Cooperative roles of BDNF expression in neurons and Schwann cells are modulated by exercise to facilitate nerve regeneration

After peripheral nerve injury, neurotrophins play a key role in the regeneration of damaged axons that can be augmented by exercise, although the distinct roles played by neurons and Schwann cells are unclear. In this study, we evaluated the requirement for the neurotrophin, brain-derived neurotrophic factor (BDNF), in neurons and Schwann cells for the regeneration of peripheral axons after injury. Common fibular or tibial nerves in thy-1-YFP-H mice were cut bilaterally and repaired using a graft of the same nerve from transgenic mice lacking BDNF in Schwann cells (BDNF(-/-)) or wild-type mice (WT). Two weeks postrepair, axonal regeneration into BDNF(-/-) grafts was markedly less than WT grafts, emphasizing the importance of Schwann cell BDNF. Nerve regeneration was enhanced by treadmill training posttransection, regardless of the BDNF content of the nerve graft. We further tested the hypothesis that training-induced increases in BDNF in neurons allow regenerating axons to overcome a lack of BDNF expression in cells in the pathway through which they regenerate. Nerves in mice lacking BDNF in YFP(+) neurons (SLICK) were cut and repaired with BDNF(-/-) and WT nerves. SLICK axons lacking BDNF did not regenerate into grafts lacking Schwann cell BDNF. Treadmill training could not rescue the regeneration into BDNF(-/-) grafts if the neurons also lacked BDNF. Both Schwann cell- and neuron-derived BDNF are thus important for axon regeneration in cut peripheral nerves.     

Neurotrophin-4/5 is required for the early growth of regenerating axons in peripheral nerves

The requirement of the trkB ligand, neurotrophin-4/5 (NT-4/5), for the growth of regenerating axons in the peripheral nervous system (PNS) is not well established. We studied regenerating axon growth in transected peripheral nerves of thy-1-YFP-H mice that had been repaired using allografts obtained from brain-derived neurotrophic factor (BDNF) or NT-4/5 knockout mice. Lengths of profiles of YFP+ axons measured in these grafts were compared with those measured in grafts obtained from wild-type donors. When compared with axon profiles measured in grafts from wild-type donors, axon profile lengths measured in grafts from homozygous (NT-4/5(-/-)) or heterozygous (NT-4/5(+/-)) mice were significantly shorter. In contrast, the lengths of axon profiles measured in grafts from BDNF(+/-) mice were not significantly different from those measured in grafts from wild-type mice. A reduced amount of BDNF, but not NT-4/5, is sufficient to promote the elongation of regenerating axons in the PNS. When grafts from wild-type or NT-4/5(-/-) mice were treated acutely at the time of surgical repair either with exogenous BDNF or NT-4/5, the lengths of axon profiles measured in the grafts were significantly longer than those measured in grafts from untreated wild-type mice. These findings are consistent with a requirement for NT-4/5 from within the pathway used by regenerating axons for the successful growth of those axons in peripheral nerves.     

Treadmill training enhances axon regeneration in injured mouse peripheral nerves without increased loss of topographic specificity

We investigated the extent of misdirection of regenerating axons when that regeneration was enhanced by using treadmill training. Retrograde fluorescent tracers were applied to the cut proximal stumps of the tibial and common fibular nerves 2 or 4 weeks after transection and surgical repair of the mouse sciatic nerve. The spatial locations of retrogradely labeled motoneurons were studied in untreated control mice and in mice receiving 2 weeks of treadmill training, according to either a continuous protocol (10 m/minute, 1 hour/day, 5 days/week) or an interval protocol (20 m/minute for 2 minutes, followed by a 5‐minute rest, repeated four times, 5 days/week). More retrogradely labeled motoneurons were found in both treadmill‐trained groups. The magnitude of this increase was as great as or greater than that found after using other enhancement strategies. In both treadmill‐trained groups, the proportions of motoneurons labeled from tracer applied to the common fibular nerve that were found in spinal cord locations reserved for tibial motoneurons in intact mice were no greater than in untreated control mice and significantly less than those found after electrical stimulation or chondroitinase treatment. Treadmill training in the first 2 weeks following peripheral nerve injury produces a marked enhancement of motor axon regeneration without increasing the propensity of those axons to choose pathways leading to functionally inappropriate targets. J. Comp. Neurol. 517:245–255, 2009. © 2009 Wiley‐Liss, Inc.     

Treadmill training promotes axon regeneration in injured peripheral nerves

Physical activity after spinal cord injury promotes improvements in motor function, but its effects following peripheral nerve injury are less clear. Although axons in peripheral nerves are known to regenerate better than those in the CNS, methods of accelerating regeneration are needed due to the slow overall rate of growth. Therefore we studied the effect of two weeks of treadmill locomotion on the growth of regenerating axons in peripheral nerves following injury. The common fibular nerves of thy-1-YFP-H mice, in which a subset of axons in peripheral nerves express yellow fluorescent protein (YFP), were cut and repaired with allografts from non-fluorescent littermates, and then harvested two weeks later. Mice were divided into groups of low-intensity continuous training (CT, 60 min), low-intensity interval training (IT; one group, 10 reps, 20 min total), and high-intensity IT (three groups, 2, 4, and 10 reps). One repetition consisted of 2 min of running and 5 min of rest. Sixty minutes of CT resulted in the highest exercise volume, whereas 2 reps of IT resulted in the lowest volume of exercise. The lengths of regenerating YFP⁺ axons were measured in images of longitudinal optical sections of nerves. Axon profiles were significantly longer than control in all exercise groups except the low-intensity IT group. In the CT group and the high-intensity IT groups that trained with 4 or 10 repetitions axons were more than twice as long as unexercised controls. The number of intervals did not impact axon elongation. Axon sprouting was enhanced in IT groups but not the CT group. Thus exercise, even in very small quantities, increases axon elongation in injured peripheral nerves whereas continuous exercise resulting in higher volume (total steps) may have no net impact on axon sprouting.     

Neurotrophin-3 deficient Schwann cells impair nerve regeneration

Neurotrophin 3 (NT-3) is an important autocrine factor supporting Schwann cell (SC) survival and differentiation in the absence of axons. Prior studies have failed to define the explicit role of SC versus axon in NT-3 deficiency in relation to nerve regeneration and associated remyelination. In the paradigm we studied, using NT-3 heterozygous (NT3^+/−) knockout mice capable of survival into adult-life, the experimental design provided a model uniquely capable of differentiating SC/axon influences. In these studies we first identified a defect in nerve regeneration characterized by fewer SCs in the regenerating nerve fibers of crushed sciatic nerves of NT3^+/− mice. Subsequent experiments differentiated SC versus axonal influences as the culprit in defective nerve regeneration using sciatic nerve transplant paradigms. Results show an impairment in nerve regeneration in NT3^+/− mice with a retardation of the myelination process, and this defect is associated with decreased SC survival and an increase in the neurofilament packing density of regenerating axons. These observations indicate that NT3^+/− status of the SCs, but not of the axons, is responsible for impaired nerve regeneration and that NT-3 is essential for SC survival in early stages of regeneration-associated myelination in the adult peripheral nerve.     

Axon regeneration in peripheral nerves is enhanced by proteoglycan degradation

Regeneration of axons in the peripheral nervous system is enhanced by the removal of glycosaminoglycan side chains (GAGs) of chondroitin sulfate proteoglycans. However, some axons regenerate poorly despite such treatment, suggesting the existence of additional inhibitors. We compared the effects of enzymatic removal of GAGs from chondroitin sulfate proteoglycans versus two other proteoglycan species, heparan sulfate and keratan sulfate proteoglycans, on the regeneration of peripheral axons. Common fibular (CF) nerves of thy-1-YFP-H mice were cut and repaired using short segments of CF nerves harvested from wild-type littermates and pre-treated with a GAG-degrading enzyme for 1 h prior to nerve repair. Axonal regeneration was assayed by measuring the lengths of profiles of YFP+ axons in optical sections of the grafted nerves 1 week later. Except for grafts treated with keratanase, more and longer axon profiles were encountered in enzyme-treated grafts than in control grafts. Heparinase III treatments induced the greatest number of axons to enter into the graft. The proportions of axon profiles longer than 1000 microm were greater in grafts treated with chondroitinase ABC or heparinase I, but not with either keratanase or heparinase III. More regenerative sprouts were observed after treatment with heparinase I than any other enzymes. Treatment with a mixture of all four enzymes resulted in an enhancement of axon regeneration which was greater than that observed after treatment with any of the enzymes individually. The effects of chondroitinase ABC and heparinase III were correlated with specific GAG degradation. We believe that enzymatic removal of GAGs is especially effective in promoting the ability of regenerating axons to select their pathway in the distal stump (or nerve graft) and, in the case of chondroitinase ABC or heparinase I, it may also promote growth within that pathway.     

Receptor protein tyrosine phosphatase sigma inhibits axon regrowth in the adult injured CNS

Recently, receptor protein tyrosine phosphatase-sigma (RPTPsigma) has been shown to inhibit axon regeneration in injured peripheral nerves. Unlike the peripheral nervous system (PNS), central nervous system (CNS) neurons fail to regenerate their axons after injury or in disease. In order to assess the role of RPTPsigma in CNS regeneration, we used the retinocollicular system of adult mice lacking RPTPsigma to evaluate retinal ganglion cell (RGC) axon regrowth after optic nerve lesion. Quantitative analysis demonstrated a significant increase in the number of RGC axons that crossed the glial scar and extended distally in optic nerves from RPTPsigma (-/-) mice compared to wild-type littermate controls. Although we found that RPTPsigma is expressed by adult RGCs in wild-type mice, the retinas and optic nerves of adult RPTPsigma (-/-) mice showed no histological defects. Furthermore, the time-course of RGC death after nerve lesion was not different between knockout and wild-type animals. Thus, enhanced axon regrowth in the absence of RPTPsigma could not be attributed to developmental defects or increased neuronal survival. Finally, we show constitutively elevated activity of mitogen-activated protein kinase (MAPK) and Akt kinase in adult RPTPsigma (-/-) mice retinas, suggesting that these signaling pathways may contribute to promoting RGC axon regrowth following traumatic nerve injury. Our results support a model in which RPTPsigma inhibits axon regeneration in the adult injured CNS.     

Regeneration of rat optic axons into peripheral nerve grafts

The regenerative response of myelinated axons of the mammalian central nervous system was investigated by inserting peripheral nerve grafts in the vicinity of traumatized rat optic axons. Peroneal nerve grafts were inserted into crushed optic nerves and assessed for the ability to support optic axon regeneration. Regenerating axons of retinal origin were present in the peripheral nerve grafts several weeks later. Horseradish peroxidase studies indicated that retinal ganglion cells were a source for regenerating nerve fibers.     

Prior collateral sprouting enhances elongation rate of sensory axons regenerating through acellular distal segment of a crushed peripheral nerve

Regenerating axons in crushed peripheral nerves grow through their distal nerve segments even in the absence of Schwann cell support, but their elongation rate is reduced by 30%. We examined whether prior exposure of sensory neurons to trophic factors achieved either by collateral sprouting or regeneration after conditioning lesion could enhance subsequent regeneration of their axons after crush, and compensate for loss of cell support. Collateral sprouting of the peroneal cutaneous sensory axons in the rat was evoked by transection of adjacent peripheral nerves in the hind leg. The segment of the peroneal nerve distal to the crush was made acellular by repeated freezing. Sensory axon elongation rate during regeneration was measured by the nerve pinch test. Prior axonal sprouting for two weeks increased the elongation rate of sensory axons through the acellular distal nerve segment back to normal value observed in control crushed nerves. The number of axons in the acellular distal segment at a fixed distance from the crush site was about 50% greater in sprouting than in control non-sprouting nerves. However, prior sprouting caused no further increase of axon elongation rate in control crushed nerves. Prior collateral sprouting, therefore, could in some respect compensate for loss of cell support in the distal nerve segment after crush lesion. This suggests that loss of cell-produced trophic factors is probably responsible for slower elongation rate through the acellular distal nerve segment. Surprisingly, prior conditioning lesion caused no enhancement of elongation rate of the sensory axons regenerating in the absence of cell support.     

Chemical communication between regenerating motor axons and Schwann cells in the growth pathway

There are receptors on denervated Schwann cells that may respond to the neurotransmitters that are released from growth cones of regenerating motor axons. In order to ascertain whether the interaction of the transmitters and their receptors plays a role during axon regeneration, we investigated whether pharmacological block of the interaction would reduce the number of motoneurons that regenerate their axons after nerve section and surgical repair. Peripheral nerves in the hindlimbs of rats and mice were cut and repaired, and various drugs were applied to the peripheral nerve stump either directly or via mini-osmotic pumps over a 2–4-week period to block the binding of acetylcholine to nicotinic and muscarinic acetylcholine receptors (AChRs: α-bungarotoxin, tubocurarine, atropine and, gallamine) and binding of ATP to P2Y receptors (suramin). In rats, the nicotinic AChR antagonistic drugs and suramin reduced the number of motoneurons that regenerated their axons through the distal nerve stump. In mice, suramin significantly reduced the upregulation of the carbohydrate HNK-1 on the Schwann cells in the distal nerve stump that normally occurs during motor axon regeneration. These data indicate that chemical communication between regenerating axons and Schwann cells during axon regeneration via released neurotransmitters and their receptors may play an important role in axon regeneration.     

N-Propionylmannosamine stimulates axonal elongation in a murine model of sciatic nerve injury

Increasing evidence indicates that sialic acid plays an important role during nerve regeneration. Sialic acids can be modified in vitro as well as in vivo using metabolic oligosaccharide engineering of the N-acyl side chain. N-Propionylmannosamine (ManNProp) increases neurite outgrowth and accelerates the reestablishment of functional synapses in vitro. We investigated the influence of systemic ManNProp application using a specific in vivo mouse model. Using mice expressing axonal fluorescent proteins, we quantified the extension of regenerating axons, the number of regenerating axons, the number of arborising axons and the number of branches per axon 5 days after injury. Sciatic nerves from non-expressing mice were grafted into those expressing yellow fluorescent protein. We began a twice-daily intraperitoneal application of either peracetylated ManNProp (200 mg/kg) or saline solution 5 days before injury, and continued it until nerve harvest (5 days after transection). ManNProp significantly increased the mean distance of axonal regeneration (2.49 mm vs. 1.53 mm; P < 0.005) and the number of arborizing axons (21% vs. 16%; P = 0.008) 5 days after sciatic nerve grafting. ManNProp did not affect the number of regenerating axons or the number of branches per arborizing axon. The biochemical glycoengineering of the N-acyl side chain of sialic acid might be a promising approach for improving peripheral nerve regeneration.     

Neurotrophin-4 delivered by fibrin glue promotes peripheral nerve regeneration

Neurotrophin-4 (NT-4) is a recently identified neurotrophic factor with potential trophic effects on subpopulations of neurons. Little is known about its role in peripheral nerve regeneration following nerve injury. To investigate this, 48 Sprague-Dawley rats underwent left sciatic nerve transection and immediate repair. Fibrin glue mixed with either NT-4 or vehicle (control) was injected around the nerve repair site. Nerve regeneration was assessed both functionally and histomorphometrically. The results showed that the NT-4-treated group had a significant increase compared with the control in the regeneration distance at 5 days. The sciatic function index was significantly greater in the NT-4 group from 40 to 60 days after nerve repair. Morphometric analysis revealed that nerves treated with NT-4 had significant improvement in the number of regenerated axons, axonal diameter, and myelin thickness. These results suggest that NT-4 is a potent factor improving rat sciatic nerve regeneration.     

Bcl-2 deficiency deprives peripheral nerves of neurotrophic activity against injured optic nerve

Optic nerve injury leads to retinal ganglion cell apoptosis, thus preventing fiber regeneration. Peripheral nerve grafts are known to promote survival and regeneration in injured adult mammalian central nervous system, including optic nerve, but the mechanisms of their activity remain unclear. It is likely that they attenuate the apoptotic cascade triggered by axotomy in retinal ganglion cells. The aim of this work was to examine the role of the antiapoptotic gene bcl-2 in the optic nerve regeneration induced by such grafts. Experiments were carried out on bcl-2-deficient and wild-type mice. We have reported previously that predegeneration markedly enhances neurotrophic activity of peripheral nerve grafts, so we applied both predegenerated and non-predegenerated implants to the transected optic nerves. We studied the neurotrophic effects of bcl-2-deficient grafts on wild-type and bcl-2 knock-out optic nerves, as well as wild-type grafts on both strains of mouse optic nerves. After application of fluorescent dye to the end of the graft, we counted the stained retinal ganglion cells. Predegenerated wild-type grafts promoted survival and outgrowth of retinal ganglion cells axons in both types of mice. By contrast, non-predegenerated and predegenerated bcl-2-deficient grafts induced little or no regeneration in the optic nerves. These results indicate that the lack of bcl-2 gene does not deprive retinal ganglion cells of their regenerative potential. At the same time, we found that bcl-2 knock-out dispossesses peripheral nerves of their neurotrophic activity.     

Differences in peripheral nerve degeneration/regeneration between wild-type and neuronal nitric oxide synthase knockout mice

Nitric oxide (NO), a unique biological messenger molecule, is synthesized by three isoforms of the enzyme NO synthase (NOS) and diffuses from the site of production across cellular membranes. A postulated role for NO in degeneration and regeneration of peripheral nerves has been explored in a sciatic nerve model comparing wild-type mice and mice lacking neuronal NOS after transection and microsurgical repair. In NOS knockout mice, regenerative delay was observed, preceded by a decelerated Wallerian degeneration (WD). In the regenerated nerve, pruning of uncontrolled sprouts was disturbed, leading to an enhanced number of axons, whereas remyelination seemed to be less affected. Delayed regeneration was associated with a delayed recovery of sensor and motor function. In such a context, possible NO targets are neurofilaments and myelin sheaths of the interrupted axon, filopodia of the growth cone, newly formed neuromuscular endplates, and Schwann cells in the distal nerve stump. The results presented suggest that 1) local release of NO following peripheral nerve injury is a crucial factor in degeneration/regeneration, 2) success of fiber regeneration in the peripheral nervous system depends on a regular WD, and 3) manipulation of NO supply may offer interesting therapeutic options for treatment of peripheral nerve lesions.     

Changes in myelinated and unmyelinated axon numbers in the proximal parts of rat sural nerves after two types of injury

Counts of myelinated and unmyelinated axon profiles have been made from normal, uninjured rat sural nerves and from nerves injured 6 months earlier in one of two ways. In one group of rats the nerve was simply cut and left to regenerate, leading to the development of a neuroma in continuity, while in the second group the nerve was cut but then ligated as well to prevent regeneration; this led to stump neuroma formation. After nerve transection and regeneration, with subsequent formation of a neuroma in continuity, there was no change in the number of myelinated axon profiles found 25 mm proximal to the old injury site when compared with control, but there was an 18% reduction (P < 0.05) in the number of unmyelinated axon profiles. Immediately proximal to the injury site the picture was similar, with there still being the same number of myelinated axon profiles as in control material but here the reduction in unmyelinated axon numbers was slightly greater at 24% (P < 0.05). In the proximal part of nerves that had been cut and stump neuroma formation induced there was a large increase (33%) in myelinated axon profiles over and above control values (P < 0.001) but the number of unmyelinated profiles was the same as in controls. Closer to the stump neuroma the number of myelinated axon profiles had increased yet further to be 88% (P < 0.001) above control while the number of unmyelinated ones remained no different from control. Our interpretation of these results is that after nerve transection and regeneration there is no loss of peripheral neurons supporting myelinated axons but some loss of those supporting unmyelinated ones. If a cut nerve is prevented from regenerating and a stump neuroma forms, however, a vigorous sprouting response is triggered in neurons with myelinated axons while those supporting unmyelinated axons are possibly prevented from dying. The reaction of peripheral neurons to injury is such that the number of axons they support varies along the nerve as one goes disto-proximally away from the injury site. Thus discrepancies in results from different laboratories have come about because material for axon counting has been taken from different points along the nerve relative to the injury site and also because the material has been taken from nerves injured in different ways.     

Influence of autograft size on peripheral nerve regeneration in cats

A combination of electrophysiological and histological techniques has been used to study the extent to which the size of an autograft affects myelinated axon regeneration in damaged cat peripheral nerves. The length of the graft and the match between the number of axons and Schwann cell basement membrane tubes in the graft and the repaired nerve were investigated. No difference was found in the success of regeneration through 10, 20 and 30 mm autografts. Changing the number of Schwann cell tubes available to regenerating myelinated axons also had no detectable effect on the success of regeneration.     

Retinal ganglion cell axons regenerate in the presence of intact sensory fibres

A novel allograft paradigm was used to test whether adult mammalian central axons regenerate within a peripheral nerve environment containing intact sensory axons. Retinal ganglion cell axon regeneration was compared following anastomosis of dorsal root ganglia grafts or conventional peripheral nerve grafts to the adult rat optic nerve. Dorsal root ganglia grafts comprised intact sensory and degenerate motor axons, whereas conventional grafts comprised both degenerating sensory and motor axons. Retinal ganglion cell axons were traced after 2 months. Dorsal root ganglia survived with their axons persisting throughout the graft. Comparable numbers of retinal ganglion cells regenerated axons into both dorsal root ganglia (1053+/-223) and conventional grafts (1323+/-881; P>0.05). The results indicate that an intact sensory environment supports central axon regeneration.     

The role of Schwann cells in the regeneration of peripheral nerve axons through muscle basal lamina grafts

Evacuated muscle is a possible substitute for nerve autografts in the repair of damaged peripheral nerves. Previous experiments have shown that killed or evacuated muscle grafts are as effective as nerve autografts for bridging gaps of up to 4 cm between proximal and distal nerve stumps. Evacuated muscle grafts are made of extracellular matrix components, which are good substrates for axon growth in vitro. However, experiments in vivo have generally demonstrated that live Schwann cells are essential for successful axon regeneration. In the present experiments we have used immunohistochemical techniques with anti-S100 and anti-neurofilament antibodies to visualize axon growth and Schwann cell migration into muscle grafts over the first 10 days following grafting. We only saw axons growing into grafts accompanied by Schwann cells, and most though not all Schwann cells were associated with axons. Schwann cell migration from the proximal stump in association with axons was much faster and more extensive than from the distal stump. We examined muscle grafts over the first 20 days after grafting by electron microscopy. Regenerating axons were always associated with Schwann cells, which were mostly in the basal lamina-lined tubes left by the evacuated myofibrils. A comparison between evacuated muscle grafts and grafts in which the muscle had been killed but not evacuated revealed that 7 days after grafting there were more than twice as many regenerated axons in and distal to the evacuated grafts, but that by 20 days the numbers of axons were similar in the two groups.     

Peripheral nerve regeneration is delayed in neuropilin 2-deficient mice

Peripheral nerve transection or crush induces expression of class 3 semaphorins by epineurial and perineurial cells at the injury site and of the neuropilins neuropilin-1 and neuropilin-2 by Schwann and perineurial cells in the nerve segment distal to the injury. Neuropilin-dependent class 3 semaphorin signaling guides axons during neural development, but the significance of this signaling system for regeneration of adult peripheral nerves is not known. To test the hypothesis that neuropilin-2 facilitates peripheral-nerve axonal regeneration, we crushed sciatic nerves of adult neuropilin-2-deficient and littermate control mice. Axonal regeneration through the crush site and into the distal nerve segment, repression by the regenerating axons of Schwann cell p75 neurotrophin receptor expression, remyelination of the regenerating axons, and recovery of normal gait were all significantly slower in the neuropilin-2-deficient mice than in the control mice. Thus, neuropilin-2 facilitates peripheral-nerve axonal regeneration.     

Regeneration and soma size changes following axotomy of the trochlear nerve

The effects of CNS and PNS axotomy of the IVth nerve on cell death, soma size, axon size, and axon number were investigated. In adult cats, the IVth nerve was axotomised by using four surgical paradigms: (1) peripheral IVth nerve crush, (2) peripheral IVth nerve cut, (3) peripheral IVth nerve resection, and (4) a CNS IVth nerve cut in the velum. The extent of cell death resulting from each surgical paradigm was determined. Following axotomy distal to the decussation of the IVth nerves, cell death was least after nerve crush, intermediate after nerve cut, and maximal after resection of 5-7 mm of the nerve. Following axotomy at the decussation--a CNS lesion--most cells died but some successful regeneration was observed. Soma size measurements following a short-term survival (3 days to 4 weeks) before the regenerating axons reached their target muscle revealed that somas of axotomised cells underwent hypotrophy within 1 week of axotomy and then gradually increased in size. They re-attained normal size by 4 weeks postoperative when regenerating axons first reach their target. Following a long-term survival (greater than 2 months), somas were significantly hypertrophied, and the degree of hypertrophy was inversely related to the extent of cell survival up to a limit of 40% soma size increase. Counts and measurements of axons revealed that mean axon diameter of regenerated axons was much smaller than normal 3 months after axotomy, increased during the third to sixth postoperative months, but then showed no subsequent increase and remained below normal. In animals with cell death varying from 10% to 70%, the number of axons in the nerve was maintained constant at approximately 1,000. These data indicate that there is a mechanism for the production and maintenance of the appropriate number of regenerative axonal branches following axotomy. In animals in which cell death exceeded 70%, the number of axons was controlled by a maximum ratio of 3 to 4 axon branches per surviving cell. The results suggest that axon number is strongly influenced by the target muscle and that hypertrophy of regenerated cells is related to the number of axonal sprouts each cell has to produce and support in order to re-establish the preoperative number of axons in the regenerated trochlear nerve.