Reinnervation of transplanted pacinian corpuscles by ventral root axons: ultrastructure of the regenerated nerve terminals

Variation in nerve autograft length increases fibre misdirection and decreases pruning effectiveness. An experimental study in the rat median nerve

OBJECTIVES

In the clinical set, autologus nerve grafts are the current option for reconstruction of nerve tissue losses. The length of the nerve graft has been suggested to affect outcomes. Experiments were performed in the rat in order to test this assumption and to detect a possible mechanism to explain differences in recovery.

METHODS

The rat median nerve was repaired by ulnar nerve grafts of different lengths. Rats were evaluated for 12 months by behavioural assessment and histological studies, including ATPase myofibrillary histochemistry and retrograde neuronal labelling.

RESULTS

It was demonstrated that graft length interferes in behavioural functional recovery that here correlates to muscle weight recovery. Short nerve grafts recovered faster and better. Reinnervation was not specific either at the trunk level or in the muscle itself. The normal mosaic pattern of Type I muscle fibres was never restored and their number remained largely augmented. An increment in the number of motor fibres was observed after the nerve grafting in a predominantly sensory branch in all groups. This increment was more pronounced in the long graft group. In the postoperative period, about a 20% reduction in the number of misdirected motor fibres occurred in the short nerve graft group only.

CONCLUSION

Variation in the length of nerve grafts interferes in behavioural recovery and increases motor fibres misdirection. Early recovery onset was related to a better outcome, which occurs in the short graft group.

Concepts of nerve regeneration and repair applied to brachial plexus reconstruction

Brachial plexus injury is a serious condition that usually affects young adults. Progress in brachial plexus repair is intimately related to peripheral nerve surgery, and depends on clinical and experimental studies. We review the rat brachial plexus as an experimental model, together with its behavioral evaluation. Techniques to repair nerves, such as neurolysis, nerve coaptation, nerve grafting, nerve transfer, fascicular transfer, direct muscle neurotization, and end-to-side neurorraphy, are discussed in light of the authors' experimental studies. Intradural repair of the brachial plexus by graft implants into the spinal cord and motor rootlet transfer offer new possibilities in brachial plexus reconstruction. The clinical experience of intradural repair is presented. Surgical planning in root rupture or avulsion is proposed. In total avulsion, the authors are in favor of the reconstruction of thoraco-brachial and abdomino-antebrachial grasping, and on the transfer of the brachialis muscle to the wrist extensors if it is reinnervated. Surgical treatment of painful conditions and new drugs are also discussed.

Functional recovery improvement is related to aberrant reinnervation trimming. A comparative study using fresh or predegenerated nerve grafts

Experimentally predegenerated nerve grafts have been demonstrated to improve recovery. In a 12 month-long study, we compared the degree of recovery of conventional and predegenerated grafts in rat median nerve repair. To induce predegeneration the ulnar donor nerve was crushed and grafting to the median nerve was performed 2 weeks later. The day of recovery and the improvement of finger flexion strength were studied by the grasping test. At 3, 6, 9, and 12 months after surgery retrograde labeling studies and flexor carpi radialis muscle ATPase histochemistry were performed. In the predegenerated grafts, the recovery of finger flexion occurred 19.6+/-1.5 days after surgery and was significantly faster than that in the conventional group. Twelve months after surgery, a similar rate of 85% of grasping strength recovery in relation to the normal control rats was demonstrated for the conventional and predegenerated grafts. After grafting, a larger number of motoneurons, compared to the normal controls, were retrograde labeled in the median nerve. This surplus of retrograde labeled motoneurons in the predominantly sensory branch of the median nerve represented misdirected motor fibers. There was a time-related decrease in the number of labeled motoneurons, which correlated to functional grasping strength recovery. Muscle reinnervation induced a predominance of type I over type II muscle fibers. Forty percent of type I fibers were grouped indicating that collateral sprouting plays a prominent role during muscle reinnervation. Regeneration in predegenerated grafts was faster but the final rate of recovery was similar to conventional grafts.

The course of aberrant reinnervation following nerve repair with fresh or denatured muscle autografts

Denatured muscle grafts obtained by freeze thawing have been proposed to replace losses in the peripheral nerves. In the present report, we compare the performance of such grafts with fresh grafts in the rat median nerve. A long-term effect of muscle interposition on reinnervation was studied by behavioral assessment, muscle ATPase histochemistry, and retrograde labeling of motoneurons. There was no difference in grasping strength recovery between fresh and denatured 10-mm-long muscle grafts. Recovery was delayed and incomplete. Twelve months after surgery, only 50% of the normal grasping strength was attained. Grasping recovery was not observed in the 20-mm-long graft groups. Pathway reinnervation was non-specific with a huge amount of motor fiber misdirection. A decrease in the number of misdirected motor fibers occurred with time and activity recovery. Muscle reinnervation was not specific with disturbance of the mosaic pattern and type-grouping formation. Preference of type I axons for reinnervating deeper zones was observed. Type I aberrant reinnervation was demonstrated in the muscle periphery. The mosaic distribution of type I and II muscle fibers was not stable, and readjustments were observed with time, correlating with grasping improvement. During grasping strength recovery, there was a decrease in the number of type I fibers peripherally located and an increase of those deeply disposed. A time- and activity-related recovery was associated with readjustment in the pathways and muscle fiber rearrangement. We suggest that muscle activity generates specificity.

Protein synthesis in axons and terminals: significance for maintenance, plasticity and regulation of phenotype. With a critique of slow transport theory

This article focuses on local protein synthesis as a basis for maintaining axoplasmic mass, and expression of plasticity in axons and terminals. Recent evidence of discrete ribosomal domains, subjacent to the axolemma, which are distributed at intermittent intervals along axons, are described. Studies of locally synthesized proteins, and proteins encoded by RNA transcripts in axons indicate that the latter comprise constituents of the so-called slow transport rate groups. A comprehensive review and analysis of published data on synaptosomes and identified presynaptic terminals warrants the conclusion that a cytoribosomal machinery is present, and that protein synthesis could play a role in long-term changes of modifiable synapses. The concept that all axonal proteins are supplied by slow transport after synthesis in the perikaryon is challenged because the underlying assumptions of the model are discordant with known metabolic principles. The flawed slow transport model is supplanted by a metabolic model that is supported by evidence of local synthesis and turnover of proteins in axons. A comparison of the relative strengths of the two models shows that, unlike the local synthesis model, the slow transport model fails as a credible theoretical construct to account for axons and terminals as we know them. Evidence for a dynamic anatomy of axons is presented. It is proposed that a distributed "sprouting program," which governs local plasticity of axons, is regulated by environmental cues, and ultimately depends on local synthesis. In this respect, nerve regeneration is treated as a special case of the sprouting program. The term merotrophism is proposed to denote a class of phenomena, in which regional phenotype changes are regulated locally without specific involvement of the neuronal nucleus.

Degeneration and regeneration of cutaneous sensory nerve formations

Auxiliary structures of the cutaneous sensory nerve formations (SNF) are dependent on sensory innervation during their critical period of development. Denervation of mature cutaneous corpuscles results in survival of the terminal Schwann cells and the capsular structures which are probably responsible for successful reinnervation of the cutaneous SNF. In addition, the basal lamina tubes of Schwann cells are connected with the terminal Schwann cells and play an important role in the guidance of regrowing axons to their original targets. Long-lasting denervation causes atrophic changes of the terminal Schwann cells and alterations of their molecular equipment. These atrophic changes in the terminal Schwann cells may be responsible for erroneous reinnervation of cutaneous SNF. A population of the cutaneous Merkel cells surviving denervation may also serve as targets for regrowing sensory axons. The basal laminae of terminal Schwann cells are produced under control of the sensory terminals during maturation of cutaneous SNF. In adult animals, the basal laminae are capable of stimulating differentiation of migrated Schwann cells to the terminal Schwann cells without the presence of the sensory terminals. Nonspecific cholinesterase (nChE) is secreted by the terminal Schwann cells and is attached to their extracellular matrix. The synthesis of these molecules in adult animals is not influenced by the sensory terminals. However, the presence of nChE molecules is associated with living terminal Schwann cells. Fetal orthotopically grafted dorsal root ganglion (DRG) neurons have the ability to reinnervate cutaneous SNF of adult hosts. When cutaneous areas are denervated, axons from adjacent sensory nerves may extend collateral branches into this area. The capacity for such extension is dependent on: (1) type of sensory nerve ending, C and A delta fibers having significantly greater capacity than sensory axons of larger caliber; (2) age of the animal, immature animals generally showing a greater capacity for collateral sprouting; (3) the state of the adjacent axons, those already in a growth mode being more capable than "resting" ones; and (4) the regional and mechanical conditions at the site of denervation, hindpaw skin being much less extensively reinnervated by collateral fibers than that of the trunk.

Reinnervation of Pacinian corpuscles by CNS axons after transplantation to the dorsal column: incidence and ultrastructure

We have investigated the capacity of injured axons in the spinal dorsal columns of young adult rats to reinnervate grafted Pacinian corpuscles. A branch of the hindlimb interosseous nerve with a group of crural Pacinian corpuscles attached to it was autotransplanted to the surface of the spinal cord and the nerve stump was implanted into the dorsal column. Two to three months later 16 grafts were removed for examination by light and electron microscopy. By 3 months after transplantation almost all Schwann cell columns of the grafted nerve branch were occupied by regenerated myelinated and unmyelinated axons. Of 41 corpuscles examined by electron microscopy 24 were reinnervated by 1-3 myelinated fibres which gave rise to multiple terminals in the inner core. The remaining corpuscles appeared to be denervated. Only two of the reinnervated corpuscles contained regenerated endings which reiterated the distinct ultrastructure of normal presynaptic terminals of CNS axons, characterized by clusters of lucent vesicles and paramembranous densities. All other corpuscles were reinnervated by terminals which resembled peripheral mechanosensory endings as they contained mitochondria and very few vesicles. One such corpuscle was coinnervated by small terminals filled with large dense cored vesicles. We assume that the majority of grafted Pacinian corpuscles have been reinnervated by dorsal column axons and that the regenerated terminals with the ultrastructure of peripheral mechanosensory endings derive from central axons of primary sensory neurons, which are apparently capable of constructing mechanosensory-like terminals in response to signals from the Pacinian corpuscles. The vesicle-filled endings are probably formed by second order sensory neurons, corticospinal neurons and small peptidergic neurons unable to adjust their terminals to the new target.

Location and completeness of reinnervation by two types of neurons at a single target: the feline muscle spindle

Muscle spindles from the tenuissimus muscle of the cat were examined microscopically to assess the precision and completeness of reinnervation of intrafusal muscle fibers by efferent and afferent neurons. Positions of motor and sensory nerve terminals were charted relative to the cross-sectional area enclosed by the outer capsule of the spindle. Profiles of nerve endings were measured for normally innervated and reinnervated spindles. The tenuissimus was deprived of innervation by freezing its nerve, sometimes in conjunction with either spinal ganglion removal or ventral rhizotomy. Sensory and motor terminals occupied separate locales along the length of normal muscle spindles. Nerve terminals of efferent and afferent neurons were located in appropriate positions along the length of spindles when axons of both types of neurons regrew together and when either category of axon regenerated alone. Precise reinnervation of muscle spindles occurred in spite of a diminished diameter of intrafusal fibers. Repopulation of the spindle with motor endings was less complete than that by sensory endings, based on the proportion and size of the regenerated terminals. We conclude that under optimal conditions for axonal regrowth, efferent and afferent neurons reinnervate their respective regions along intrafusal muscle fibers but motor lags sensory reinnervation within the spindle. The mechanism by which positional specificity happens during reinnervation of intrafusal fibers requires neither an interaction between terminals of the two types of neurons nor target cells of normal bulk.