Michael Streppel

Factors limiting motor recovery after facial nerve transection in the rat: combined structural and functional analyses

It is believed that a major reason for the poor functional recovery after peripheral nerve lesion is collateral branching and regrowth of axons to incorrect muscles. Using a facial nerve injury protocol in rats, we previously identified a novel and clinically feasible approach to combat axonal misguidance – the application of neutralizing antibodies against neurotrophic factors to the injured nerve. Here, we investigated whether reduced collateral branching at the lesion site leads to better functional recovery. Treatment of rats with antibodies against nerve growth factor, brain‐derived neurotrophic factor, fibroblast growth factor, insulin‐like neurotrophic factor I, ciliary neurotrophic factor or glial cell line‐derived neurotrophic factor increased the precision of reinnervation, as evaluated by multiple retrograde labelling of motoneurons, more than two‐fold as compared with control animals. However, biometric analysis of vibrissae movements did not show positive effects on functional recovery, suggesting that polyneuronal reinnervation – rather than collateral branching – may be the critical limiting factor. In support of this hypothesis, we found that motor end‐plates with morphological signs of multiple innervation were much more frequent in reinnervated muscles of rats that did not recover after injury (51% of all end‐plates) than in animals with good functional performance (10%). Because polyneuronal innervation of muscle fibres is activity‐dependent and can be manipulated, the present findings raise hopes that clinically feasible and effective therapies could be soon designed and tested.

Focal application of neutralizing antibodies to soluble neurotrophic factors reduces collateral axonal branching after peripheral nerve lesion

A major reason for the insufficient recovery of function after motor nerve injury are the numerous axonal branches which often re‐innervate muscles with completely different functions. We hypothesized that a neutralization of diffusable neurotrophic factors at the lesion site in rats could reduce the branching of transected axons. Following analysis of local protein expression by immunocytochemistry and by in situ hybridization, we transected the facial nerve trunk of adult rats and inserted both ends into a silicon tube containing (i) collagen gel with neutralizing concentrations of antibodies to NGF, BDNF, bFGF, IGF‐I, CNTF and GDNF; (ii) five‐fold higher concentrations of the antibodies and (iii) combination of antibodies. Two months later, retrograde labelling was used to estimate the portion of motoneurons the axons of which had branched and projected into three major branches of the facial trunk. After control entubulation in collagen gel containing non‐immune mouse IgG 85% of all motoneurons projecting along the zygomatic branch sprouted and sent at least one twin axon to the buccal and/or marginal‐mandibular branches of the facial nerve. Neutralizing concentrations of anti‐NGF, anti‐BDNF and anti‐IGF‐I significantly reduced sprouting. The most pronounced effect was achieved after application of anti‐BDNF, which reduced the portion of branched neurons to 18%. All effects after a single application of antibodies were concentration‐dependent and superior to those observed after combined treatment. This first report on improved quality of reinnervation by antibody‐therapy implies that, in rats, the post‐transectional collateral axonal branching can be reduced without obvious harmful effects on neuronal survival and axonal elongation.

An example of neural plasticity evoked by putative behavioral demand and early use of vibrissal hairs after facial nerve transection.

Abnormally associated movements inevitably occur after surgical repair of the facial nerve. The reason for this postparalytic syndrome is poor navigation of regrowing axons. Despite the valuable functional advantage provided by the easily detected movement of vibrissae in rats, the major investigative tools for establishing the degree of misdirected reinnervation are still electrophysiologic recordings and retrograde tracing. In the present study we complemented data from pre- and postoperative retrograde labeling (FluoroGold, Fast Blue, DiI) of facial motoneurons with an evaluation of whisker movements. Using a video-based motion analysis system, we compared the recovery of vibrissae motor performance in visually normal and blind rats of the Sprague-Dawley strain. The analysis of whisker movement after facial nerve surgery revealed a striking discrepancy between morphologic and functional estimates. Whereas retrograde labeling displayed poor accuracy of target reinnervation and supernumerary axonal branching in both groups, the video-based motion analysis showed a perfect recovery of vibrissae movements in the blind rats. Attributing the complete recovery of whisker movement in the blind rats to an extraordinary plasticity of the facial motoneurons induced by putative behavioral demand and forced overuse, we conclude that the video-based analysis of whisker movement is a valuable tool for studying the progress in functional recovery.

The axotomy-induced neuropeptides galanin and pituitary adenylate cyclase-activating peptide promote axonal sprouting of primary afferent and cranial motor neurones

The neuropeptides galanin and pituitary adenylate cyclase‐activating peptide (PACAP) are markedly up‐regulated in response to peripheral nerve lesion. Both peptides are involved in neuronal differentiation and neurite outgrowth during development. In this study, we investigated the effects of galanin and PACAP on axonal elongation and sprouting by adult rat sensory neurones in vitro and facial motor neurones in vivo. Dissociated rat dorsal root ganglion neurones were plated on laminin substrate and analysed morphometrically. Both the mean axonal length and the number of branch points significantly increased in the presence of galanin or PACAP (2–5 µm). Effects on axonal collateralization were investigated in the rat facial nerve lesion model by direct application of the peptides to collagen‐filled conduits entubulating the transected facial nerve stumps. Triple retrograde labelling of brainstem neurones confirmed that the peptides potently induce axonal sprouting of cranial motor neurones. The number of neurones regenerating into identified rami of the facial nerve increased up to fivefold. Biometrical analysis of whisking behaviour revealed that galanin and PACAP impaired the functional outcome when compared with vehicle‐treated animals 8 weeks after surgery. In conclusion, although galanin and PACAP have been established as neurotrophic molecules with respect to axonal development and regeneration, their potential as treatments for peripheral nerve lesions appears limited because of the extensive stimulation of collateral axon branching. These branches are misrouted towards incorrect muscles and cause impairment in their coordinated activity.

Contralateral trigeminal nerve lesion reduces polyneuronal muscle innervation after facial nerve repair in rats

Functional recovery after facial nerve surgery is poor. Axotomized motoneurons (hyperexcitable upon intracellular current injections, but unable to discharge upon afferent stimulation) outgrow supernumerary branches which are misrouted towards improper muscles. We hypothesized that alterations in the trigeminal input to axotomized electrophysiologically silent facial motoneurons might improve specificity of reinnervation. To test this we compared, in the rat, behavioural, electrophysiological, and morphological parameters after transection and suture of the buccal facial nerve (buccal–buccal anastomosis, BBA) with those after BBA plus excision of the ipsi‐ or contralateral infraorbital nerve (ION). After BBA, the mystacial vibrissae dropped and remained motionless until 18–21 days post operation (days PO). After BBA plus ipsilateral ION excision, there was no recovery of vibrissae whisking at all. Following BBA plus contralateral ION excision, full restoration of whisking occurred at 7–10 days PO. Electromyography of whiskerpad muscles showed normal waveform and amplitude was also most rapidly restored after BBA plus contralateral ION excision. Neuron counts after retrograde tracing showed that the intact buccal nerve contained axons of the superior (91%) and inferior (9%) buccolabial nerves. After BBA, the superior nerve comprised 56%, the inferior 21%, and 23% of the motoneurons projected within both nerves. After BBA plus ipsilateral ION excision, misdirection worsened and values changed to 48, 39 and 13%, respectively. After BBA plus contralateral ION excision, portions improved to 69, 23 and 8%. We conclude that, by reducing the redundant axon branching, lesion of contralateral ION provides the best conditions for recovery of vibrissae rhythmical whisking after reconstructive surgery on the facial nerve.

Transoral CO2 laser for surgical management of glottic carcinoma in situ.

Objectives/Hypothesis In carcinoma in situ (CIS) tumors malignant cells have not penetrated the basement membrane and therefore have no metastatic potential. Treatment strategies of CIS are topics of ongoing discussion. The aim of this study was to evaluate long‐term results of CO2 laser therapy in laryngeal CIS.

Hypoglossal and reticular interneurons involved in oro-facial coordination in the rat.

Chewing, swallowing, breathing, and vocalization in mammals require precise coordination of tongue movements with concomitant activities of the mimetic muscles. The neuroanatomic basis for this oro‐facial coordination is not yet fully understood. After the stereotaxic microinjection of retrograde and anterograde neuronal tracers (biotin‐dextran, Fluoro‐Ruby, Fluoro‐Emerald, and Fluoro‐Gold) into the facial and hypoglossal nuclei of the rat, we report here a direct bilateral projection of hypoglossal internuclear interneurons onto facial motoneurons. We also confirm the existence of a small pool of neurons in the dorsal part of the brainstem reticular formation that project ipsilaterally to both facial and hypoglossal nuclei. For precise tracer injections, both motor nuclei were located and identified by the electrical antidromic activation of their constituent motoneurons. Injections of retrograde tracers into the facial nucleus consistently labeled neurons in the hypoglossal nucleus. These neurons prevalently lay in the ipsilateral side, were small in size, and, like classic intrinsic hypoglossal local‐circuit interneurons, had several thin dendrites. Reverse experiments — injections of anterograde tracers into the hypoglossal nucleus — labeled fine varicose nerve fiber terminals in the facial nucleus. These fiber terminals were concentrated in the intermediate subdivision of the facial nucleus, with a strong ipsilateral prevalence. Double injections of different tracers into the facial and the hypoglossal nuclei revealed a small, but constant, number of double‐labeled neurons located predominantly ipsilateral in the caudal brainstem reticular formation. Hypoglossal internuclear interneurons projecting to the facial nucleus, as well as those neurons of the parvocellular reticular formation that project to both facial and hypoglossal nuclei, could be involved in oro‐facial coordination. J. Comp. Neurol. 433:364–379, 2001.

ED2‐positive perivascular cells act as neuronophages during delayed neuronal loss in the facial nucleus of the rat

Injection of Fluoro‐Gold (FG) into the whisker pad of rats yields a stable retrograde labeling of facial motoneurons. After removal of 10 mm from the facial nerve the microglia phagocytose the FG‐prelabeled dead neurons and assume the label. A subsequent brightfield immunostaining of the sections with HRP‐DAB as end‐product fully quenches the fluorescence of FG from all specifically stained structures (immunoquenching). Combining FG‐labeling of neuronophages with immunoquenching, we recently described a population of enigmatic fluorescent cells, found in immediate vicinity to the motoneurons after the general neuronofugal migration of microglia. As the fluorescence of these cells was not quenched after a triple immunostaining with anti neuron‐specific enolase, anti‐GFAP, and OX‐42 (quenching all fluorescence from neurons and glia), they seemed to represent a new, immunologically not identified neuronophage. Now we have further characterized this cell type. Following triple immunostaining, we tested a broad panel of mabs (OX‐33, OX‐19, OX‐18, OX‐6, R73, ED1, and ED2) to stain, quench fluorescence, and thus immunotype the unknown phagocytes. Only the mab ED2, the classical marker for perivascular cells, specifically stained the small round neuronophages. This surprising migration of perivascular cells toward decaying neurons was additionally tested and confirmed by intracerebroventricular application of FG prior to resection of the facial nerve Providing evidence for neuronophagia by ED2‐positive cells, our results strongly support the hypothesis that the latter are the APC (antigen presenting cells) of the CNS.

Manually-stimulated recovery of motor function after facial nerve injury requires intact sensory input

We have recently shown in rat that daily manual stimulation (MS) of vibrissal muscles promotes recovery of whisking and reduces polyinnervation of muscle fibers following repair of the facial nerve (facial-facial anastomosis, FFA). Here, we examined whether these positive effects were: (1) correlated with alterations of the afferent connections of regenerated facial motoneurons, and (2) whether they were achieved by enhanced sensory input through the intact trigeminal nerve. First, we quantified the extent of total synaptic input to motoneurons in the facial nucleus using synaptophysin immunocytochemistry following FFA with and without subsequent MS. We found that, without MS, this input was reduced compared to intact animals. The number of synaptophysin-positive terminals returned to normal values following MS. Thus, MS appears to counteract the deafferentation of regenerated facial motoneurons. Second, we performed FFA and, in addition, eliminated the trigeminal sensory input to facial motoneurons by extirpation of the ipsilateral infraorbital nerve (IONex). In this paradigm, without MS, vibrissal motor performance and pattern of end-plate reinnervation were as aberrant as after FFA without MS. MS did not influence the reinnervation pattern after IONex and functional recovery was even worse than after IONex without MS. Thus, when the sensory system is intact, MS restores normal vibrissal function and reduces the degree of polyinnervation. When afferent inputs are abolished, these effects are eliminated or even reversed. We conclude that rehabilitation strategies must be carefully designed to take into account the extent of motor and/or sensory damage.

Delayed Rat Facial Nerve Repair Leads to Accelerated and Enhanced Muscle Reinnervation with Reduced Collateral Axonal Sprouting during a Definite Denervation Period Using a Cross-Anastomosis Paradigm

To establish the influence of prolonged denervation on the recovery of a motor nerve, the rat facial nerve was transected and denervated for 0 to 224 days. Then, the freshly transected hypoglossal nerve was sutured to the predegenerated facial nerve (hypoglossal-facial nerve anastomosis, HFA). Using this nerve cross-anastomosis paradigm we analyzed the nerve regeneration and muscle reinnervation 7 to 112 days post-suture operation (DPSO). After HRP injection into the whiskerpad 931+/-27 hypoglossal neurons were labeled at 112 DPSO after immediate HFA. Following 14 to 112 days denervation the number of labeled neurons increased to 138% (14 days delay), 154% (56 days), and 145% (112 days). In contrast, the reinnervation was poorer after 7 days denervation with the number of neurons increasing to 84%, and after long-term denervation of 224 days the number of neurons increased to 81%. The increase in amplitude of evoked electromyography wave after nerve suture correlated with the number of labeled neurons. After immediate HFA each regenerated motoneuron established on average 5.1 myelinated sprouts at 112 DPSO. The number of sprouts remained constant after delayed suture of 14 to 112 days, whereas the slower reinnervation after 7 or 224 days delay was accompanied by a massive sprouting of 9.1 or 8.1, respectively, sprouts per neuron. The muscles showed recovery after any denervation time. The muscle cross-sectional area continuously decreased with longer denervation time. This decrease was only significant after 224 days denervation (67% of the normal value). We conclude that motor nerve reconstruction achieves better functional results after a definite period of denervation when using a nerve cross-anastomosis paradigm.