Mary E. Bernstein

Axonal regeneration and formation of synapses proximal to the site of lesion following hemisection of the rat spinal cord

Although the central nervous system of the mammal is reported not to regenerate, axonal sprouting has recently been demonstrated in various regions of brain and spinal cord. The following study investigates the regenerative capacity of 72 rat spinal cords following left hemisection at T2. In addition to nine normals, rats were killed at 5, 7, 14, 30, 60, and 90 days after making the lesion. Three animals per groups were prepared for Golgi, Protargol-eosin-cresyl violet staining, and electron microscopy. Three additional animals had hemisections at C5 and the cord from C5 to T5 was stained by the Fink-Heimer method. Six animals had hemisections at T2 and the cords were subsequently rehemisected 90 days later at C5; three animals were prepared for the Fink-Heimer stain and three for electron microscopy. The dendrites of the reactive motor neurons proximal to the site of lesion became varicose from day 10 to day 60–90 after the lesion, until the entire dendrite was replete with irregularly shaped varicosities. At 15 days postoperatively, processes grew into the reactive neural zone and by day 30 could be identifified as axons (0.1–0.5 μ in diameter) and dendrites. The axons grew in fascicles, usually free of neuroglial cell processes. The amjority of the axons formed axodendritic synapses on the indentations of the dendritic varicosities by 60–90 days postoperatively. The Fink-Heimer stain reveled that a limited number of axons regenerated from long tracts into the reactive neural zone. Most regenerated axons appear to be axonal sprouts from the operated and unoporated portions of the spinal cord. The central nervous system of the rat does regenerate, but the regenerating axons do not grow past the reactive neural zone and thus do not reach the neuroglial scar.

Neuronal Alteration and Reinnervation Following Axonal Regeneration and Sprouting in Mammalian Spinal Cord (Part 1 of 2)

Recent findings have shown that the spinalcord of mammals is capable of limited regeneration. The reparative process involves very limited regeneration of severed nerve fibers and growth of axonal spr

Ventral horn synaptology in the rat.

SummaryThe synaptology of the normal ventral horn of the rat was studied. Presynaptic boutons were classified as S (spherical vesicles), F (flattened vesicles), and G (predominance of 700–1200 Å granular vesicles). In addition, Cf, Cs, M, and T synaptic complexes were defined and quantitated. Synaptology was studied on α-motoneuron somata, α-motoneuron primary dendrites, peripheral dendrites and interneuron somata. In addition, organelles were quantified for the pre- and postsynaptic members of the synaptic complex. All counts were made on coded material and these data were analyzed statistically.Motoneuron somata had significantly more (P < 0.01) F (58%) than S (33%) boutons. This was also the case for the motoneuron primary dendrite (P < 0.01; F, 61%; S, 37%). The small dendrites had more (P < 0.05) S (56%) than F (44%) boutons. More Cf bulbs (P < 0.01) were found on motoneuron somata (9%) than on motoneuron primary dendrites (2%) or interneuron somata (3%). The C complex presynaptic bouton contained spherical (Cs) or flattened (Cf) synaptic vesicles which were attributed to the fixation employed. Cf bulbs were not observed on small dendrites. G bulbs were observed (< 1%) only on small dendrites. M bulbs were not observed on any postsynaptic structure.The boutons of the motoneuron primary dendrites (15% of total afferents) and peripheral dendrites (14% of total afferents) were frequently branched whereas there was significantly (P < 0.01) less branching of boutons on motoneuron and interneuron somata. Small postsynaptic subsurface cisterns were associated with boutons of both the S and F type on all structures. In addition, these cisterns were observed in motoneuron somata (4%) and interneuron somata (2%) without an accompanying bouton. C postsynaptic organelles were observed in motoneuron somata (3%) and primary dendrites (1%) with an overlying neuroglial cell process and no presynaptic bouton.The synaptology of the rat ventral horn is comparable to that in the cat and monkey. However, M (R) and P bulbs were not observed in the rat. This could be due to the sampling method which indicated that synapses with less than 1% occurrence fall at the level of statistical resolution in quantitative electron microscopy. The presence of postsynaptic specialization usually associated with presynaptic boutons with no presynaptic component may be a reflection of the dynamics of normal bouton renewal in the rat ventral horn.

Ultrastructure of normal regeneration and loss of regenerative capacity following teflon blockage in goldfish spinal cord.

Abstract Although the spinal cord of goldfish normally regenerates, 30 or more days of blockage by Teflon inserts will result in the loss of this regenerative capacity. The following studies were undertaken to investigate the ultrastructure of normal regeneration and the loss of regenerative capacity following Teflon blockage. The spinal cords of 13 normal and 27 cord-transected goldfish were examined with the electron microscope. Cord-transected goldfish were killed 1 day to 3 months postoperatively. In 15 additional goldfish the spinal cords were transected and Teflon sheet placed between the severed stumps for 1–3 months, after which time it is known the spinal cord will not regenerate following Teflon removal. The spinal cords were immersed in buffered 1% osmic acid and prepared for electron microscopy. In the absence of Teflon, regenerating axons were repelte with smooth endoplasmic reticulum and mitochondria. The majority of axons had regenerated across the site of lesion by day 30 and synapsed in the first segment of the caudal spinal cord stump. Dendrites regenerated in neurite fascicles along with axons. After blockage by Teflon, the nerve fiber tips grew into juxtaposition in zones subjacent to a glialependymal scar. These axons possessed multibulbous processes at the tip which usually formed axoaxonic and axodendritic synapses. Retransection (five goldfish) of the spinal cord one segment rostral to the Teflon block resulted in the degeneration of approximately 50% of all the synapses in the synaptic zone subjacent to the glial-ependymal scar.

Regeneration of axons and synaptic complex formation rostral to the site of hemisection in the spinal cord of the monkey.

Historically, injury of the mammalian spinal cord resulted in abortive regeneration. Recent findings have shown regeneration of the spinal cord occurs by limited regrowth of severed nerve fibers and massive regrowth of axonal sprouts of normal axons. The foregoing study investigates the regenerative capacity of the spinal cord in 14 Rhesus monkeys (Macaca mulatto) following left hemisection at vertebral segment T2. In addition to three normals, animals were utilized 7, 14, 21 days and 1, 2, 3, and 4 months posthemisection. Tissue was prepared for Golgi impregnation, Protargol-cresyl violet-eosin staining, and electron microscopy. The motor horn rostral to the site of lesion was investigated. Dendrites of motor horn cells adjacent to and facing the lesion developed varicosities which formed at the terminal end of the dendrite by day 7 and progressed to include the entire dendrite by day 14–30 posthemisection. Motor horn cell dendrites (0–5 mm from lesion) were varicose and many possessed only two short, va...

Effect of glial-ependymal scar and Teflon arrest on the regenerative capacity of goldfish spinal cord

Abstract The regenerative capacity of the goldfish spinal cord is not affected by connective tissue or parenchymal scars in the normal course of regeneration. The present experiments were undertaken to study the effects of both mechanical arrest and a glialependymal scar blockade on the regenerative capacity of the spinal cord. Spinal cords of fifty goldfish were transected and Teflon was placed between the severed stumps. Fourteen, 30, 60, or 90 days later, the Teflon was removed and regeneration assessed up to 6 months postoperatively. Regeneration of the spinal cord was severely retarded after 14 days of Teflon arrest and did not occur after 30 or more days of arrest. After abortive regeneration, nerve fibers were found to terminate at a glial-ependymal scar. To ascertain if the loss in regenerative capacity was due to the glial-ependymal scar or Teflon arrest, the spinal cord was cut and Teflon inserted for 30 days. The Teflon was then removed and the cord transected one segment rostral to the former mechanical block. Thirty days later the axons had regenerated through the isolated spinal cord segment and past the glial-ependymal scar. Therefore, the inability of the axons to regenerate after Teflon removal was due to the inability of the neurons to resume growth and not to any mechanical effects of the scar.

Alteration of neuronal synaptic complement during regeneration and axonal sprouting of rat spinal cord

Abstract Spinal cord hemisection results in limited regeneration of nerve fibers and the sprouting of intact nerve fibers proximal to the site of lesion. Electron microscopically the nerve fibers form new synaptic complexes. The following study was undertaken in the rat to determine the synaptic profile of neurons during the regenerative process. The spinal cords of 40 rats were hemisected at vertebral segment T-2 and 5 animals per group (in addition to 5 normals) were utilized 10, 20, 30, 45, 60, and 90 days posthemisection. Tissue was prepared by the Rasmussen technique for the light microscopic demonstration of boutons. In addition, five animals had the cord hemisected and the spinal cord tissue was prepared for electron microscopy 40 days postoperatively. Bouton counts were made on perikaryon and primary dendrite (of the same neuron) of neurons in lamina IV, lamina VII, and on motoneurons. Statistical analysis of counts from coded material was carried out utilizing a polynomial regression, analysis of variance, and a Neuman-Keuls a posteriori analysis. The number of boutons on the perikaryon of neurons on the operated side demonstrated a significant decrease in boutons at 10–20 days postoperative followed by a significant increase in boutons at 30 days to levels below normal innervation. This reinnervation was followed by a significant secondary loss of boutons from 30 to 60 days. This trend of bouton alteration was repeated (with minor variations) on the perikaryon of neurons on the unoperated side of the spinal cord. This general trend was repeated on the primary dendrite on both operated and unoperated side of the spinal cord. However, the number of boutons on primary dendrite usually returned to normal. The regenerated boutons (30 days) frequently contained granular vesicles (predominantly 900 A). The boutons that degenerated 30–60 days had all combinations of synaptic vesicles. This is an indication that former and new synaptic complexes were degenerating. The alteration of boutons on perikaryon and primary dendrite following hemisection suggests: that there is a critical period of initial denervation (0–20 days); that reinnervation occurs within the first 30 days; and that a secondary denervation occurs after 30–60 days. This recombination of neuronal circuits may represent an accelerated view of a dynamic continuum which imparts information to denervated neurons about altered neuronal circuitry adding a further dimension to the plasticity of the adult nervous system.

Synaptic frequency alteration on rat ventral horn neurons in the first segment proximal to spinal cord hemisection: an ultrastructural statistical study of regenerative capacity

SummaryTo study the regenerative capacity of the spinal cord in adult rat, presynaptic boutons were classified as S (spherical vesicles), F (flattened vesicles) and C complexes, and analysed statistically on α-motoneuron somata and lamina VII interneurons on the operated side in the first segment rostral to a spinal cord hemisection. Following chloral hydrate anaesthesia left spinal cord hemisections were made on twenty adult rats (225 gms) at vertebral level T-2. Animals were prepared for electron microscopy at 7, 14, 30, 45, 60 and 90 DPO and compared with normals. All counts were made on coded material and subjected to statistical analysis. The normal frequency of presynaptic bouton types on α-motoneuron somata and primary dendrites was altered over the entire postoperative period. S presynaptic boutons were increased on α-motoneuron somata at 30 DPO. At 45 DPO, massive degeneration with concomitant synaptic remodeling resulted in a return to near normal frequencies of S and F presynaptic boutons. At 60 and 90 DPO a gain in S presynaptic boutons and a concomitant loss in F presynaptic boutons resulted in frequencies different from normal and decreased absolute numbers of presynaptic boutons. The interneuron somata also exhibited alterations over the postoperative period. There was a reversal of frequency of presynaptic boutons at 45 DPO. However unlike on α-motoneuron somata the frequency of S and F presynaptic boutons returned to normal at 60 and 90 DPO. The α-motoneuron somata appeared to be cyclically and nonselectively reinnervated by ventral horn interneurons over 90 DPO.

Effect of puromycin treatment on the regeneration of hemisected and transected rat spinal cord.

SummaryThe effect of puromycin on spinal cord regeneration was studied following implantation into the site of spinal cord hemi- or transection of Gel-foam saturated with puromycin (1mM) in a saline carrier, implantation of Gel-foam sponge saturated with saline (carrier control), or lesion alone (lesion control). The spinal cords of 107 rats were studied with light and electron microscopy 7, 14, 30, 60 and 90 days postoperative (DPO). Spinal cord hemisected animals developed a dense cicatrix at the site of lesion replete with connective tissue, blood vessels, and myelinated and unmyelinated nerve fibres which could be traced to peripheral sources. Rostrally at the C.N.S.-cicatrix interface, there were reactive neuroglial cells, occasional nerve fibres and finger-like projections of spinal cord (due to cavitation lesions) which contained neuroglia, axons and dendrites. Implantation of saline in Gel-foam resulted in the same morphology as in hemisected animals except for increased lesion size due to mechanical factors and decreased cicatrix density during the first 30 DPO. Puromycin treatment resulted in a cicatrix with initial decreased cell density but which contained a new class of nerve fibres at 30 DPO. These nerve fibres were oriented in a rostro—caudal direction, were unmyelinated, 0.1–0.2 μm in diameter and had expanded smooth endoplasmic reticulum. Some of these nerve fibres were degenerating at 30 DPO and all were absent by 60 DPO. The puromycin-treated spinal cord within 200 μm rostral to the basal lamina contained nerve terminal conglomerates, which resembled boutons, in fascicles from 30–90 DPO (duration of experiment). Hemisection of the spinal cord by crushing 1-11/2 segments rostral to the site of puromycin implantation at 30 DPO resulted in degeneration of these nerve fibres in the cicatrix as well as the degeneration of nerve terminal conglomerates just rostral to the basal lamina. The regenerative capacity of the spinal cord is discussed in relationship to these findings.

Dendritic growth cone and filopodia formation as a mechanism of spinal cord regeration

Abstract Following hemisection of the rat spinal cord at the T1–T2 vertebral interface, dendritic sprouting was observed on neurons in the motoneuron pool at 45 days postoperative. The sprouting resulted in increased numbers of dendritic terminals, dendritic growth cones, and dendritic filopodial growth. The morphologic characteristics of the dendritic sprouts are similar to neurons in developing as well as in mature animals, and are correlated with synaptic renewal during spinal cord regeneration.