery, the range and velocity of movements remained severely impaired even four months after reconstructive surgery.

3.2 A slight trend to reduce the degree of collateral axonal branching at the lesion site after reconstruction of the facial nerve with a 3-way-conduit

3.2.1 Intact rats

Motoneurons innervating muscles through the zygomatic, buccal or marginal mandibular branches are localized in three distinct subnuclei of the facial nucleus – the dorsal, lateral and intermediate subnucleus, respectively (Fig. 5). This is the so called myotopic or somatotopic organization of the motoneurons in the facial nucleus (Aldskogius & Thomander, 1986; Ito & Kudo, 1994; Klein & Rhoades, 1985; Semba & Egger, 1986).

3.2.2 Operated rats

After any type of reconstructive (but not sham) surgery, two major changes, characteristic for the post-transectional axonal regrowth, were detected. First, myotopic organization into facial subnuclei was no longer present, i.e. all retrogradely labeled motoneurons were scattered throughout the facial nucleus (Fig. 5). This lack of myotopy was due to the regrowth of axons into the wrong facial nerve branch (r. zygomaticus, r. buccalis and r. marginalis mandibulae). Second, the entire facial nucleus contained double- (FR+FB, FR+FE, FB+FE) and even triple-labeled (FR + FB + FE) motoneuronal perikarya (arrows in Fig. 5), which arose due to collateral axonal branching at the lesion site; the collateral branches grew simultaneously into different facial nerve rami and retrogradely transported the different fluorescent dyes to their parent motoneurons in the facial nucleus.

Contrary to our hypothesis, the amounts of double- and triple-labeled perikarya were insignificantly reduced in animals receiving a 3-way-conduit for FFA (17.3%) and for HFA (19.5%) when compared to their coaptation equivalents (19.3 % for FFA and 23.5% for HFA). Thus we may conclude that there was only a trend for reduction in the proportion of double-labelled perikarya after FFA+3-way-conduit vs. FFA classic and after HFA+3-way-conduit vs. HFA-classic (see last column of Table 2).

Table 2
Collateral axonal branching at the lesion site

Group Neurons projecting only in r. zygomaticus (FR only) Neurons projecting only in r. buccalis (FB only) Neurons projecting only in ramus marginalis mandibulae (FE only) Neurons projecting in r. zygomaticus and in r. buccalis (FR+FB) Neurons projecting in r.zygomaticus and in ramus (FR+FE) Neurons projecting in r. buccalis and in ramus marginalis mandibulae (FB+FE) Neurons projecting in r. zygomaticus, r. buccalis and in ramus marginalis mandibulae (FR+FB+FE) Degree of collateral axonal branching In %

Group 1, Intact 570±72 1947±584 1422±424 0 0 0 0 0

Group 2: Sham operated 612±104 1911±408 1446±419 0 0 0 0 0

Group 3: FFA 519±115 1590±203 1083±365 201±66 99±30 303±99 15±3 19.3

Group 4: FFA+3 way conduit 558±142 1713±297 1209±313 240±72 108±54 234±102 21±6 17.3

Group 5: HFA 756±128 1888±402 1364±378 234±81 232±52 436±140 40±8 23.5

Group 6: HFA +3 way conduit 788±96 1968±504 1440±402 171±24 208±44 404±144 36±4 19.5

Group	Neurons projecting only in r. zygomaticus (FR only)	Neurons projecting only in r. buccalis (FB only)	Neurons projecting only in ramus marginalis mandibulae (FE only)	Neurons projecting in r. zygomaticus and in r. buccalis (FR+FB)	Neurons projecting in r.zygomaticus and in ramus (FR+FE)	Neurons projecting in r. buccalis and in ramus marginalis mandibulae (FB+FE)	Neurons projecting in r. zygomaticus, r. buccalis and in ramus marginalis mandibulae (FR+FB+FE)	Degree of collateral axonal branching In %
Group 1, Intact	570±72	1947±584	1422±424	0	0	0	0	0
Group 2: Sham operated	612±104	1911±408	1446±419	0	0	0	0	0
Group 3: FFA	519±115	1590±203	1083±365	201±66	99±30	303±99	15±3	19.3
Group 4: FFA+3 way conduit	558±142	1713±297	1209±313	240±72	108±54	234±102	21±6	17.3
Group 5: HFA	756±128	1888±402	1364±378	234±81	232±52	436±140	40±8	23.5
Group 6: HFA +3 way conduit	788±96	1968±504	1440±402	171±24	208±44	404±144	36±4	19.5

3.3 Increased degree of polyinnervation of the neuromuscular junctions (NMJ) after reconstruction of the facial nerve with a 3-way-conduit

The quality of target muscle reinnervation was evaluated in levator labii superioris (LLS) muscles. LLS is an extrinsic vibrissal muscle, which, similar to the intrinsic vibrissal muscles, is innervated by the “six longitudinal branches” of the buccal branch of the facial nerve (the short common trunk of the fused ramus buccolabialis superior and ramus buccolabialis inferior (Dorfl, 1985).

3.3.1 Intact animals

All NMJ were innervated by one axon and designated as monoinnervated (no picture included, see Hizay et al., 2012).

3.3.2 Operated rats

Four months after surgery, numerous polyinnervated (i.e., innervated by two or more axons) NMJ were observed in LLS (no picture provided, see Hizay et al., 2012). Their proportion was insignificantly higher in both animal groups that were subjected to 3-way-conduit reconstruction (27±3 % for FFA and 22±3 % for HFA) when compared to the values obtained in rats that received classic coaptation of the nerve trunks (18±4 % for FFA and 17±4 % for HFA). We may thus conclude that there was a trend for increased polyinnervation of the NMJ after reconstructive surgery with a 3-way conduit. No polyinnervation of NMJ was observed in sham-operated rats (Table 3).

Table 3
Quality of reinnervation (polyinnervation degree) of the levator labii superioris NMJ: Neuromuscular junction Small numbers in the superscripts indicate the group with significantly different values. (ANOVA and post-hoc Tukey’s test, p < 0.05)

Group Monoinnervated NMJ in % Polyinnervated NMJ in % Non-innervated NMJ in %

Group 1, Intact 100^3,4,5,6 0^3,4,5,6 0^3,4,5,6

Group 2: Sham operated 100^3,4,5,6 0^3,4,5,6 0^3,4,5,6

Group 3: FFA 77±7^1,2,4 18±4^1,2,4 5±2^1,2,4

Group 4: FFA+3 way conduit 64±3^1,2,3 27±3^1,2,3 9±4^1,2,3

Group 5: HFA 79±3^1,2,6 17±4^1,2,6 6±2^1,2,6

Group 6: HFA+3 way conduit 70±4^1,2,5 22±3^1,2,5 8±3^1,2,5

Group	Monoinnervated NMJ in %	Polyinnervated NMJ in %	Non-innervated NMJ in %
Group 1, Intact	100^3,4,5,6	0^3,4,5,6	0^3,4,5,6
Group 2: Sham operated	100^3,4,5,6	0^3,4,5,6	0^3,4,5,6
Group 3: FFA	77±7^1,2,4	18±4^1,2,4	5±2^1,2,4
Group 4: FFA+3 way conduit	64±3^1,2,3	27±3^1,2,3	9±4^1,2,3
Group 5: HFA	79±3^1,2,6	17±4^1,2,6	6±2^1,2,6
Group 6: HFA+3 way conduit	70±4^1,2,5	22±3^1,2,5	8±3^1,2,5

4 Discussion

Here we show that, when compared to coaptation alone, the use of an aortic 3-way-guide for reconstructive surgery of the facial nerve (FFA or HFA) promoted a slight reduction in the degree of collateral axonal branching at the injury site. This trend, however, was not accompanied by a reduction of NMJ-polyinnervation, nor by improved whisking function. Taken together, the results suggest that, although the 3-way-conduit is not an obstacle for axonal regrowth after surgical reconstruction of the facial nerve, it does not confer any added functional benefit.

To our knowledge, this is the first study to show that aortic 3-way-conduits do not impede axon regeneration of the zygomatic, buccal and marginal mandibular branches of the facial nerve in rats. Our findings support the current view that isogeneic aortic tubes promote robust peripheral nerve regeneration and do so over 5–10 mm gaps (Brushart, 1987; Brushart, 1988; Lundborg et al., 1986; Lundborg et al., 1997; Mackinnon et al., 1986; Politis, 1985; Politis et al., 1982; Seckel et al., 1986).

4.1 Role of intrafascicular topography for postlesional axonal pathfinding

Knowledge of the intrafascicular topography of a nerve may be of great importance (Sun et al., 2009). For example, awareness of the intrafascicular topography of the extratemporal portion of the facial nerve should improve outcome after microsurgery (Guerrissi & Gil Miranda, 2007). Unfortunately, the topography of the axons within the facial nerve trunk has yet to be determined.

Within the facial nerve trunk, one may only suppose that the majority of axons supplying the zygomatic branch presumably lie in its uppermost portion, with respect to the animal’s head. Accordingly, the majority of axons supplying the buccal branch may lie just below. Finally, the axons of the marginal mandibular branch may occupy the lowest part of the trunc. One could logically anticipate that the proximal portions of the zygomatic, buccal and marginal mandibular axons would simply regrow towards their respective distal branches that are contained in the short arms of the 3-way-conduit. Regretfully we do not know, whether our experiment restored the fascicular topography adequately: at present we are not able to perform such a study.

4.2 Collateral sprouting after peripheral nerve regeneration

Following injury, each lesioned axon gives rise to another 25 axons (axonal branches) (Brushart et al., 1998). Over the following time of regeneration, in a period of almost up to a year, some of the daughter axons are pruned off (Angelov et al., 2005; Brushart et al., 1998; Mackinnon et al., 1991). It is thought that these are the axons that during this time have failed to establish neuronal connections.

Immoderate collateral axonal sprouting and axonal inability to connect with peripheral targets are thought to be the main factors to lead to poor functional recovery. The reduction of collateral sprouting has been the goal of some studies. Usually the attempt is made by altering the extracellular matrix. Dohm et al. (2000) and Hristov et al. (2005) performed entubulations of transected rat facial nerves in a silicon chambers filled with phosphate buffered saline (PBS) pH 7.4, collagen type I (100μg/ml in PBS), laminin (20μg/ml in collagen type I), fibronectin (20μg/ml in collagen type I), tenascin-R (20μg/ml in collagen type I), semaphorin 3A/Fc chimera (120 ng/ml in collagen type I), neuropilin-1/Fc chimera (3μg/ml in collagen type I). Unfortunately, the efforts were not successful and excessive collateral sprouting seems to be a striking feature of peripheral nerve regeneration (Dohm et al., 2000). Nevertheless, experiments with modern techniques that attempt to improve guidance of regenerating peripheral axons are still going on, e.g. application of semaphorin 6A and nerve growth factor to improve guidance of sensory axons (Curley et al., 2014), use of nanoparticles (Huang et al., 2018), improved tubulation of the sciatic nerve (Liu et al., 2018) and studies showing that Schwann cells can direct regenerating motor axons that have been destroyed in colorectal carcinoma (Rosenberg et al., 2014).

We were therefore pleasantly surprised to document that the use of a 3-way-conduit promoted a trend for reduced the axonal collateralisation, as established by retrograde labelling which identified double- or triple-labelled motor neurons. It was anticipated that this effect would be even stronger after HFA+3-way-conduit. As already described by our group (Ozsoy et al., 2011) the guidance of the transected hypoglossal axons into the “facial-nerve periphery”, provides very beneficial conditions for their regrowth, because HFA creates a situation in which the sectioned nerve fibers of the proximal hypoglossal stump sprout into the distal facial stump, guided by 40% more rows of Schwann cells left over from degenerated facial axons. This situation of course differs from that of FFA in which the ratio of regenerated and degenerated nerve fibers is 1:1. This negative result may be attributed to increased number of the sprouting daughter axons into the 3-way-conduit since there were three branches of the facial nerve (zygomatic, buccal, and marginal mandibular) growing towards their distal muscle targets.

4.3 Polyinnervation of the neuro-muscular junctions (NMJ)

Regenerating axons undergo sprouting all the way until they approach their target. When reaching the target, there is additional branching (terminal or intramuscular) which leads to the reinnervation of multiple incorrect muscle fibers. This results to newly formed, large motor units (Son et al., 1996) and polyinnervation, that is the innervation of motor end-plates by more than one motoneuron, This type of polyinnervation at NMJ during regeneration persists for long periods of time, whereas transient polyneuronal innervation during development is quickly restored (Brown et al., 1981; Esslen, 1960; Grant et al., 2002; Greulich, 2005; Ijkema-Paassen et al., 2002; Jergovic< <et al., 2001; Mackinnon et al., 1991). Corroborating to the above, abnormally dense meshworks of intramuscular axonal branches were observed on a target muscle (LLS), a similar finding described after facial nerve crush, in the zygomatic muscles in transgenic Thy1-GFP mice (Magill et al., 2010).

Persistent polyinnervation of muscle fibers after peripheral nerve regeneration is thought to be the main factor restricting recovery (Brown et al., 1981; Esslen, 1960; Grant et al., 2002; Greulich, 2005; Ijkema-Paassen et al., 2002; Jergovic< <et al., 2001; Magill et al., 2010; Rich & Lichtman, 1989). A muscle fiber is controlled by two or even more motoneurons, which are often functionally different and may operate asynchronously (Barry & Ribchester, 1995; Friede & Bischhausen, 1980; Gorio et al., 1983; Grimby et al., 1989; Schröder, 1968a; Tam & Gordon, 2003; Trojan et al., 1991).

The study demonstrates that the application of a 3-way-conduit for facial nerve reconstruction does not impede axonal pathfinding across the lesion. However, it does not significantly improve vibrissal whisking, a fact that highlights the importance of NMJ polyinnervation and depicts its significant role in recovery.

Recent studies which focus on the reduction of polyinnervation in muscle fibers have been more successful in terms of functional recovery. Following facial and hypoglossal nerve injury, motor end-plate polyneuronal innervation was reduced and whisking, blink reflexes and tongue position were improved in studies where the whisker-pads, the eyelids and the extrinsic/intrinsic muscles of the tongue, were mechanically stimulated (Angelov et al., 2007; Bischoff et al., 2009; Evgenieva et al., 2008; Guntinas-Lichius et al., 2005). It seems that the training of the muscles during the recovery period, when reinnervation takes place, enhances improvement.

In conclusion, physical axonal pathfinding reduces the extent of axonal collateralization, but does not lead to recovery of function. There is increasing evidence suggesting that research should focus on reducing the extent of polyinnervation at the level of neuromuscular junction. Additional interventions such as physical therapy of the reinnervated distal targets should be kept in mind instead of leaving them to their fate. We also think that further studies using conduits filled with relevant neurotrophic factors (Bendella et al., 2017) should be carried out to elucidate some molecular mechanisms which govern axonal pathfinding. In this way, the future use of 3-way-conduits would promote better functional recovery after facial nerve reconstruction.

Conflict of interest

There was no conflict of interest for any of the authors.

Footnotes

Acknowledgments

This study was supported by TUBITAK (The Scientific and Technical Research Council of Turkey, project number: 113S032) and the TUR-INTEN-C-025 joint project of TUBITAK and BMBF (German Federal Ministry for Education and Research).

References

Aldskogius

, Thomander

(1986). Selective reinnervation of somatotopically appropriate muscles after facial nerve transection and regeneration in the neonatal rat. Brain Research, 375(1), 126–134.

Angelov

D.N.

, Ceynowa

, Guntinas-Lichius

, Streppel

, Grosheva

, Kiryakova

S.I.

, Skouras

, Maegele

, Irintchev

, Neiss

W.F.

, Sinis

, Alvanou

, Dunlop

S.A.

(2007). Mechanical stimulation of paralyzed vibrissal muscles following facial nerve injury in adult rat promotes full recovery of whisking. Neurobiology of Disease, 26(1), 229–242.

Angelov

D.N.

, Guntinas-Lichius

, Wewetzer

, Neiss

W.F.

, Streppel

(2005). Axonal branching and recovery of coordinated muscle activity after transection of the facial nerve in adult rats. In: Beck

F.F.

, Christ

, Clascá

, Haines

D.E.

, Korf

H.W.

, Kummer

, Marani

, Putz

, Sano

, Schiebler

T.H.

, & Zillers

, editors. Advances in Anatomy Embryology and Cell Biology. Berlin, Heidelberg, Germany: Springer Verlag, pp. 3–25

Barry

J.A.

, Ribchester

R.R.

(1995). Persistent polyneuronal innervation in partially denervated rat muscle after reinnervation and recovery from prolonged nerve conduction block. The Journal of Neuroscience, 15(10), 6327–6339.

Bendella

, Brackmann

D.E.

, Goldbrunner

, Angelov

D.N.

(2016). Nerve crush but not displacement-induced stretch of the intra-arachnoidal facial nerve promotes facial palsy after cerebellopontine angle surgery. Experimental Brain Research, 234(10), 2905–2913.

Bendella

, Rink

, Grosheva

, Sarikcioglu

, Gordon

, Angelov

D.N.

(2017). Putative roles of soluble trophic factors in facial nerve regeneration, target reinnervation, and recovery of vibrissal whisking. Experimental Neurology, 300, 100–110.

Bermejo

, Harvey

, Gao

, Zeigler

H.P.

(1996). Conditioned whisking in the rat. Somatosensory and Motor Research, 13(3-4), 225–233.

Bischoff

, Grosheva

, Irintchev

, Skouras

, Kaidoglou

, Michael

, Angelova

S.K.

, Kuerten

, Sinis

, Dunlop

S.A.

, Angelov

D.N.

(2009). Manual stimulation of the orbicularis oculi muscle improves eyelid closure after facial nerve injury in adult rats. Muscle & Nerve, 39(2), 197–205.

Brown

M.C.

, Holland

R.L.

, Hopkins

W.G.

, Keynes

R.J.

(1981). An assessment of the spread of the signal for terminal sprouting within and between muscles. Brain Research, 210(1-2), 145–151.

10.

Brushart

T.M.

(1987). Neurotropism and neurotrophism. The Journal of Hand Surgery, 12(5 Pt 1), 808–809.

11.

Brushart

T.M.

(1988). Preferential reinnervation of motor nerves by regenerating motor axons. The Journal of Neuroscience, 8(3), 1026–1031.

12.

Brushart

T.M.

, Gerber

, Kessens

, Chen

Y.G.

, Royall

R.M.

(1998). Contributions of pathway and neuron to preferential motor reinnervation. The Journal of Neuroscience, 18(21), 8674–8681.

13.

Büngner

(1891). Ueber die Degenerations- und Regenerationsvorgänge am Nerven nach Verletzungen.. Beitråge Pathologische Anatomie, 10, 321–387.

14.

Carvell

G.E.

, Simons

D.J.

(1990). Biometric analyses of vibrissal tactile discrimination in the rat. The Journal of Neuroscience, 10(8), 2638–2648.

15.

Curley

J.L.

, Catig

G.C.

, Horn-Ranney

E.L.

, Moore

M.J.

(2014). Sensory axon guidance with semaphorin 6A and nerve growth factor in a biomimetic choice point model. Biofabrication, 6(3), 035026.

16.

Dohm

, Streppel

, Guntinas-Lichius

, Pesheva

, Probstmeier

, Walther

, Neiss

W.F.

, Stennert

, Angelov

D.N.

(2000). Local application of extracellular matrix proteins fails to reduce the number of axonal branches after varying reconstructive surgery on rat facial nerve. Restorative Neurology and Neuroscience, 16(2), 117–126.

17.

Dorfl

(1985). The innervation of the mystacial region of the white mouse: A topographical study. Journal of Anatomy, 142, 173–184.

18.

Esslen

(1960). Electromyographic findings on two types of misdirection of regenerating axons. Electroencephalography and Clinical Neurophysiology, 12, 738–741.

19.

Evgenieva

, Schweigert

, Guntinas-Lichius

, Pavlov

, Grosheva

, Angelova

, Streppel

, Irintchev

, Skouras

, Kuerten

, Sinis

, Dunlop

, Radeva

, Angelov

D.N.

(2008). Manual stimulation of the suprahyoid-sublingual region diminishes polynnervation of the motor endplates and improves recovery of function after hypoglossal nerve injury in rats. Neurorehabilitation and Neural Repair, 22(6), 754–768.

20.

Fansa

, Keilhoff

, Plogmeier

, Frerichs

, Wolf

, Schneider

(1999). Successful implantation of Schwann cells in acellular muscles. Journal of Reconstructive Microsurgery, 15(1), 61–65.

21.

Fansa

, Keilhoff

, Wolf

, Schneider

, Gold

B.G.

(2001). Tissue Engineering of Peripheral Nerves: A Comparison of Venous and Acellular Muscle Grafts with Cultured Schwann Cells. Plastic and Reconstructive Surgery, 107(2), 495–496.

22.

Fansa

, Schneider

, Wolf

, Keilhoff

(2002). Influence of insulin-like growth factor-I (IGF-I) on nerve autografts and tissue-engineered nerve grafts. Muscle & Nerve, 26(1), 87–93.

23.

Friede

R.L.

, Bischhausen

(1980). The fine structure of stumps of transected nerve fibers in subserial sections. Journal of the Neurological Sciences, 44(2-3), 181–203.

24.

Gorio

, Carmignoto

, Finesso

, Polato

, Nunzi

M.G.

(1983). Muscle reinnervation–II. Sprouting, synapse formation and repression. Neuroscience, 8(2), 403–416.

25.

Grant

G.A.

, Rostomily

R.R.

, Kim

D.K.

, Mayberg

M.R.

, Farrell

, Avellino

, Duckert

L.G.

, Gates

G.A.

, Winn

H.R.

(2002). Delayed facial palsy after resection of vestibular schwannoma. Journal of Neurosurgery, 97(1), 93–96.

26.

Greulich

(2005). Anchoring the nasolabial fold. In: Beurskens

C.H.G.

, van Gelder

R.S.

, Heymans

, Manni

J.J.

, Nicolai

J.P.A.

, editors. The Facial Palsies. Utrecht: Lemma Publishers. pp. 235–243.

27.

Grimby

, Einarsson

, Hedberg

, Aniansson

(1989). Muscle adaptive changes in post-polio subjects. Scandinavian Journal of Rehabilitation Medicine, 21(1), 19–26.

28.

Grosheva

, Rink

, Jansen

, Bendella

, Pavlov

S.P.

, Sarikcioglu

, Angelov

D.N.

, Dunlop

S.A.

(2018). Early and continued manual stimulation is required for long-term recovery after facial nerve injury. Muscle & Nerve, 57(1), 100–106.

29.

Guerrissi

J.O.

, Gil Miranda

M.F.

(2007). Intraneural topography of the extratemporal facial nerve: Microsurgical nerve reconstruction. Journal of Craniofacial Surgery, 18(3), 578–585.

30.

Gundersen

H.J.

(1986). Stereology of arbitrary particles. A review of unbiased number and size estimators and the presentation of some new ones, in memory of William R. Thompson. Journal of Microscopy, 143(Pt 1), 3–45.

31.

Guntinas-Lichius

, Angelov

D.N.

, Tomov

T.L.

, Dramiga

, Neiss

W.F.

, Wewetzer

(2001). Transplantation of olfactory ensheathing cells stimulates the collateral sprouting from axotomized adult rat facial motoneurons. Experimental Neurology, 172(1), 70–80.

32.

Guntinas-Lichius

, Hundeshagen

, Paling

, Streppel

, Grosheva

, Irintchev

, Skouras

, Alvanou

, Angelova

, Kuerten

, Sinis

, Dunlop

S.S.

, Angelov

D.N.

(2007). Manual stimulation of facial muscles improves functional recovery after hypoglossal-facial anastomosis and interpositional nerve grafting of the facial nerve in adult rats. Neurobiology of Disease, 28(1), 101–112.

33.

Guntinas-Lichius

, Irintchev

, Streppel

, Lenzen

, Grosheva

, Wewetzer

, Neiss

W.F.

, Angelov

D.N.

(2005). Factors limiting motor recovery after facial nerve transection in the rat: Combined structural and functional analyses. European Journal of Neuroscience, 21(2), 391–402.

34.

Guntinas-Lichius

, Wewetzer

, Tomov

T.L.

, Azzolin

, Kazemi

, Streppel

, Neiss

W.F.

, Angelov

D.N.

(2002). Transplantation of olfactory mucosa minimizes axonal branching and promotes the recovery of vibrissae motor performance after facial nerve repair in rats. The Journal of Neuroscience, 22(16), 7121–7131.

35.

Hadlock

, Elisseeff

, Langer

, Vacanti

, Cheney

(1998). A tissue-engineered conduit for peripheral nerve repair. Archives of Otolaryngology Head & Neck Surgery, 124(10), 1081–1086.

36.

Hizay

, Ozsoy

, Demirel

B.M.

, Ozsoy

, Angelova

S.K.

, Ankerne

, Sarikcioglu

S.B.

, Dunlop

S.A.

, Angelov

D.N.

, Sarikcioglu

(2012). Use of a Y-tube conduit after facial nerve injury reduces collateral axonal branching at the lesion site but neither reduces polyinnervation of motor endplates nor improves functional recovery. Neurosurgery, 70(6), 1544–1556.

37.

Hristov

, Guntinas-Lichius

, Smith

, Angelov

D.N.

(2005). Focal application of the cell signaling molecules semaphorin-3A and neuropilin-1 fails to improve recovery of function after peripheral nerve lesion. 22. Arbeitstagung der Anatomischen Gesellschaft, Würzburg. 28. bis 30. September 2005. 62 [Abstract].

38.

Huang

Y.A.

, Ho

C.T.

, Lin

Y.H.

, Lee

C.J.

, Ho

S.M.

, Li

M.C.

, Hwang

(2018). Nanoimprinted anisotropic topography preferentially guides axons and enhances nerve regeneration. Macromolecular Bioscience, 18(12), e1800335.

39.

Hudson

T.W.

, Evans

G.R.

, & Schmidt

C.E.

(2000). Engineering strategies for peripheral nerve repair. Orthopedic Clinics of North America, 31(3), 485–498.

40.

Hwang

, Kim

S.G.

, Kim

D.J.

(2006). Facial-hypoglossal nerve anastomosis using laser nerv welding. Journal of Craniofacial Surgery, 17(4), 687–691.

41.

Ijkema-Paassen

, Meek

M.F.

, Gramsbergen

(2002). Reinnervation of muscles after transection of the sciatic nerve in adult rats. Muscle & Nerve, 25(6), 891–897.

42.

Ito

, Kudo

(1994). Reinnervation by axon collaterals from single facial motoneurons to multiple muscle targets following axotomy in the adult guinea pig. Acta Anatomica (Basel), 151(2), 124–130.

43.

Jergovic<

<, D.

, Stal

, Lidman

, Lindvall

, Hildebrand

(2001). Changes in a rat facial muscle after facial nerve injury and repair. Muscle & Nerve, 24(9), 1202–1212.

44.

Johnson

E.O.

, Soucacos

P.N.

(2008). Nerve repair: Experimental and clinical evaluation of biodegradable artificial nerve guides. (Suppl) S30–36.

45.

Kerrebijn

J.D.

, Freeman

J.L.

(1998). Facial nerve reconstruction: Outcome and failures. The Journal of Otolaryngology, 27(4), 183–186.

46.

Kitahara

A.K.

, Suzuki

, Qi

, Nishimura

, Suzuki

, Kiyotani

, Takimoto

, Nakamura

, Shimizu

, Endo

(1999). Facial nerve repair using a collagen conduit in cats. Scandinavian Journal of Plastic and Reconstructive Surgery and Hand Surgery, 33(2), 187–193.

47.

Klein

B.G.

, Rhoades

R.W.

(1985). Representation of whisker follicle intrinsic musculature in the facial motor nucleus of the rat. The Journal of Comparative Neurology, 232(1), 55–69.

48.

Komisaruk

B.R.

(1970). Synchrony between limbic system theta activity and rhythmical behavior in rats. Journal of Comparative and Physiological Psychology, 70(3), 482–492.

49.

Liu

, Mi

, Zhang

, Yan

, Wang

, Tan

, Yuan

, Yang

, Gu

, Hu

(2018). Tubulation repair mitigates misdirection of regenerating motor axons across a sciatic nerve gap in rats. Scientific Reports, 8(1), 3443.

50.

Lundborg

, Dahlin

L.B.

, Danielsen

, Nachemson

A.K.

(1986). Tissue specificity in nerve regeneration. Scandinavian Journal of Plastic and Reconstructive Surgery, 20(3), 279–283.

51.

Lundborg

, Dahlin

, Danielsen

, Zhao

(1994). Trophism, tropism, and specificity in nerve regeneration. Journal of Reconstructive Microsurgery, 10(5), 345–354.

52.

Lundborg

, Rosén

, Dahlin

, Danielsen

, Holmberg

(1997). Tubular versus conventional repair of median and ulnar nerves in the human forearm: Early results from a prospective, randomized, clinical study. The Journal of Hand Surgery, 22(1), 99–106.

53.

Mackinnon

S.E.

, Dellon

A.L.

, Lundborg

, Hudson

A.R.

, Hunter

D.A.

(1986). A study of neurotrophism in a primate model. The Journal of Hand Surgery, 11(6), 888–894.

54.

Mackinnon

S.E.

, Dellon

A.L.

, O’Brien

J.P.

(1991). Changes in nerve fiber numbers distal to a nerve repair in the rat sciatic nerve model. Muscle & Nerve, 14(11), 1116–1122.

55.

Magill

C.K.

, Moore

A.M.

, Borschel

G.H.

, Mackinnon

S.E.

(2010). A new model for facial nerve research: The novel transgenic Thy1-GFP rat. Archives of Facial Plastic Surgery, 12(5), 315–320.

56.

Ozsoy

, Hizay

, Demirel

B.M.

, Ozsoy

, Bilmen Sarikcioglu

, Turhan

, Sarikcioglu

(2011). The hypoglossal-facial nerve repair as a method to improve recovery of motor function after facial nerve injury. Annals of Anatomy, 193(4), 304–313.

57.

Politis

M.J.

(1985). Specificity in mammalian peripheral nerve regeneration at the level of the nerve trunk. Brain Research, 328(2), 271–276.

58.

Politis

M.J.

, Ederle

, Spencer

P.S.

(1982). Tropism in nerve regeneration in vivo. Attraction of regenerating axons by diffusible factors derived from cells in distal nerve stumps of transected peripheral nerves. Brain Research, 253(1-2), 1–12.

59.

Rich

M.M.

, Lichtman

J.W.

(1989). In vivo visualization of pre- and postsynaptic changes during synapse elimination in reinnervated mouse muscle. The Journal of Neuroscience, 9(5), 1781–1805.

60.

Rosenberg

A.F.

, Isaacman-Beck

, Franzini-Armstrong

, Granato

(2014). Schwann cells and deleted in colorectal carcinoma direct regenerating motor axons towards their original path. The Journal of Neuroscience, 34(44), 14668–14681.

61.

Schröder

J.M.

(1968a). Die Hyperneurotisation Büngnerscher Bänder bei der experimentellen Isoniazid-Neuropathie: Phasenkontrast- und elektronenmikroskopische Untersuchungen. Virchows Archiv Abteilung, 1(1), 131–156.

62.

Seckel

B.R.

, Ryan

S.E.

, Gagne

R.G.

, Chiu

T.H.

, Watkins

Jr. (1986). Target-specific nerve regeneration through a nerve guide in the rat. Plastic and Reconstructive Surgery, 78(6), 793–800.

63.

Semba

, Egger

M.D.

(1986). The facial “motor” nerve of the rat: Control of vibrissal movement and examination of motor and sensory components. The Journal of Comparative Neurology, 247(2), 144–158.

64.

Semba

, Szechtman

, Komisaruk

B.R.

(1980). Synchrony among rhythmical facial tremor, neocortical ‘alpha’ waves, and thalamic non-sensory neuronal bursts in intact awake rats. Brain Research, 195(2), 281–298.

65.

Skouras

, Ozsoy

, Sarikcioglu

, Angelov

D.N.

(2011). Intrinsic and therapeutic factors determining the recovery of motor function after peripheral nerve transection. Annals of Anatomy, 193(4), 286–303.

66.

Son

Y.J.

, Trachtenberg

J.T.

, Thompson

W.J.

(1996). Schwann cells induce and guide sprouting and reinnervation of neuromuscular junctions. Trends in Neurosciences, 19(7), 280–285.

67.

Spector

J.G.

, Lee

, Derby

, Roufa

D.G.

(1995). Comparison of rabbit facial nerve regeneration in nerve growth factor-containing silicone tubes to that in autologous neural grafts. Annals of Otology, Rhinology, and Laryngology, 104(11), 875–885.

68.

Sun

, Zhang

, Chen

, Li

, Hu

(2009). Three-dimensional reconstruction and visualization of the median nerve from serial tissue sections. Microsurgery, 29(7), 573–577.

69.

Tam

S.L.

, Gordon

(2003). Mechanisms controlling axonal sprouting at the neuromuscular junction. Journal of Neurocytology, 32(5-8), 961–974.

70.

Tomita

, Kubo

, Matsuda

, Yano

, Tohyama

, Hosokawa

(2007). Myelin-associated glycoprotein reduces axonal branching and enhances functional recovery after sciatic nerve transection in rats. Glia, 55(14), 1498–1507.

71.

Tomov

T.L.

, Guntinas-Lichius

, Grosheva

, Streppel

, Schraermeyer

, Neiss

W.F.

, Angelov

D.N.

(2002). An example of neural plasticity evoked by putative behavioral demand and early use of vibrissal hairs after facial nerve transection. Experimental Neurology, 178(2), 207–218.

72.

Trojan

D.A.

, Gendron

, Cashman

N.R.

(1991). Electrophysiology and electrodiagnosis of the post-polio motor unit. Orthopedics, 14(12), 1353–1361.

73.

Weiss

(1944). The technology of nerve regeneration: A review. Sutureless tubulation and related methods of nerve repair. Journal of Neurosurgery, 1(6), 400–450.

74.

Weiss

, Hoag

(1946). Competitive reinnervation of rat muscles by their own and foreign nerves. Journal of Neurophysiology, 9, 413–418.

75.

Weiss

, Taylor

A.C.

(1943). Histomechanical analysis of nerve reunion in the rat after tubular splicing. Archives of Surgery, 47(5), 419–447.

76.

Weiss

, Taylor

(1944). Further experimental evidence against “neurotropism” in nerve regeneration. Journal of Experimental Zoology, 95(2), 233–257.

77.

Welker

W.I.

(1964). Analysis of sniffing of the albino rat. Behaviour, 22(3), 223–244.

78.

Williams

L.R.

(1987). Rat aorta isografts possess nerve regeneration-promoting properties in silicone Y chambers. Experimental Neurology, 97(3), 555–563.

79.

Wineski

L.E.

(1985). Facial morphology and vibrissal movement in the golden hamster. Journal of Morphology, 183(2), 199–217.