Notes: In this report, we examined the possible functions of the cell death protease, caspase-3, in the axotomy-induced apoptosis of facial motoneurons in newborn rodents. Using in situ hybridization and Western blot, we found higher levels of caspase-3 mRNA and pro-caspase-3 protein expression in motoneurons of neonatal and 2-week-old rats than adult rats. Following facial motoneuron axotomy, caspase-3 mRNA and protein expression increased in motoneurons of both neonatal and adult rats. However, using an antibody directed to the activated form of the caspase-3 protease, we found that catalytically active caspase-3 was present only in axotomized neonatal motoneurons. As motoneurons in neonatal but not adult rodents are susceptible to axotomy-induced apoptosis, we hypothesized that caspase-3 may play a role in their demise. To determine the necessity of caspase-3 activation in axotomy-induced apoptosis, we counted the number of surviving motoneurons at 4 and 7 days following axotomy in wild type mice and caspase-3 gene-deleted mice. There were nearly three times more surviving motoneurons in caspase-3 gene-deleted mice than in wild type mice at both 4 days (mean 1074 vs. 464, P 〈 0.005) and 7 days (mean 469 vs. 190, P 〈 0.005) following injury, indicating a slower rate of death. Examination of the dying motoneurons using TUNEL staining (for fragmented DNA) and bisbenzimide staining (for nuclear morphology) revealed incomplete nuclear condensation in caspase-3-deficient motoneurons. These results demonstrate that caspase-3 activation plays important roles in the rapid demise of axotomized neonatal motoneurons.

Notes: Motoneurons of the adult survive after axotomy even though they are deprived of putative target derived trophic factors. Alternative sources of trophic support may substitute. In this study we test the hypothesis that the immediate environment of the motoneuronal cell body or the cell body itself increases the production of trophic factors after axonal injury. Using in situ hybridization (ISH) and reverse transcription–polymerase chain reaction (RT-PCR), we report that after axotomy, rat facial motoneurons increase the expression of mRNA for brainderived neurotrophic factor (BDNF) and its receptor trkB. After transection of the facial nerve, we measured a 2-to 4-fold increase in BDNF mRNA expression which had its onset between 3 and 8 h after injury. The BDNF mRNA levels peaked at ∼1–2 days and gradually declined thereafter to return to contralateral levels within 7 days of injury. Western blotting revealed a several-fold increase in BDNF as early as 24 h, which subsequently reached a maximum in ∼5–7 days and was still sustained at 2 weeks post-axotomy. Using exon-specific primers, we determined that the increase in BDNF mRNA is largely due to an increased expression from the promoters of exons IV and III, and to a lesser extent from exons I and II. Analysing the mRNA expression for the BDNF receptor, trkB, we found a 2- to 3-fold increase in full-length trkB mRNA expression starting 2 days after axotomy which lasted for 2–3 weeks. These findings suggest that BDNF might act locally on axotomized motoneurons in an autocrine fashion, providing support for axotomized motoneurons during the first weeks after axotomy.

Notes: We tested whether regeneration of transected rubrospinal tract (RST) axons is facilitated by a prolonged electrical stimulation of these axons. A peripheral nerve was grafted to the transected RST at the cervical level (C4/5) of adult rats, providing a permissive environment for regeneration of rubrospinal axons. Direct antidromic stimulation of the RST was applied immediately after grafting through a microwire inserted just rostral to the RST lesion, using a 1-h 20-Hz supramaximal stimulation protocol. Stimulation caused no direct damage to rubrospinal axons, and was sufficient to recruit the entire rubrospinal tract. In control animals that had a nerve graft and implanted microwire with no stimulation, there were 42.7 ± 10.2 rubrospinal neurons regenerated into the graft at 8 weeks, as assessed by retrograde labelling. In test animals that were stimulated there were 28.2 ± 7.4 backlabelled neurons, not significantly different from control, indicating that this stimulation did not improve the regenerative capacity of rubrospinal neurons. Furthermore, reverse-transcriptase polymerase chain reaction and in situ hybridization for brain-derived neurotrophic factor (BDNF) and/or growth-associated protein-43 (GAP-43) expression in rubrospinal neurons revealed no significant difference between stimulated and unstimulated groups at 48 h after injury, with either 1 or 8 h of stimulation. In summary, direct stimulation of the injured RST axons for the periods tested does not increase expression of GAP-43 and BDNF, and ultimately does not promote regeneration of these central nervous system axons.

Notes: Myelin-derived molecules inhibit axonal regeneration in the CNS. The Long–Evans Shaker rat is a naturally occurring dysmyelinated mutant, which although able to express the components of myelin lacks functional myelin in adulthood. Given that myelin breakdown exposes axons to molecules that are inhibitory to regeneration, we sought to determine whether injured dorsal column axons in a Shaker rat would exhibit a regenerative response absent in normally myelinated Long–Evans (control) rats. Although Shaker rat axons did not regenerate beyond the lesion, they remained at the caudal end of the crush site. Control rat axons, in contrast, retracted and died back from the edge of the crush. The absence of retraction/dieback in Shaker rats was associated with a reduced phagocytic reaction to dorsal column crush around the caudal edge of the lesion. Systemic injection of minocycline, a tetracycline derivative, in control rats reduced both the macrophage response and axonal retraction/dieback following dorsal column injury. In contrast, increasing macrophage activation by spinal injection of the yeast particulate zymosan had no effect on axonal retraction/dieback in Shaker rats. Schwann cell invasion was reduced in minocycline-treated control rats compared with untreated control rats, and was almost undetectable in Shaker rats, suggesting that like axonal retraction/dieback, spinal Schwann cell infiltration is dependent upon macrophage-mediated myelin degeneration. These results indicate that following spinal cord injury the phagocyte-mediated degeneration of myelin and subsequent exposure of inhibitory molecules to the injured axons contributes to their retraction/dieback.

Notes: Brain-derived neurotrophic factor (BDNF) and neurotrophin-3 (NT-3) have been identified as survival factors for adult axotomized rat corticospinal neurons (CSN) in vivo. Axotomy of corticospinal neurons at the level of the internal capsule induced death of 46% of the CSN within the first week after axotomy. The surviving population of CSN displayed severe atrophy with mean cross-sectional area 49% of their unlesioned contralateral counterparts 7 days after axotomy. Using in situ hybridization to assess the expression of the receptors for the family of neurotrophins, we found trkB and trkC but not trkA mRNA expression in CSN. Intraparenchymal application of BDNF or NT-3 at doses of 12 μg/day for 7 days via an osmotic minipump fully prevented the axotomy-induced death of CSN. Interestingly, no neuronal atrophy was seen after BDNF application while NT-3 had only a partial effect on the size of the axotomized CSN. Nerve growth factor did not prevent death or cell atrophy, consistent with the lack of trkA mRNA expression in these neurons. These findings show that BDNF and NT-3 are survival factors for adult rat CSN in vivo, and may contribute to the development of therapeutic strategies aiming at the prevention of CSN degeneration in human motor neuron diseases.

Notes: Summary Tight and gap junctions are described on the basis of freeze-fractures in normal chicken sciatic nerves as well as during Wallerian degeneration and subsequent regeneration. 1. Small calibre nerve fibres display a fairly continuous tight junction contact zone in the membranes of the mesaxons, paranodal loops and Schmidt-Lanterman incisures. Large fibres with more than 40 lamellae have only focal tight junction contacts in the mesaxonal membranes. 2. With the onset of Wallerian degeneration (days 2–4 post-crush, distal stump) myelinic tight junctions become arranged as maculae composed of one circular or several polygonally oriented strands that are criss-crossed by other tight junctional strands. These maculae are subsequently found in the membranes of cytoplasmic vacuoles of the Schwann cells, indicating an endocytotic mode of uptake. Tight junctions are not found between the 5th and 6th day after crush. 3. During the proliferation phase of the Schwann cells and the arrangement of these cells into Büngner cell bands (2 to 8 days post-crush) gap junctions appear between the Schwann cells of the bands. These junctions then disappear with the onset of remyelination (8 days post-crush). 4. With the onset of remyelination (from the 8th day onwards) short focal tight junctions appear in the membranes of the outer mesaxons. Shortly thereafter, when the sheaths possess 4 to 8 lamellae, tight junctions also appear in the membranes of the inner mesaxons, the paranodal loops and the cytoplasmic inclusions. The characteristic differences of tight junction elaboration in small versus large nerve fibres are re-established after three months of regeneration. The elaborated tight junctions in small and early remyelinating fibres point to a specific function; in small fibres (versus large fibres) the tight junctions might effect a separation of the intramyelinic extracellular space as a single compartment. The tight junction contacts in early remyelinating fibres support the hypothesis that myelin growth occurs within the myelin spiral and not by a free rotation and elongation of the Schwann cell tongues. It is assumed that the gap junctions between the Schwann cells contribute to the co-ordination of the Schwann cell band formation, which is involved in the guidance of sprouting axons.