Impaired axonal regeneration in acrylamide intoxication

Acrylamide is a neurotoxin known to impair regeneration of axons following nerve crush and to produce structurally abnormal regenerating sprouts. To investigate the mechanism of these abnormalities, protein synthesis and fast axonal transport were studied in acrylamide-intoxicated and control rats 2 weeks after sciatic nerve crush. Using an in vitro preparation of sciatic nerve-dorsal root ganglion, there was no difference in ganglion ³H-leucine incorporation between the two groups. In these preparations of sensory axons, as well as in motor axons studied in vivo, a smaller proportion of rapidly transported radioactivity was carried beyond the crush in the acrylamide-regenerating nerves compared to the control-regenerating nerves. Correlative ultrastructural studies demonstrated that this difference reflected the impaired outgrowth of the acrylamide-regenerating nerves, rather than an abnormality in fast transport. The acrylamide-treated sprouts often developed swellings filled with whorls of neurofilaments; in addition, many sprouts ended in massively enlarged growth cones containing membranous organelles. EM autoradiography showed labeled, rapidly transported organelles accumulated in the neurofilamentous whorls, and therefore suggested that these organelles might be “trapped” or impeded in passage through these regions. However, there was no evidence that the growth cones received insufficient amounts of transported protein; in fact, the distended endings were densely labeled and apparently “ballooned” by transported organelles. These results suggest that acrylamide intoxication does not impair regeneration by diminishing the delivery of rapidly transported materials to the growing tip. Rather, the marked distention of the growth cones is interpreted as the morphological consequence of continued delivery of rapidly transported organelles into sprouts unable to utilize them in outgrowth.     

Association of synapsin I with neuronal cytoskeleton. Identification in cytoskeletal preparations in vitro and immunocytochemical localization in brain of synapsin I

Calmodulin-dependent protein kinase II (CaM kinase II) is associated with microtubule preparations and phosphorylates several endogenous proteins including microtubule-associated protein 2, tubulin, and an 80,000-dalton protein doublet (pp80). We now report that pp80 is identical to synapsin I by all criteria studied including molecular weight, isoelectric point, phosphopeptide mapping of cAMP- and calmodulin-dependent phosphorylated protein, comigration with authentic synapsin I, and sensitivity to digestion with collagenase. Synapsin I and CaM kinase II were found in association with both microtubule preparations and preparations enriched in neurofilaments. Antibodies to synapsin I specifically labeled neurofilaments prepared in vitro. Immunocytochemical studies on rat brain tissue demonstrated synapsin I immunoreactivity specifically associated with the neuronal cytoskeleton as well as synaptic vesicles. The observed synapsin I staining on cytoskeletal elements was considerably diminished or abolished by the inclusion of Triton X-100 in the staining solutions. These results indicate that synapsin I is associated with the cytoskeleton and may be an important link between cytoskeletal elements as well as between the cytoskeleton and membrane.     

Neurofilament Phosphorylation in Peripheral Nerve: Changes with Axonal Length and Growth State

We have previously reported that the intrinsic ability of motor axons to sprout can vary considerably from nerve to nerve with an inverse correlation to axonal length. In this study we asked whether this variation might be associated with differences in one axonal component, neurofilaments, near the site of outgrowth. The phosphorylation of epitopes on the heavy subunit of neurofilaments (NF-H) was compared in normal and regenerating axons from long and short nerves in the rat. Quantitative determination of phosphorylation states on NF-H epitopes was made by measuring immunoreactivity to monoclonal antibodies using an enzyme-linked immunosorbent assay system. Our results showed a much higher degree of phosphorylation of epitopes on NF-H in terminal portions of short compared to long axons. There was a significant inverse correlation between phosphorylation of NF-H epitopes and axonal length. In newly formed sprouts NF-H phosphorylation was reduced compared to normal. However, the absolute levels were related to the degree of NF-H phosphorylation in the parent axons. The ability to phosphorylate axonal proteins near the site of outgrowth may correlate with the potential for plastic changes in the axon such as sprouting.     

Axon regeneration across the site of injury in the optic nerve of the newt Triturus pyrrhogaster

Summary The process by which axons regenerate following a freeze injury to the optic nerve of the newt was analyzed by light and electron microscopy. Freezing destroys cellular constituents in a one millimeter segment of the nerve, leaving intact the basal lamina and the blood supply to the eye. No axons are seen at the site of injury one to seven days post lesion. This contrasts with the persistence of normal-appearing but severed unmyelinated axons within the cranial stump which thus give a false appearance of early regeneration. The first axon sprouts traverse the lesion and enter the cranial stump by ten days. The number of regenerating axons increases rapidly thereafter with no signs of random growth at the site of injury. These axon sprouts tend to be somewhat larger than normal unmyelinated axons and contain dense core vesicles and abnormal organelles similar to those in growing axons in tissue culture. The persisting basal lamina inside the optic sheath appears to provide continuity across the site of injury, to orient axon sprouts, and to favor an orderly process of axon regeneration without neuroma formation.The authors wish to express their gratitude to Barbara Heindel and Jill Jones for extremely helpful technical assistance. This work was supported by grants NS 10864 and NS 05666 from the U.S. Public Health Service and by the Medical Research Service of the Veterans Administration     

Incorporation of axonally transported glycoproteins into axolemma during nerve regeneration

The insertion of axonally transported fucosyl glycoproteins into the axolemma of regenerating nerve sprouts was examined in rat sciatic motor axons at intervals after nerve crush. [(3)H]Fucose was injected into the lumbar ventral horns and the nerves were removed at intervals between 1 and 14 d after labeling. To follow the fate of the “pulse- labeled” glycoproteins, we examined the nerves by correlative radiometric and EM radioautographic approaches. The results showed, first, that rapidly transported [(3)H]fucosyl glycoproteins were inserted into the axolemma of regenerating sprouts as well as parent axons. At 1 d after delivery, in addition to the substantial mobile fraction of radioactivity still undergoing bidirectional transport within the axon, a fraction of label was already associated with the axolemma. Insertion of labeled glycoproteins into the sprout axolemma appeared to occur all along the length of the regenerating sprouts, not just in sprout terminals. Once inserted, labeled glycoproteins did not undergo extensive redistribution, nor did they appear in sprout regions that formed (as a result of continued outgrowth) after their insertion. The amount of radioactivity in the regenerating nerves decreased with time, in part as a result of removal of transported label by retrograde transport. By 7-14 d after labeling, radioautography showed that almost all the remaining radioactivity was associated with axolemma. The regenerating sprouts retained increased amounts of labeled glycoproteins; 7 or 14 d after labeling, the regenerating sprouts had over twice as much of radioactivity as comparable lengths of control nerves or parent axons. One role of fast axonal transport in nerve regeneration is the contribution to the regenerating sprout of glycoproteins inserted into the axolemma; these membrane elements are added both during longitudinal outgrowth and during lateral growth and maturation of the sprout.     

Axonal growth during regeneration: a quantitative autoradiographic study

The intraaxonal distribution of labeled glycoproteins in the regenerating hypoglossal nerve of the rabbit was studied by use of quantitative electron microscope autoradiography. 9 d after nerve crush, glycoproteins were labeled by the administration of [3H]fucose to the medulla. The distribution of transported 3H-labeled glycoproteins was determined 18 h later in segments of the regenerating nerve and in the contralateral, intact nerve. At the regenerating tip, the distribution was determined both in growth cones and in non-growth cone axons, 6 and 18 h after labeling. The distribution within the non-growth cone axons of the tips was quite different at 6 and 18 h. At 6 h, the axolemma region contained < 10% of the radioactivity; at 18 h, it contained virtually all the radioactivity. In contrast, the distribution within the growth cones was similar at both time intervals, with 30% of the radioactivity over the axolemmal region. Additional segments of the regenerating nerve also showed a preferential labeling of the axolemmal region. In the intact nerve, 3H-labeled glycoproteins were uniformly distributed. These results suggest that: (a) in this system the labeled glycoproteins reaching the tip of the regenerating axons are inserted into the axolemma between 6 and 18 h after leaving the neuronal perikaryon; (b) at the times studied, there is a fairly constant ratio between glycoproteins reaching the growth cone through axoplasmic transport and glycoproteins inserted into the growth cone axolemma; (c) the axolemma elongates by continuous insertion of membrane precursors at the growth cone; the growth cone then advances, leaving behind an immature axon with a newly formed axolemma; and (d) glycoproteins are preferentially inserted into the axolemma along the entire regenerating axon.     

Developmental changes of synapsin I subcellular localization in rat cerebellar neurons

Synapsin I, one of the major synaptic proteins, is thought to associate with synaptic vesicles and to play a regulatory role in neurotransmitter release. In mature neurons, it is concentrated almost exclusively in presynaptic nerve endings. Here, we studied the subcellular localization of synapsin I during the development of rat cerebellar cortices by immunocytochemistry, using anti-synapsin I antibodies and found that during the development of rat cerebellar cortices it tentatively exists in the dendritic growth cones of immature internal granule cells and in the axonal growth cones of mossy fibers as well as mature presynaptic endings. Also, we found that synapsin I, in the axonal and dendritic growth cones does not necessarily associate with vesicles, but rather with fuzzy filamentous structures in the cytoplasm. In search of the structure of synapsin I in vivo, we employed the quick-freeze, deep-etch technique after immunogold labeling. Synapsin I seems to thereby connect synaptic vesicles or anchor them to cytoskeletons in presynaptic endings.     

Distribution of GAP-43, β-III tubulin and F-actin in developing and regenerating axons and their growth cones in vitro, following neurotrophin treatment

Brain derived neurotrophic factor (BDNF) when added to explant cultures of both embryonic and adult retinal ganglion cell (RGC) axons exerted a marked effect on their growth cone size and complexity and also on the intensity of GAP-43, ß-III tubulin and F-actin immunoreaction product in their axons. GAP-43 was distributed in axons, lamellipodia, and filopodia whereas ß-III tubulin was distributed along the length of developing and adult regenerating axons and also in the C-domain of their growth cones. BDNF-treated developing RGC growth cones were larger and displayed increased numbers of GAP-43 and microtubule-containing branches. Although filopodia and lamellipodia were lost from both developing and adult RGC growth cones following trkB-IgG treatment, the intensity of the immunoreaction product of all these molecules was reduced and trkB-IgGs had no effect on the axonal distribution of ß-III tubulin and GAP-43. BDNF-treated growth cones also displayed increased numbers of F-actin containing filopodia and axonal protrusions. This study demonstrates, for the first time, that trkB-IgG treatment causes the loss of F-actin in the P-domain of growth cone tips in developing and regenerating RGC axons. Although microtubules and F-actin domains normally remained distinct in cultured growth cones, ß-III tubulin and F-actin overlapped within the growth cone C-domain, and within axonal protrusions of adult RGC axons, under higher concentrations of BDNF. The collapse of RGC growth cones appeared to correlate with the loss of F-actin. In vitro, trkB signalling may therefore be involved in the maintenance and stabilisation of RGC axons, by influencing F-actin polymerisation, stabilisation and distribution.     

Control of axonal caliber by neurofilament transport

The role of neurofilaments, the intermediate filaments of nerve cells, has been conjectural. Previous morphological studies have suggested a close relationship between neurofilament content and axonal caliber. In this study, the regenerating neuron was used as a model system for testing the hypotheses that neurofilaments are intrinsic determinants of axonal caliber, and that neurofilament content is controlled by the axonal transport of neurofilaments. This system was chosen because previous studies had shown that, after axotomy, axonal caliber was reduced within the proximal stump of the regenerating nerve and, because the relative amount of neurofilament protein undergoing axonal transport in regenerating axons was selectively reduced. The relationship between axonal caliber and neurofilament number was examined in a systematic fashion in both regenerating and control motor axons in rat L5 ventral root. Reconstruction of the spatial and temporal sequences of axonal atrophy in the proximal stump after axotomy showed that reductions in axonal caliber were first detected in the most proximal region of the root and subsequently progressed in a proximal-to-distal direction at a rate of 1.7 mm/day, which is identical to the rate of neurofilament transport in these neurons. Quantitative ultrastructural studies showed that these reductions in caliber correlated with a proportional decrease in the number of axonal neurofilaments but not microtubules. These results support the hypotheses that neurofilament content is a major intrinsic determinant of axonal caliber and that neurofilament content is controlled by the axonal transport of neurofilaments. On this basis, we suggest a role for neurofilaments in the control of axonal volume.     

Dynamic behavior of the ends of growing parallel fibers in early postnatal rat cerebellum

The molecular layer of the cerebellum contains parallel fibers, the axons of granule neurons. We have examined the morphology and behavior of parallel fiber growth cones in the early postnatal rat cerebellum using the fluorescent tracer DiI. Parallel fiber growth cones distributed into three categories based on size and shape: short torpedo-like, long torpedo-like, and lamellopodial in form. The torpedo-like growth cones were modified by the addition of lamellopodia and/or filopodia, and the lamellopodial growth cones were often decorated with a filopodium. These three different growth cone morphologies were found throughout the growing region of the molecular layer. The nascent axons elaborated by premigratory granule neurons differed from the longer axons of more developed neurons in that they often had forked growth cones and extensive lamellopodial decoration along the axon shaft. Growth cones in living slices closely resembled those observed in the fixed preparations. The living growth cones exhibited frequent lamellopodial rearrangement and a side-to-side head-waving movement. The axon proximal to the growth cone was also dynamic. The axons curved and undulated, and mobile swellings formed along the axon shaft. These observations show that the growth cones of parallel fibers are similar to growth cones described for axons in other developing systems in terms of size, morphological characteristics, and dynamic behavior. © 1998 John Wiley & Sons, Inc. J Neurobiol 36: 91–104, 1998     

ULTRASTRUCTURE AND FUNCTION OF GROWTH CONES AND AXONS OF CULTURED NERVE CELLS

Dorsal root ganglion nerve cells undergoing axon elongation in vitro have been analyzed ultrastructurally. The growth cone at the axonal tip contains smooth endoplasmic reticulum, vesicles, neurofilaments, occasional microtubules, and a network of 50-A in diameter microfilaments. The filamentous network fills the periphery of the growth cone and is the only structure found in microspikes. Elements of the network are oriented parallel to the axis of microspikes, but exhibit little orientation in the growth cone. Cytochalasin B causes rounding up of growth cones, retraction of microspikes, and cessation of axon elongation. The latter biological effect correlates with an ultrastructural alteration in the filamentous network of growth cones and microspikes. No other organelle appears to be affected by the drug. Removal of cytochalasin allows reinitiation of growth cone-microspike activity, and elongation begins anew. Such recovery will occur in the presence of the protein synthesis inhibitor cycloheximide, and in the absence of exogenous nerve growth factor. The neurofilaments and microtubules of axons are regularly spaced. Fine filaments indistinguishable from those in the growth cone interconnect neurofilaments, vesicles, microtubules, and plasma membrane. This filamentous network could provide the structural basis for the initiation of lateral microspikes and perhaps of collateral axons, besides playing a role in axonal transport.     

Axonal cytoskeletal changes after non-disruptive axonal injury

In animal models of human diffuse axonal injury, axonal swellings leading to secondary axotomy occur between 2 and 6 h after injury. But, analysis of cytoskeletal changes associated with secondary axotomy has not been undertaken. We have carried out a quantitative analysis of cytoskeletal changes in a model of diffuse axonal injury 4 h after stretch-injury to adult guinea-pig optic nerves. The major site of axonal damage was the middle portion of the nerve. There was a statistically significant increase in the proportion of small axons with a diameter of 0.5 μm and smaller in which there was compaction of neurofilaments. Axons with a diameter greater than 2.0 μm demonstrated an increased spacing between cytoskeletal elements throughout the length of the nerve. However, in the middle segment of the nerve these larger axons demonstrated two different types of response. Either, where periaxonal spaces occurred, there was a reduction in axonal calibre, compaction of neurofilaments but no change in their number, and a loss of microtubules. Or, where intramyelinic spaces occurred there was an increased spacing between neurofilaments and microtubules with a significant loss in the number of both. Longitudinal sections showed foci of compaction of neurofilaments interspersed between regions where axonal structure was apparently normal. Neurofilament compaction was correlated with disruption of the axolemma at these foci present some hours after injury. We suggest that the time course of these axonal cytoskeletal changes after stretch-injury to central axons is shorter than those changes documented to occur during Wallerian degeneration.     

Demonstration and localization of synapsin I in vestibular receptors in the cat: immunocytochemical study

The presence and localization of synapsin I, a neuron-specific phosphoprotein, was investigated in the cat vestibular epithelium, using a rabbit antisynapsin I anti-serum. The staining was performed by immunofluorescence or by a peroxidase-antiperoxidase (PAP) technique. A strong immunoreactivity was observed with both methods. This immunoreactivity appeared as spherical patches distributed in the lower part of the epithelium. This distribution pattern is very similar to that of the efferent synaptic endings which form axodendritic synapses with the afferent nerve chalice of type I hair cells, or axosomatic synapses with type II hair cells. Some of the nerve chalices were also labelled; in this case, the immunoreactivity was more evident with PAP staining. These results thus suggest the presence of large amounts of synapsin I in the vestibular efferent nerve endings. These endings are known to be filled with numerous synaptic vesicles. This localization of synapsin I is well correlated with previous work that report a close association between synapsin I and small synaptic vesicles. The presence of synapsin I in sensory endings such as the afferent nerve chalices was unexpected and is under investigation.     

Morphological changes of the terminations of afferent pathways in a neurodegenerative lesion induced by kainic acid

30 days after kainic acid injection into the rat ventrobasal thalamus, lemniscal afferents were labeled using wheat-germ agglutinin conjugated to HRP. They appeared considerably swollen in the area where neuronal post-synaptic targets had been eliminated. Electron microscopic analysis of the lesioned tissue revealed the presence of large profiles containing numerous organelles, particularly smooth endoplasmic reticulum, and giving rise to thin excrescences filled with neurofilaments. Since these morphological features are typical of regenerating "growth cones", we conclude that afferent terminals deprived of their post-synaptic targets undergo morphological changes preparing them for new synapses.     

Glycosylation Sites Flank Phosphorylation Sites on Synapsin I

Synapsin I is concentrated in nerve terminals, where it appears to anchor synaptic vesicles to the cytoskeleton and thereby ensures a steady supply of fusion-competent synaptic vesicles. Although phosphorylation-dependent binding of synapsin I to cytoskeletal elements and synaptic vesicles is well characterized, little is known about synapsin I's O-linked N-acetylglucosamine (O-GlcNAc) modifications. Here, we identified seven in vivo O-GlcNAcylation sites on synapsin I by analysis of HPLC-purified digests of rat brain synapsin I. The seven O-GlcNAcylation sites (Ser55, Thr56, Thr87, Ser516, Thr524, Thr562, and Ser576) in synapsin I are clustered around its five phosphorylation sites in domains B and D. The proximity of phosphorylation sites to O-GlcNAcylation sites in the regulatory domains of synapsin I suggests that O-GlcNAcylation may modulate phosphorylation and indirectly affect synapsin I interactions. With use of synthetic peptides, however, the presence of an O-GlcNAc at sites Thr562 and Ser576 resulted in only a 66% increase in the Km of calcium/calmodulin-dependent protein kinase II phosphorylation of site Ser566 with no effect on its Vmax. We conclude that O-GlcNAcylation likely plays a more direct role in synapsin I interactions than simply modulating the protein's phosphorylation.     

Immuno-electron-microscopic localization of basic fibroblast growth factor in the dystrophic mdx mouse masseter muscle

Summary The localization of basic fibroblast growth factor (bFGF)-like immunoreactivity in the masseter muscle of dystrophic mdx mice on postnatal day 28 was investigated by immunoblot analysis and electron microscopy. Crude homogenate of the masseter muscle, when subjected to immunoblotting with a bFGF antiserum, exhibited a main band with the same molecular weight (18 kDa) as bovine bFGF. By electron microscopy, bFGF immunoreactivity was detected in small regenerating myocytes; the smaller cells were the premature myocytes, the most intense staining was the immunoreactivity within the cytoplasm. Putative precursors of the muscle cells with a few myofilaments, which were most intensely labeled with anti-bFGF, contacted each other and possibly developed into multinucleated myocytes through cell fusion. Mature myocytes with densely packed myofilaments and peripherally located nuclei did not exhibit bFGF immunoreactivity; they formed myoneural junctions with motor nerve endings immunoreactive for bFGF. Early differentiating myocytes with intense bFGF-like immunoreactivity did not make contact with immunoreactive nerve terminals. Degenerating large myocytes with a limited number of distorted and/or disrupted myofilaments exhibited electron-dense deposits in the cristae of mitochondria; these deposits were not abolished by immunoadsorption control experiments. Thus, the cell-size-dependent decrease in bFGF immunoreactivity in regenerating but not in degenerating myocytes provides a morphological basis for an autoregulatory role of bFGF in muscle regeneration. This study suggests that neuronal bFGF is not involved in initial muscle regeneration in the dystrophic mdx mouse.     

A Stochastic Multiscale Model That Explains the Segregation of Axonal Microtubules and Neurofilaments in Neurological Diseases

The organization of the axonal cytoskeleton is a key determinant of the normal function of an axon, which is a long thin projection of a neuron. Under normal conditions two axonal cytoskeletal polymers, microtubules and neurofilaments, align longitudinally in axons and are interspersed in axonal cross-sections. However, in many neurotoxic and neurodegenerative disorders, microtubules and neurofilaments segregate apart from each other, with microtubules and membranous organelles clustered centrally and neurofilaments displaced to the periphery. This striking segregation precedes the abnormal and excessive neurofilament accumulation in these diseases, which in turn leads to focal axonal swellings. While neurofilament accumulation suggests an impairment of neurofilament transport along axons, the underlying mechanism of their segregation from microtubules remains poorly understood for over 30 years. To address this question, we developed a stochastic multiscale model for the cross-sectional distribution of microtubules and neurofilaments in axons. The model describes microtubules, neurofilaments and organelles as interacting particles in a 2D cross-section, and is built upon molecular processes that occur on a time scale of seconds or shorter. It incorporates the longitudinal transport of neurofilaments and organelles through this domain by allowing stochastic arrival and departure of these cargoes, and integrates the dynamic interactions of these cargoes with microtubules mediated by molecular motors. Simulations of the model demonstrate that organelles can pull nearby microtubules together, and in the absence of neurofilament transport, this mechanism gradually segregates microtubules from neurofilaments on a time scale of hours, similar to that observed in toxic neuropathies. This suggests that the microtubule-neurofilament segregation can be a consequence of the selective impairment of neurofilament transport. The model generates the experimentally testable prediction that the rate and extent of segregation will be dependent on the sizes of the moving organelles as well as the density of their traffic.     

Arrival, reversal, and departure of neurofilaments at the tips of growing axons

We have investigated the movement of green fluorescent protein-tagged neurofilaments at the distal ends of growing axons by using time-lapse fluorescence imaging. The filaments moved in a rapid, infrequent, and asynchronous manner in either an anterograde or retrograde direction (60% anterograde, 40% retrograde). Most of the anterograde filaments entered the growth cone and most of the retrograde filaments originated in the growth cone. In a small number of cases we were able to observe neurofilaments reverse direction, and all of these reversals occurred in or close to the growth cone. We conclude that neurofilament polymers are delivered rapidly and infrequently to the tips of growing axons and that some of these polymers reverse direction in the growth cone and move back into the axon. We propose that 1) growth cones are a preferential site of neurofilament reversal in distal axons, 2) most retrograde neurofilaments in distal axons originate by reversal of anterograde filaments in the growth cone, 3) those anterograde filaments that do not reverse direction are recruited to form the neurofilament cytoskeleton of the newly forming axon, and 4) the net delivery of neurofilament polymers to growth cones may be controlled by regulating the reversal frequency.     

Qualitative and quantitative comparison of the distribution of phosphorylated and non-phosphorylated neurofilament epitopes within central and peripheral axons of adult hamster (Mesocricetus auratus)

Summary The distribution of phosphorylated and nonphosphorylated neurofilament epitopes was determined immunocytochemically in adjacent 2 m-thick sections of sciatic nerve, ventral root and spinal cord. Staining was scored as either intense, moderate or absent and the proportion of labeled axons was calculated for each category. Nearly all sciatic nerve and ventral root axons were immunoreactive with both antibodies against phosphorylated and non-phosphorylated neurofilaments and there were no significant differences in the number of intensely- or moderately-labeled axons. Within the spinal cord however, while the majority of large caliber axons was stained with both antibodies, there was a significant number of small caliber axons which stained only with antibodies against phosphorylated neurofilaments. These results show that phosphorylated and nonphosphorylated neurofilaments are extensively codistributed in CNS and PNS axons, and that in the CNS, staining intensity for non-phosphorylated epitopes is less in the smaller axons.     

Fine structure of growth cones in the upper dorsal horn of the adult primate spinal cord in the course of reactive synapto-neogenesis

Summary Following transganglionic degenerative atrophy of primary afferent terminals induced by a crush-injury of the sciatic nerve, a regenerative process takes places in the upper dorsal horn of the lumbar spinal cord in the primate Macacus rhesus. Axonal growth cones are characterized by cisterns of axoplasmic reticulum; filopodia emanating from growth cones are electron-optically translucent sheet-like expansions, often containing growth-cone vesicles. Axoplasmic reticulum appears also in preterminal portions of regenerating axons. Dendritic growth cones contain a fine, filamentous matrix; electron-dense membrane specializations can be seen in well-defined areas of their surfaces. Immature synapses are formed between filopodia of axonal growth cones and dendritic growth cones. Electron-microscopic structures of this unique CNS regeneration are similar to those seen in the course of embryonic development of the spinal cord.