- Email: [email protected]
Survival of chronically-injured neurons can be prolonged by treatment with neurotrophic factors
Neuroscience Vol. 94, No. 3, pp. 929–936, 1999 929 Copyright q 1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/99 $20.00+0.00
Trophic factors promote long-term neuron survival
Pergamon PII: S0306-4522(99)00359-0
SURVIVAL OF CHRONICALLY-INJURED NEURONS CAN BE PROLONGED BY TREATMENT WITH NEUROTROPHIC FACTORS J. D. HOULE* and J.-H. YE Department of Anatomy, University of Arkansas for Medical Sciences, 4301 West Markham Street, Little Rock, AR 72205, U.S.A.
Abstract—Axonal regeneration by chronically-injured supraspinal neurons can be enhanced by neurotrophic factor treatment at the site of injury, although the number of regenerating neurons decreases as the interval between spinal cord injury and treatment increases. This study investigated whether this decline in regenerative response could be due to continued loss of neurons during the post-injury period. Adult rats received a cervical hemisection lesion and axotomized neurons were labeled by retrograde transport of True Blue from the lesion site. Animals were killed one, four or eight weeks after injury and surviving neurons (True Bluelabeled) were counted in the red nucleus and lateral vestibular nucleus. The neuron number in the lateral vestibular nucleus was stable for eight weeks after spinal cord injury, while survival in the red nucleus decreased by 25% between four and eight weeks. To test how neurons respond to a second injury with or without trophic factor treatment, at four, eight, 14 or 22 weeks after injury the lesion cavity was enlarged by 0.5 mm in a rostral direction. Gel foam saturated with ciliary neurotrophic factor, brain-derived neurotrophic factor or basic fibroblast growth factor was placed into the cavity. Animals were killed four weeks later. Re-injury of the spinal cord caused a significant decrease in neuron survival in both the red nucleus and lateral vestibular nucleus, the effects of which were lessened by treatment with ciliary neurotrophic factor or brain-derived neurotrophic factor for the red nucleus and with ciliary neurotrophic factor for the lateral vestibular nucleus, when re-injured at four or eight weeks. Basic fibroblast growth factor did not affect neuron survival at any time post-injury. Ciliary neurotrophic factor was not effective with longer delays (14 or 22 weeks) between the initial injury and re-injury. These results indicate a delayed pattern of secondary neuronal cell loss after spinal cord injury that is exaggerated by re-injury, but which can be ameliorated by treatment with neurotrophic factors. q 1999 IBRO. Published by Elsevier Science Ltd. Key words: spinal cord injury, transplantation, CNTF, BDNF, FGF2, axotomy.
approximately 40% of these cells, 9,11,24,28 although there is some dispute as to whether there is severe atrophy exhibited by many of these neurons, which is miscalculated as cell death. 30 Fetal spinal cord tissue transplants 4,24 can affectively rescue dying RN neurons and acute treatment with neurotrophic factors 21 can prevent atrophy of adult RN neurons and promote survival of injured neonatal RN neurons. 7 In contrast to the RN, little is known about atrophy and/or cell death of other brainstem neurons following SCI. One aim of this study was to examine the temporal effects of SCI on another major contributor to descending spinal pathways, the lateral vestibular nucleus (LVN). A second aim was to determine how RN and LVN neurons would respond to the type of second injury that occurs routinely during removal of scar tissue and expansion of the size of the lesion cavity as the original lesion site is prepared for a neural tissue transplant. 14,15 Following injury to the adult rat spinal cord there is the potential for axonal regrowth by brainstem neurons that contribute to most descending spinal pathways, provided that a suitable environment (e.g., fetal neural tissue, peripheral nerve graft, Schwann cell graft, etc.) for growth is provided. Many investigations have demonstrated that these injured neurons can respond to specific trophic or growth factors, such as ciliary neurotrophic factor (CNTF) or basic fibroblast growth factor (FGF2), leading to an enhanced axonal regrowth. 12,17,36 When there is a delay between the time of injury and when the chosen substratum is made available, as in a chronic injury situation, the regenerative effort diminishes in conjunction with increasing delays. 15 Similarly, with a long delay between injury and treatment with trophic or growth factors, the regenerative effort appears to decline. 16
Previous studies have shown the effective use of neurotrophic and growth factors for the rescue of injured neurons in several different systems. Retinal ganglion neurons are extremely sensitive to optic nerve injury, with over 75% of the neurons dying within two weeks. 35 Intraocular treatment with brainderived neurotrophic factor (BDNF) or neurotrophin (NT)-4 can delay cell death for a short period, 22,27 however, a decline in survival is observed one to two weeks after treatment. Similar short-term benefits have been obtained with treatment with glial cell-derived neurotrophic factor. 20 Long-term benefits to a single treatment with growth or trophic factor after injury have been shown for neurons of Clarke’s nucleus 29 and neurons of the lateral geniculate nucleus. 1 Other neurons, such as those forming the corticospinal tract appear to survive for extended periods if trophic factors are provided for several weeks via a cannula placed into the cerebral cortex. 6,10 Injury to the spinal cord causes disruption of descending pathways from brainstem and cortical neurons and initiates a neuronal response which can lead to atrophy or death of the injured cells. 3,31,32 Neurons of the red nucleus (RN) in neonatal animals appear to be very sensitive to spinal cord injury (SCI), as over 55% die after a midthoracic injury. 4 Axotomy of adult rubrospinal neurons is thought to result in death of *To whom correspondence should be addressed. Tel.: 1 1-501-6865849; fax: 1 1-501-6866382. E-mail address: [email protected] (J. D. Houle) Abbreviations: BDNF, brain-derived neurotrophic factor; C, cervical; C3Hx, cervical level 3 hemisection; CNTF, ciliary neurotrophic factor; CSA, cross-sectional area; FGF2, basic fibroblast growth factor; Hx, hemisection; LVN, lateral vestibular nucleus; NT, neurotrophin; PBS, phosphate-buffered saline; PN, peripheral nerve; RN, red nucleus; SCI, spinal cord injury; TB, True Blue. 929
930
J. D. Houle and J.-H. Ye
This observation suggests that injured neurons probably continue to change during the post-injury period, such that an effective intervention strategy at one post-injury interval might no longer be effective after a longer interval before intervention. For effective repair of the chronically-injured spinal cord it is necessary to determine how different parameters of the neuronal response to injury might change with increasing post-injury time and how such changes may be affected by trophic factor support. The present study provides information about long-term neuron survival after SCI which can be incorporated into the development of strategies for intervention to be used at different post-injury periods. The decision to test three different trophic/growth factors was based upon studies 17,36 of the chronically injured spinal cord in which these specific factors (BDNF, CNTF and FGF2) demonstrated the ability to enhance axonal regrowth from a variety of supraspinal neuron groups. EXPERIMENTAL PROCEDURES
Surgical procedures and trophic factor treatment Adult female Sprague–Dawley rats (225–250 g, Charles River) were anesthetized with a mixture of ketamine (60 mg/kg) and xylazine (10 mg/kg). A laminectomy was performed on the second and third cervical (C) vertebrae, the meningeal membranes were excised along the midline and a hemisection (Hx) lesion cavity was created by aspiration of the C3 spinal cord segment. The cavity was approximately 3 mm in length and was complete to the lateral and ventral boundaries formed by the meninges. Upon hemostasis, gel foam saturated with a 2% solution of True Blue (TB, Sigma) was placed into the cavity for 60 min to label injured neurons by retrograde transport of this fluorescent dye. After removing the gel foam the cavity was rinsed with saline and the dorsal surface of the cavity was covered with a piece of silastic film inserted beneath the meninges. The dura mater was sutured and overlying muscle and skin closed with wound clips. Penicillin G (3000 units s.c., every third day for one week) and Buprenorphine (0.1 mg/kg i.m., once daily as needed for analgesia) were provided. Efforts were made to minimize animal suffering and all procedures were in accordance with United States Public Health Service Guidelines for the Care and Use of Laboratory Animals and with approval of the Institutional Animal Care and Use Committee. There were three main experimental groups for this study. (1) Spinal cord-injured rats were allowed to survive for one (n 3), four (n 4) or eight (n 4) weeks with no further manipulation of the injury site. These subgroups were used to determine possible changes in neuron survival at progressively longer post-injury periods. (2) After a C3Hx, the lesion site was exposed four weeks later, scar tissue was removed by aspiration and the cavity was expanded by 0.5 mm in a rostral direction. Gel foam saturated with BDNF (50 mg/ml, Regeneron/ Amgen Partners, n 7), CNTF (100 mg/ml, Regeneron Pharmaceuticals, n 4), FGF2 (1 mg/ml, Collaborative Research, n 5) or phosphate-buffered saline (PBS, n 4) was placed into the cavity. After 5 min the gel foam was removed and replaced with fresh factor-saturated gel foam every 15 min over a 60 min period. Approximately 5 ml of factor was provided with each change of gel foam. Factor-saturated gel foam was left in the cavity after the final change, the cavity was covered with silastic film and the dura was sutured. Animals were maintained for four weeks with no further manipulation. The doses of factors used for this study were those previously found to be effective in promoting axonal regeneration by chronically-injured neurons. 16,17,36 Animals in this subgroup were used to determine the effects of a second injury on neuronal survival and to test whether trophic and/or growth factors would influence survival of chronically injured neurons after a second injury. Removal of scar tissue and reinjury is a necessary step to cavity preparation for a peripheral nerve graft in axonal regeneration studies. 36 (3) After a C3HX, the lesion site was exposed at eight (n 3), 14 (n 3) or 22 (n 3) weeks after spinal cord injury, scar tissue was removed by aspiration and the cavity expanded in a rostral direction by 0.5 mm. Gel foam saturated with CNTF (100 mg/ml) was placed in the cavity as described above. Animals were maintained for four weeks with no additional
manipulation. These subgroups were used to determine neuronal survival following a second injury and trophic factor treatment at postinjury periods longer than four weeks. At the appropriate post-injury or post-treatment period, animals were killed by administration of an overdose of Nembutal (120 mg/ kg) followed by cardiac perfusion with a 4% paraformaldehyde solution. Brains were removed, postfixed with 4% paraformaldehyde overnight at 48C and then immersed in 20% sucrose at 48C. Tissue preparation Tissue blocks were prepared from brainstem regions containing the red nucleus or the vestibular nuclear complex. Two sets of alternating sections (25-mm-thick) were prepared using a cryostat and mounted separately in serial order on glass slides. One set of sections was coverslipped with Fluoromount (Bio/medical Specialties, Santa Monica, CA) and used for detection and analysis of TB labeled neurons within the RN and LVN. The second set of sections was stained with Thionin, coverslipped with Permount (Fisher Scientific) and used for confirmation of nuclear regions used for analysis. For identification of neurons in the RN, the caudal most section through the nucleus magnocellularis where TB labeled neurons could be observed was designated as the first section for analysis, with succeeding sections more rostral to the first. For identification of neurons in the LVN, the caudal-most section where TB labeled neurons were found within the inferior cerebellar peduncle ipsilateral to the lesion was designated as the first section for analysis, with succeeding sections more rostral to the first. The small number of TB labeled neurons in the contralateral LVN were not included in this quantitation. Analysis All neurons labeled with True Blue were counted in the first ten sections (spaced 25 mm apart from each other) through the RN and LVN, to provide a measure of the number of injured neurons that survived for varying periods or after exposure to different trophic or growth factors. Serial sections were compared to insure that duplicate counts of labeled neurons were not made. Labeled neurons were easily distinguished from non-labeled neurons using an ultraviolet light filter package on a Zeiss Axioskop fluorescent microscope. The entire cell soma and short portions of several processes were labeled with TB, giving the cells a characteristic neuronal appearance. Any nonneuronal cells labeled with TB were easily distinguished from neurons based upon their smaller cell size and lack of labeled processes. Stereological counting methods were not used because we were interested in determining the absolute number of neurons present at different time intervals following spinal cord injury. Because the same number of sections were used for quantitative analysis, covering the same caudal to rostral extent of each nuclear region, no corrections to cell counts were made. A one-way analysis of variance was used to test for statistically significant differences within the three large groups of experimental animals for the mean number of TB-labeled neurons. When significant differences were present, individual post hoc comparisons were made with the Bonferroni t-test (P , 0.05). Differences between different experimental groups were determined by a Student’s t-test (P , 0.05). The results from subgroups in experimental group 2 that were treated with trophic or growth factors were compared to the eight week survival, no re-injury subgroup of experimental group 1, because of their common overall length of animal survival. The results from subgroups in experimental group 3 treated with CNTF at different postinjury intervals were compared to the four week re-injury, CNTF treated subgroup of experimental group 2, because of their common use of CNTF as a trophic agent after a second injury. The cross-sectional area (CSA) of surviving neurons at different post-injury periods (experimental group 1) or after a second injury (experimental group 2) was measured from fluorescent photomicrographs with the NIH Image Program. All TB labeled neurons on three sections of the RN (from rostral, medial and caudal portions) from each animal were measured. Neurons of the RN were classified as small (,600 mm 2), medium (600–900 mm 2) or large (.900 mm 2), based upon the size distribution of TB-labeled RN neurons from a previous study. 15 Neurons of the LVN were classified as small (,400 mm 2), medium (400–800 mm 2) or large (.800 mm 2). Significant differences in the percentage of neurons within each size category were determined by an analysis of variance, with individual post hoc comparisons made with the Bonferroni t-test (P , 0.05).
Trophic factors promote long-term neuron survival
931 RESULTS
Change in neuron number and size one to eight weeks after spinal cord injury Neurons within the red nucleus RN and lateral vestibular nucleus that had been injured during preparation of a C3Hx lesion contained the fluorescent dye True Blue (TB) within their cell soma (Figs 1A, 3A). Based upon cell size and spread of the dye that labeled short portions of contiguous cell processes, injured neurons easily were distinguished from surrounding non-neuronal cells. Because the aim of this study was to examine long-term cell survival and the effects of neurotrophic or growth factors on injured neurons only, we included in our examination only those neurons that contained TB. One week after injury, TB labeled neurons were observed throughout the caudal to rostral regions of the RN and LVN that were chosen for study. The number of labeled cells at one week after injury represents those neurons that survived the initial trauma inflicted by SCI and is the maximum number of cells that might be present at any longer post-injury interval or after trophic factor treatment. In the 10 tissue sections examined from each nuclear region, there were 381 ^ 78 RN neurons and 267 ^ 37 LVN neurons labeled with TB (Table 1). Four weeks after injury there was no change in the number of neurons in the RN (Fig. 1B) or LVN compared to those present at one week. There was a significant decrease (to 75% of the number at four weeks post-injury, P , 0.05) in the number of RN neurons that survived for eight weeks after SCI (Fig. 1C, Table 1), whereas, there was no significant change in the number of surviving LVN neurons, compared to those at four weeks post-injury. Effects of a second injury and trophic factor treatment
Fig. 1. Fluorescence micrographs of True Blue-labeled neurons in the red nucleus following SCI at (A) one, (B) four or (C) eight weeks prior to sacrifice. Most labeled neurons are located in the ventral-lateral portion of the magnocellular nucleus of the RN, with no detectable decrease in number between one and four weeks after injury, compared to an obvious decrease in survival of labeled neurons at eight weeks after injury. Scale bar 200 mm. Table 1. Changes in neuron survival in the red nucleus and lateral vestibular nucleus between one and eight weeks after spinal cord injury Group
Red nucleus
Lateral vestibular nucleus
One week post C3Hx Four weeks post C3Hx Eight weeks post C3Hx
381 ^ 78 358 ^ 48 269 ^ 43*
267 ^ 37 239 ^ 15 283 ^ 41
Values are mean cell number ^ S.D. *Indicates significant difference from four weeks post C3Hx at P , 0.05.
Following the removal of scar tissue from the rostral cavity wall and expansion of the cavity in this direction, animals were allowed to survive for four weeks, which was eight weeks after the initial injury and comparable to the longest post-injury period in the previous group of animals. Without a second injury, 75% of the RN neurons survived for eight weeks as did 100% of the injured LVN neurons (see above). Following a second injury and treatment of the lesion site with PBS, there was a significant decrease in survival in both the RN (44% of those present eight weeks after injury) and LVN (48% of those present eight weeks after injury) (Fig. 2A, Table 2). When BDNF was provided after the second injury, over 73% of previously injured RN neurons survived, which was a significant increase over the PBS-treated group (Table 2). In contrast, BDNF treatment did not have a noticeable effect upon the survival of LVN neurons, with 52% survival, compared to 48% after PBS treatment. CNTF treatment after a second injury provided even greater protection to RN neurons, as the same number survived a second injury as did those surviving a single injury (Fig. 2C, D; Table 2). Survival was significantly increased compared to PBS treatment and was improved by 37% compared to BDNF treatment after a second injury. In the LVN, neuron survival after a second injury and CNTF treatment was 65% of that found eight weeks after a single injury, which was not a significant decrease from the single injury, eight week survival group. Providing FGF2 to the lesion site after a second injury had no effect on neuron survival in either the RN or LVN (Fig. 2B),
932
J. D. Houle and J.-H. Ye
Fig. 2. Fluorescent micrograph depicting changes in neuron survival in the RN, after (A) a second injury at four weeks after the initial injury and treatment with PBS or after treatment with (B) FGF2 or (C, D) CNTF. Treatment with PBS or FGF2 failed to rescue RN neurons, compared to the effects of CNTF, where significantly more chronically-injured neurons remained at four weeks after a second injury. There is no obvious atrophy of True Blue-labeled neurons treated with CNTF after a second injury. RN neurons of all sizes are present in this region. Scale bar 200 mm.
as the number of TB-labeled cells was comparable to that found after PBS treatment, which was significantly lower than the number that had survived for eight weeks after only a single injury. Effects of ciliary neurotrophic factor at longer post-injury periods Because only CNTF had a significant effect on neuron survival after a delayed second injury to RN and LVN neurons, only CNTF treatment at longer post-injury periods was examined. When the second injury occurred eight weeks after the initial injury and the lesion site was treated with CNTF, neuron survival four weeks later (12 weeks after the initial injury) in both the RN and LVN was comparable to that recorded following CNTF treatment of a second lesion four weeks after the initial injury (Fig. 3A, B; Table 3). That is, Table 2. Changes in neuron survival in the red nucleus and lateral vestibular nucleus after a second injury and trophic factor treatment Group
Eight weeks post C3Hx PBS four weeks post C3Hx, killed four weeks later BDNF four weeks post C3Hx, killed four weeks later CNTF four weeks post C3Hx, killed four weeks later FGF2 four weeks post C3Hx, killed four weeks later
Red nucleus
Lateral vestibular nucleus
269 ^ 43 118 ^ 14*
283 ^ 41 136 ^ 45*
197 ^ 55 1
146 ^ 36*
270 ^ 34 1
183 ^ 57
108 ^ 12*
103 ^ 25*
Values are mean cell number ^ S.D. *Indicates significant difference from eight weeks post C3Hx at P , 0.05. 1Indicates significant difference from PBS-treatment after a second injury at P , 0.05.
Fig. 3. Fluorescent micrograph of changes in cell survival in the LVN after a second injury at (A) four weeks or (B) eight weeks after injury and treatment with CNTF in each case. There is no significant difference in the number and distribution of neurons in the LVN at either time-point. These labeled neurons survive for at least four weeks after re-injury and CNTF treatment. Scale bar 200 mm.
Trophic factors promote long-term neuron survival Table 3. Changes in the ability to rescue neurons in the red nucleus and lateral vestibular nucleus by ciliary neurotrophic factor treatment at long post-injury periods Group
CNTF four weeks post C3Hx, killed four weeks later CNTF eight weeks post C3Hx, killed four weeks later CNTF 14 weeks post C3Hx, killed four weeks later CNTF 22 weeks post C3Hx, killed four weeks later
Red nucleus
Lateral vestibular nucleus
270 ^ 34
183 ^ 57
262 ^ 61
175 ^ 68
89 ^ 24*
103 ^ 8*
57 ^ 8*
57 ^ 6*
Values are mean cell number ^ S.D. *Indicates significant difference from a second injury and CNTF treatment at eight weeks post C3Hx, P , 0.05.
CNTF treatment significantly reduced the extent of neuron loss resulting from a second spinal cord injury, whether this second injury occurred at four or eight weeks after the initial injury. When the second injury was delayed for 14 or 22 weeks, treatment with CNTF was not effective in reducing neuronal loss in either the RN or LVN (Table 3), as there was a significant decrease in the number of neurons that survived, compared to those re-injured at eight weeks after the initial injury. Neurons of the LVN, however, did demonstrate a more robust nature at these longer post-injury periods than did the RN neurons. Over 60% of LVN neurons survived when CNTF was provided after a second injury 14 weeks after the initial injury, compared to 34% survival of RN neurons. Changes in cell size after spinal cord injury Following SCI there was no change in the CSA of the overall population of injured neurons within the RN between one and four weeks post-injury, although the percentage of large neurons was slightly increased at four weeks post-injury (Table 4). In the period between four and eight weeks postinjury there was a significant increase in the mean CSA for all injured RN neurons, the result of a decrease in the presence (survival) of small neurons and an increase in the representation of medium and large neurons (Fig. 2D, Table 4). This change in survival of different sized RN neurons between four and eight weeks post-injury was not influenced by a second injury and PBS treatment, although treatment with CNTF in this situation profoundly affected survival of different RN neurons. The mean CSA after a second injury and CNTF treatment decreased by 25% compared to PBS-treated neurons, as the percentage of surviving small neurons was
933
significantly increased (Fig. 2D, Table 4), approaching the percentage of small neurons present at four weeks post-injury. The effect of SCI on the size of LVN neurons was different from the RN neurons in that there was a significant decrease in overall CSA between one and four weeks post-injury (Table 5), with no further change between weeks 4 and 8. This decrease resulted from a sharp decline in the number of large neurons present at this interval. A second injury at four weeks after the initial injury, followed by PBS treatment had no detectable effect on overall LVN neuron size, although the distribution of small- and medium-sized neurons was reversed following reinjury, compared to the percentage found at eight weeks after a single injury (Table 5). Following a second injury, CNTF treatment led to a significant increase in mean cell size (Table 5) compared to PBS treatment. As a population, LVN neurons were restored to their size recorded at one week post-injury as a result of the increase in neurons with a CSA greater than 800 mm 2. DISCUSSION
This study addressed issues related to the chronicallyinjured spinal cord and the determination of appropriate post-injury intervals for the implementation of interventions designed to affect a structural repair. Our previous reports on the effects of growth and trophic factors on axonal regeneration by chronically-injured neurons indicated that consideration be given to the possibility that certain features of injured neurons continue to change during the post-injury period. Several necessary manipulations of the spinal cord which affect brainstem neurons are made during the procedures designed to promote regeneration, and it was not clear how such manipulations might influence the overall regenerative effort of chronically-injured neurons. Accordingly, this study examined: (i) the survival of two different populations of brainstem neurons at post-injury periods up to eight weeks, (ii) the effects of a second injury at different times after the initial injury on neuron survival and the potential for neurotrophic or growth factors to rescue re-injured neurons, and (iii) whether chronically-injured neurons could sustain the ability to respond to neurotrophic factors for post-injury periods as long as 26 weeks. Neuron survival after spinal cord injury The study of neuron survival at sub-chronic and chronic post-injury time-points was carried out to determine how stable the injured cell population might be after SCI. No change in the number of LVN neurons from one to eight
Table 4. Changes in red nucleus neuron size after spinal cord injury Group
One week post C3Hx Four weeks post C3Hx Eight weeks post C3Hx PBS four weeks post C3Hx, killed four weeks later CNTF four weeks post C3Hx, killed four weeks later
Mean cell size (mm 2)
Soma size distribution (%) , 600 mm 2
600–900 mm 2
. 900 mm 2
445 ^ 57 468 ^ 37 651 ^ 104* 637 ^ 30*
82 ^ 14 83 ^ 9 51 ^ 8* 48 ^ 11*
17 ^ 13 12 ^ 7 27 ^ 5 30 ^ 8*
2^1 5^3 21 ^ 13* 22 ^ 10*
483 ^ 106
78 ^ 12
12 ^ 5
10 ^ 7
Values are mean ^ S.D. *Indicates significant difference from one week post C3Hx at P , 0.05.
934
J. D. Houle and J.-H. Ye Table 5. Changes in lateral vestibular nucleus neuron size after spinal cord injury
Group
One week post C3Hx Four weeks post C3Hx Eight weeks post C3Hx PBS four weeks post C3Hx, killed four weeks later CNTF four weeks post C3Hx, killed four weeks later
Mean cell (mm 2)
Soma size distribution (%) , 400 mm 2
400–800 mm 2
. 800 mm 2
654 ^ 74 475 ^ 42* 509 ^ 95 520 ^ 45
24 ^ 9 44 ^ 15 43 ^ 11 30 ^ 2
46 ^ 3 48 ^ 14 40 ^ 7 53 ^ 10
30 ^ 10 8 ^ 1* 17 ^ 16 17 ^ 9
651 ^ 28**
15 ^ 5
60 ^ 3
25 ^ 2
Values are mean ^ S.D. *Indicates significant difference from one week post C3Hx at P , 0.05. **Indicates significant difference from PBS treatment at P , 0.05.
weeks after injury indicated that a fairly uniform population of neurons would be available for therapeutic intervention at any time during this period. Our data agree with the observations of Jin et al. 18 that LVN neuron survival does not change significantly between two and six weeks after a cervical lesion. In contrast, the number of RN neurons remained stable for four weeks, but there was a significant decline in survival when examined eight weeks after SCI. This observation suggests a delayed secondary response to injury that is severe enough to lead to cell death and further, that an effective intervention applied at four weeks after injury may not be as effective if applied at a later post-injury period. Other studies have examined the effects of SCI at fixed intervals after injury, without giving consideration to the possibility that injured neurons may be unstable after these short or long intervals. Reports of RN neuron response to injury indicate that most neurons survive for four to five weeks 9,37 and that there is little difference in the survival of RN neurons at eight or 16 weeks after a C3Hx. 24 It is possible that the second wave of cell death that we have detected between four and eight weeks after SCI subsides, such that no further decrease in survival would be observed, although we do not have data to support this suggestion. One important aspect to be aware of in many of these studies involving the neuronal response to injury is that the level of spinal cord injury greatly influences the severity of the response. Clearly, a mid-thoracic injury results in a less dramatic increase in expression of regeneration-associated genes and relatively little cell death of RN neurons compared to an upper cervical injury. 30,31 One of the more important findings of this study was that the additional loss of neurons following re-injury of the spinal cord after long post-injury periods (at least eight weeks), could be prevented by providing CNTF (or BDNF in the case of RN neurons) to the lesion site. Furthermore, the effects of these factors were not transient, in that neuronal survival was evident for at least four weeks and survival was observed after factors were provided at the lesion site over a short timeperiod (involving several changes of factor-saturated gel foam over 1 h). Although it is not known how long these particular factors may remain present and active at the lesion site, NT-3 applied on saturated gel foam to the injured spinal cord will retain biological activity for at least five days. 29 The relatively short period of application of fresh factor in our study is in contrast to studies in which trophic factors have been infused for several weeks into the nuclear region of axotomized neurons. 21 Extended survival of spinal motoneurons has been shown following a four-week infusion of BDNF into
the subarachnoid space. 26 Whether survival of chronicallyinjured neurons can be enhanced and extended for longer periods of time by increasing the time of exposure to trophic factors remains to be determined. Because some neurotrophic factors appear to have only a transient effect on neuron survival 7,8 after axotomy, there is the possibility that the capacity to respond to certain environmental conditions diminishes as the post-injury interval increases. Several investigations have shown an enhanced and prolonged neuron survival and phenotypic expression after exposure to combinations of various neurotrophic factors. 2,5,13,19,25,34 The ability to reduce neuronal death after a second injury is significant because of the necessity to remove scar tissue from a lesion site in regeneration studies, for optimal apposition of a neural tissue transplant to the injured spinal cord tissue. While this second injury appears to be harmful to neurons in the absence of exogenous trophic factor support, it may be that the overall regenerative effort is enhanced by the combination of a second injury with trophic factor support. Injury to the peripheral process of sensory neurons is known to prime these cells so that their efforts to regrow after a second injury are more vigorous. 23 Whether CNS neurons are primed to exhibit a similar heightened response after SCI is not known, however, it is clear that chronically-injured neurons subjected to a second lesion require trophic factor support, not only for axonal regeneration, 12,17,36 but also for neuron survival as indicated in the present study. Re-injured neurons of the RN could be rescued by two different trophic factors, while the LVN were responsive to just CNTF. It is interesting that FGF2 was not effective in promoting cell survival in either region, given the fact that it did enhance axonal regeneration from both LVN and RN neurons when provided four weeks after injury. 17 This observation indicates separate mechanisms for regulation of the neuronal response to injury, i.e. cell survival likely is controlled differently than is axonal regrowth and different growth or trophic factors could have different effects when applied at different times after injury. It is possible that a change in axonal transport after injury 33 plays a role in the effectiveness of these different factors. There was a close correlation between the number of surviving neurons and the number of regenerating neurons in both RN and LVN. While this is not a surprising finding, it is interesting that the proportion of surviving chronicallyinjured neurons that exhibited regeneration did not change at any of the post-injury periods or after factor treatment. Regardless of the number of RN neurons present, only 8–10%
Trophic factors promote long-term neuron survival
were found to regenerate into a peripheral nerve (PN) graft, while in the LVN approximately 35–40% of surviving neurons regenerated their axon after treatment with CNTF. 36 This suggests a mechanism for regulating axonal regrowth that is not related directly to the number of neurons available or to the physical constraints of the PN graft (i.e. being too small to accommodate more axons). Up-regulation of mRNA for regeneration-associated genes, such as Ta1tubulin, actin and growth-associated protein 43, has been demonstrated following the administration of BDNF to the RN within one week after a spinal cord injury 31 or as long as one year after injury. 21 The number of RN neurons that regenerate their axon into a PN graft also is increased following infusion of BDNF into the RN. 21
935
by treatment with CNTF. The sensitivity of small neurons after a second injury resulted in an increase in overall CSA of RN neurons, which was not observed after CNTF treatment. Several studies have shown significant atrophy of axotomized RN neurons at short post-injury periods, which can be reversed or prevented by infusion of BDNF into the RN, 21,31 but not following intraspinal transplantation of fetal spinal cord tissue. 24 Because only TB-labeled cells were included in the present analysis, we are confident that our measurements accurately reflect the status of only injured neurons. The effect of CNTF on re-injured LVN neurons was similar to the results obtained with BDNF infusion into the RN. These results suggest that both cell survival and cell shrinkage should be considered in future studies of the neuronal response to injury and to trophic factor treatment.
Neuronal atrophy or death after spinal cord injury The quantification of neuron number and CSA provide a measure of the overall neuronal response to SCI and to exogenous trophic factors provided after a second injury. Changes in the CSA can be indicative of how different types of neurons (based upon size) within a defined nucleus respond to these different interventions. A decrease in overall CSA may reflect atrophy of injured neurons or death of a selective population of neurons, as suggested by the current study. No change in mean CSA of injured RN neurons was detected between one and four weeks post C3Hx, however, between four and eight weeks after injury there was a significant increase in CSA. Analysis of neurons within different size categories indicated that the percentage of small neurons decreased by 40% during this time period, with consequent increases in the percentage of medium and large neurons. Interestingly, small RN neurons appeared to be more sensitive to a second injury, the effects of which could be ameliorated
CONCLUSIONS
Overall, this study indicates that brainstem neurons injured by a spinal cord trauma may be reduced in number gradually during long post-injury periods. These chronically-injured neurons are very sensitive to a second injury, but their survival can be enhanced by exposure to specific neurotrophic factors, namely CNTF and BDNF. These observations provide an indication of the potential for interventive treatment with trophic factors long after the initial injury, such that the number of neurons available for regeneration can be increased, which is directly correlated with the number of regenerated axons detected within a peripheral nerve graft. Acknowledgements—We thank Regeneron/Amgen Partners for providing BDNF and CNTF for this study. Charles Perrin, Ginger Brown and Kelly Ball provided excellent technical assistance. This work was supported by NIH grant NS26380.
REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
Agarwala S. and Kalil R. E. (1998) Long-term protection of axotomized neurons in the dorsal lateral geniculate nucleus in the rat following a single administration of basic fibroblast growth factor or ciliary neurotrophic factor. J. comp. Neurol. 392, 264–272. Arakawa Y., Sendtner M. and Thoenen H. (1990) Survival effect of ciliary neurotrophic factor (CNTF) on chick embryonic motoneurons in culture: comparison with other neurotrophic factors and cytokines. J. Neurosci. 10, 3507–3515. Barron K. D., Banerjee M., Dentinger M. P., Scheibly M. E. and Mankes R. (1989) Cytological and cytochemical (RNA) studies on rubral neurons after unilateral rubrospinal tractotomy: the impact of GM1 ganglioside administration. J. Neurosci. Res. 22, 331–337. Bregman B. S. and Reier P. J. (1986) Neural tissue transplants rescue axotomized rubrospinal cells from retrograde death. J. comp. Neurol. 244, 86–95. Bregman B. S., McAtee M., Dai H. N. and Kuhn P. L. (1997) Neurotrophic factors increase axonal growth after spinal cord injury and transplantation in the adult rat. Expl Neurol. 148, 475–494. Dale S. M., Kuang R. Z., Wei X. and Varon S. (1995) Corticospinal motor neurons in the adult rat: degeneration after intracortical axotomy and protection by ciliary neurotrophic factor (CNTF). Expl Neurol. 135, 667–673. Diener P. S. and Bregman B. S. (1994) Neurotrophic factors prevent the death of CNS neurons after spinal cord lesions in newborn rats. NeuroReport 5, 1913–1917. Eriksson N. P., Lindsay R. M. and Aldskogius H. (1994) BDNF and NT-3 rescue sensory but not motoneurones following axotomy in the neonate. NeuroReport 5, 1445–1448. Feringa E. R., McBride R. L. and Pruitt J. N. (1988) Loss of neurons in the red nucleus after spinal cord transection. Expl Neurol. 100, 112–120. Giehl K. M., Schacht C. M., Yan Q. and Mestres P. (1997) GDNF is a trophic factor for adult rat corticospinal neurons and promotes their long-term survival after axotomy in vivo. Eur. J. Neurol. 9, 2479–2489. Goshgarian H. G., Koistinen J. M. and Schmidt E. R. (1983) Cell death and changes in the retrograde transport of horseradish peroxidase in rubrospinal neurons following spinal cord hemisection in the adult rat. J. comp. Neurol. 214, 251–257. Grill R., Blesch A. and Tuszynski M. H. (1997) Robust growth of chronically injured spinal cord axons induced by grafts of genetically modified NGFsecreting cells. Expl Neurol. 148, 444–452. Himes B. T., Goldberger M. E. and Tessler A. (1994) Grafts of fetal central nervous system tissue rescue axotomized Clarke’s nucleus neurons in adult and neonatal operates. J. comp. Neurol. 339, 117–131. Houle J. D. and Reier P. J. (1988) Transplantation of fetal spinal cord tissue into the chronically injured rat spinal cord. J. comp. Neurol. 269, 535–547. Houle J. D. (1991) Demonstration of the potential for chronically injured neurons to regenerate axons into intraspinal peripheral nerve grafts. Expl Neurol. 113, 1–9. Houle J. D. and Ye J. H. (1997) Changes occur in the ability to promote axonal regeneration as the post-injury period increases. NeuroReport 8, 751–755. Houle J. D., Ye J. H. and Kane C. J. M. (1997) Axonal regeneration by chronically injured supraspinal neurons can be enhanced by exposure to insulinlike growth factor, basic fibroblast growth factor, or transforming growth factor beta. Restor. Neurol. Neurosci. 10, 205–215. Jin Y., Tay D. and So K.-F. (1997) The effect of distance of axotomy in the spinal cord on NOS expression and neuronal survival in the lateral vestibular nucleus of rats. Soc. Neurosci. Abstr. 23, 899.
936
J. D. Houle and J.-H. Ye
19. Kato A. C. and Lindsay R. M. (1994) Overlapping and additive effects of neurotrophins and CNTF on cultured human spinal cord neurons. Expl Neurol. 130, 196–201. 20. Klo¨cker N., Bra¨unling F., Isenmann S. and Ba¨hr M. (1997) In vivo neurotrophic effects of GDNF on axotomized retinal ganglion cells. NeuroReport 8, 3439–3442. 21. Kobayashi N. R., Fan D.-P., Giehl K. M., Bedard A. M., Wiegand S. J. and Tetzlaff W. (1997) BDNF and NT-4/5 prevent atrophy of rat rubrospinal neurons after cervical axotomy, stimulate GAP-43 and Ta1-tubulin mRNA expression, and promote axonal regeneration. J. Neurosci. 17, 9583–9595. 22. Mansour-Robaey S., Clarke D. B., Wang Y.-C., Bray G. M. and Aguayo A. J. (1994) Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells. Proc. natn. Acad. Sci. U.S.A. 91, 1632–1636. 23. McQuarrie I. G. and Grafstein B. (1981) Effect of a conditioning lesion in optic nerve regeneration in goldfish. Brain Res. 216, 253–264. 24. Mori F., Himes B. T., Kowada M., Murray M. and Tessler A. (1997) Fetal spinal cord transplants rescue some axotomized rubrospinal neurons from retrograde cell death in adult rats. Expl Neurol. 143, 45–60. 25. Nishi R. (1994) Neurotrophic factors: two are better than one. Science 265, 1052–1053. 26. Novikov L., Novikova L. and Kellerth J.-O. (1997) Brain-derived neurotrophic factor promotes axonal regeneration and long-term survival of adult rat spinal motoneurons in vivo. Neuroscience 79, 765–774. 27. Peinado-Ramon P., Salvador M., Villegas-Perez M. P. and Vidal-Sanz M. (1996) Effects of axotomy and intraocular administration of NT-4, NT-3 and brain-derived neurotrophic factor on the survival of adult rat retinal ganglion cells. Invest. Ophthal. vis. Sci. 37, 489–500. 28. Prendergast J. and Stelzner D. J. (1976) Changes in the magnocellular portion of the red nucleus following thoracic hemisection in the neonatal and adult rat. J. comp. Neurol. 166, 163–172. 29. Shibiyama M., Hattori S., Himes B. T., Murray M. and Tessler A. (1998) Neurotrophin-3 prevents death of axotomized Clarke’s nucleus neurons in adult rat. J. comp. Neurol. 390, 102–111. 30. Tetzlaff W., Alexander S. W., Miller F. D. and Busby M. A. (1991) Response of facial and rubrospinal neurons to axotomy: changes in mRNA expression for cytoskeletal proteins and GAP-43. J. Neurosci. 8, 3181–3189. 31. Tetzlaff W., Kobayashi N. R., Giehl K. M. G., Tsui B. J., Cassar S. L. and Bedard A. M. (1994) Response of rubrospinal and corticospinal neurons to injury and neurotrophins. Prog. Brain Res. 103, 271–286. 32. Theriault E. and Tator C. H. (1994) Persistence of rubrospinal projections following spinal cord injury in the rat. J. comp. Neurol. 342, 249–258. 33. Tseng G.-F., Shu J., Huang S.-J. and Wang Y.-J. (1995) A time dependent loss of retrograde transport ability in distally axotomized rubrospinal neurons. Anat. Embryol. 191, 243–249. 34. Vejsada R., Sagot Y. and Kato A. C. (1995) Quantitative comparison of the transient rescue effects of neurotrophic factors on axotomized motoneurons in vivo. Eur. J. Neurosci. 7, 108–115. 35. Villegas-Perez M. P., Vidal-Sanz M., Rasminsky M., Bray G. M. and Aguayo A. J. (1993) Rapid and protracted phases of retinal ganglion cell loss follow axotomy in the optic nerve of adult rats. J. Neurobiol. 24, 23–36. 36. Ye J. H. and Houle J. D. (1997) Treatment of the chronically injured spinal cord with neurotrophic factors can promote axonal regeneration from supraspinal neurons. Expl Neurol. 143, 70–81. 37. Xu X. M. and Martin G. F. (1990) The response of rubrospinal neurons to axotomy in the adult opossum Didelphis virginiana. Expl Neurol. 108, 46–54. (Accepted 29 June 1999)
Trophic factors promote long-term neuron survival
Pergamon PII: S0306-4522(99)00359-0
SURVIVAL OF CHRONICALLY-INJURED NEURONS CAN BE PROLONGED BY TREATMENT WITH NEUROTROPHIC FACTORS J. D. HOULE* and J.-H. YE Department of Anatomy, University of Arkansas for Medical Sciences, 4301 West Markham Street, Little Rock, AR 72205, U.S.A.
Abstract—Axonal regeneration by chronically-injured supraspinal neurons can be enhanced by neurotrophic factor treatment at the site of injury, although the number of regenerating neurons decreases as the interval between spinal cord injury and treatment increases. This study investigated whether this decline in regenerative response could be due to continued loss of neurons during the post-injury period. Adult rats received a cervical hemisection lesion and axotomized neurons were labeled by retrograde transport of True Blue from the lesion site. Animals were killed one, four or eight weeks after injury and surviving neurons (True Bluelabeled) were counted in the red nucleus and lateral vestibular nucleus. The neuron number in the lateral vestibular nucleus was stable for eight weeks after spinal cord injury, while survival in the red nucleus decreased by 25% between four and eight weeks. To test how neurons respond to a second injury with or without trophic factor treatment, at four, eight, 14 or 22 weeks after injury the lesion cavity was enlarged by 0.5 mm in a rostral direction. Gel foam saturated with ciliary neurotrophic factor, brain-derived neurotrophic factor or basic fibroblast growth factor was placed into the cavity. Animals were killed four weeks later. Re-injury of the spinal cord caused a significant decrease in neuron survival in both the red nucleus and lateral vestibular nucleus, the effects of which were lessened by treatment with ciliary neurotrophic factor or brain-derived neurotrophic factor for the red nucleus and with ciliary neurotrophic factor for the lateral vestibular nucleus, when re-injured at four or eight weeks. Basic fibroblast growth factor did not affect neuron survival at any time post-injury. Ciliary neurotrophic factor was not effective with longer delays (14 or 22 weeks) between the initial injury and re-injury. These results indicate a delayed pattern of secondary neuronal cell loss after spinal cord injury that is exaggerated by re-injury, but which can be ameliorated by treatment with neurotrophic factors. q 1999 IBRO. Published by Elsevier Science Ltd. Key words: spinal cord injury, transplantation, CNTF, BDNF, FGF2, axotomy.
approximately 40% of these cells, 9,11,24,28 although there is some dispute as to whether there is severe atrophy exhibited by many of these neurons, which is miscalculated as cell death. 30 Fetal spinal cord tissue transplants 4,24 can affectively rescue dying RN neurons and acute treatment with neurotrophic factors 21 can prevent atrophy of adult RN neurons and promote survival of injured neonatal RN neurons. 7 In contrast to the RN, little is known about atrophy and/or cell death of other brainstem neurons following SCI. One aim of this study was to examine the temporal effects of SCI on another major contributor to descending spinal pathways, the lateral vestibular nucleus (LVN). A second aim was to determine how RN and LVN neurons would respond to the type of second injury that occurs routinely during removal of scar tissue and expansion of the size of the lesion cavity as the original lesion site is prepared for a neural tissue transplant. 14,15 Following injury to the adult rat spinal cord there is the potential for axonal regrowth by brainstem neurons that contribute to most descending spinal pathways, provided that a suitable environment (e.g., fetal neural tissue, peripheral nerve graft, Schwann cell graft, etc.) for growth is provided. Many investigations have demonstrated that these injured neurons can respond to specific trophic or growth factors, such as ciliary neurotrophic factor (CNTF) or basic fibroblast growth factor (FGF2), leading to an enhanced axonal regrowth. 12,17,36 When there is a delay between the time of injury and when the chosen substratum is made available, as in a chronic injury situation, the regenerative effort diminishes in conjunction with increasing delays. 15 Similarly, with a long delay between injury and treatment with trophic or growth factors, the regenerative effort appears to decline. 16
Previous studies have shown the effective use of neurotrophic and growth factors for the rescue of injured neurons in several different systems. Retinal ganglion neurons are extremely sensitive to optic nerve injury, with over 75% of the neurons dying within two weeks. 35 Intraocular treatment with brainderived neurotrophic factor (BDNF) or neurotrophin (NT)-4 can delay cell death for a short period, 22,27 however, a decline in survival is observed one to two weeks after treatment. Similar short-term benefits have been obtained with treatment with glial cell-derived neurotrophic factor. 20 Long-term benefits to a single treatment with growth or trophic factor after injury have been shown for neurons of Clarke’s nucleus 29 and neurons of the lateral geniculate nucleus. 1 Other neurons, such as those forming the corticospinal tract appear to survive for extended periods if trophic factors are provided for several weeks via a cannula placed into the cerebral cortex. 6,10 Injury to the spinal cord causes disruption of descending pathways from brainstem and cortical neurons and initiates a neuronal response which can lead to atrophy or death of the injured cells. 3,31,32 Neurons of the red nucleus (RN) in neonatal animals appear to be very sensitive to spinal cord injury (SCI), as over 55% die after a midthoracic injury. 4 Axotomy of adult rubrospinal neurons is thought to result in death of *To whom correspondence should be addressed. Tel.: 1 1-501-6865849; fax: 1 1-501-6866382. E-mail address: [email protected] (J. D. Houle) Abbreviations: BDNF, brain-derived neurotrophic factor; C, cervical; C3Hx, cervical level 3 hemisection; CNTF, ciliary neurotrophic factor; CSA, cross-sectional area; FGF2, basic fibroblast growth factor; Hx, hemisection; LVN, lateral vestibular nucleus; NT, neurotrophin; PBS, phosphate-buffered saline; PN, peripheral nerve; RN, red nucleus; SCI, spinal cord injury; TB, True Blue. 929
930
J. D. Houle and J.-H. Ye
This observation suggests that injured neurons probably continue to change during the post-injury period, such that an effective intervention strategy at one post-injury interval might no longer be effective after a longer interval before intervention. For effective repair of the chronically-injured spinal cord it is necessary to determine how different parameters of the neuronal response to injury might change with increasing post-injury time and how such changes may be affected by trophic factor support. The present study provides information about long-term neuron survival after SCI which can be incorporated into the development of strategies for intervention to be used at different post-injury periods. The decision to test three different trophic/growth factors was based upon studies 17,36 of the chronically injured spinal cord in which these specific factors (BDNF, CNTF and FGF2) demonstrated the ability to enhance axonal regrowth from a variety of supraspinal neuron groups. EXPERIMENTAL PROCEDURES
Surgical procedures and trophic factor treatment Adult female Sprague–Dawley rats (225–250 g, Charles River) were anesthetized with a mixture of ketamine (60 mg/kg) and xylazine (10 mg/kg). A laminectomy was performed on the second and third cervical (C) vertebrae, the meningeal membranes were excised along the midline and a hemisection (Hx) lesion cavity was created by aspiration of the C3 spinal cord segment. The cavity was approximately 3 mm in length and was complete to the lateral and ventral boundaries formed by the meninges. Upon hemostasis, gel foam saturated with a 2% solution of True Blue (TB, Sigma) was placed into the cavity for 60 min to label injured neurons by retrograde transport of this fluorescent dye. After removing the gel foam the cavity was rinsed with saline and the dorsal surface of the cavity was covered with a piece of silastic film inserted beneath the meninges. The dura mater was sutured and overlying muscle and skin closed with wound clips. Penicillin G (3000 units s.c., every third day for one week) and Buprenorphine (0.1 mg/kg i.m., once daily as needed for analgesia) were provided. Efforts were made to minimize animal suffering and all procedures were in accordance with United States Public Health Service Guidelines for the Care and Use of Laboratory Animals and with approval of the Institutional Animal Care and Use Committee. There were three main experimental groups for this study. (1) Spinal cord-injured rats were allowed to survive for one (n 3), four (n 4) or eight (n 4) weeks with no further manipulation of the injury site. These subgroups were used to determine possible changes in neuron survival at progressively longer post-injury periods. (2) After a C3Hx, the lesion site was exposed four weeks later, scar tissue was removed by aspiration and the cavity was expanded by 0.5 mm in a rostral direction. Gel foam saturated with BDNF (50 mg/ml, Regeneron/ Amgen Partners, n 7), CNTF (100 mg/ml, Regeneron Pharmaceuticals, n 4), FGF2 (1 mg/ml, Collaborative Research, n 5) or phosphate-buffered saline (PBS, n 4) was placed into the cavity. After 5 min the gel foam was removed and replaced with fresh factor-saturated gel foam every 15 min over a 60 min period. Approximately 5 ml of factor was provided with each change of gel foam. Factor-saturated gel foam was left in the cavity after the final change, the cavity was covered with silastic film and the dura was sutured. Animals were maintained for four weeks with no further manipulation. The doses of factors used for this study were those previously found to be effective in promoting axonal regeneration by chronically-injured neurons. 16,17,36 Animals in this subgroup were used to determine the effects of a second injury on neuronal survival and to test whether trophic and/or growth factors would influence survival of chronically injured neurons after a second injury. Removal of scar tissue and reinjury is a necessary step to cavity preparation for a peripheral nerve graft in axonal regeneration studies. 36 (3) After a C3HX, the lesion site was exposed at eight (n 3), 14 (n 3) or 22 (n 3) weeks after spinal cord injury, scar tissue was removed by aspiration and the cavity expanded in a rostral direction by 0.5 mm. Gel foam saturated with CNTF (100 mg/ml) was placed in the cavity as described above. Animals were maintained for four weeks with no additional
manipulation. These subgroups were used to determine neuronal survival following a second injury and trophic factor treatment at postinjury periods longer than four weeks. At the appropriate post-injury or post-treatment period, animals were killed by administration of an overdose of Nembutal (120 mg/ kg) followed by cardiac perfusion with a 4% paraformaldehyde solution. Brains were removed, postfixed with 4% paraformaldehyde overnight at 48C and then immersed in 20% sucrose at 48C. Tissue preparation Tissue blocks were prepared from brainstem regions containing the red nucleus or the vestibular nuclear complex. Two sets of alternating sections (25-mm-thick) were prepared using a cryostat and mounted separately in serial order on glass slides. One set of sections was coverslipped with Fluoromount (Bio/medical Specialties, Santa Monica, CA) and used for detection and analysis of TB labeled neurons within the RN and LVN. The second set of sections was stained with Thionin, coverslipped with Permount (Fisher Scientific) and used for confirmation of nuclear regions used for analysis. For identification of neurons in the RN, the caudal most section through the nucleus magnocellularis where TB labeled neurons could be observed was designated as the first section for analysis, with succeeding sections more rostral to the first. For identification of neurons in the LVN, the caudal-most section where TB labeled neurons were found within the inferior cerebellar peduncle ipsilateral to the lesion was designated as the first section for analysis, with succeeding sections more rostral to the first. The small number of TB labeled neurons in the contralateral LVN were not included in this quantitation. Analysis All neurons labeled with True Blue were counted in the first ten sections (spaced 25 mm apart from each other) through the RN and LVN, to provide a measure of the number of injured neurons that survived for varying periods or after exposure to different trophic or growth factors. Serial sections were compared to insure that duplicate counts of labeled neurons were not made. Labeled neurons were easily distinguished from non-labeled neurons using an ultraviolet light filter package on a Zeiss Axioskop fluorescent microscope. The entire cell soma and short portions of several processes were labeled with TB, giving the cells a characteristic neuronal appearance. Any nonneuronal cells labeled with TB were easily distinguished from neurons based upon their smaller cell size and lack of labeled processes. Stereological counting methods were not used because we were interested in determining the absolute number of neurons present at different time intervals following spinal cord injury. Because the same number of sections were used for quantitative analysis, covering the same caudal to rostral extent of each nuclear region, no corrections to cell counts were made. A one-way analysis of variance was used to test for statistically significant differences within the three large groups of experimental animals for the mean number of TB-labeled neurons. When significant differences were present, individual post hoc comparisons were made with the Bonferroni t-test (P , 0.05). Differences between different experimental groups were determined by a Student’s t-test (P , 0.05). The results from subgroups in experimental group 2 that were treated with trophic or growth factors were compared to the eight week survival, no re-injury subgroup of experimental group 1, because of their common overall length of animal survival. The results from subgroups in experimental group 3 treated with CNTF at different postinjury intervals were compared to the four week re-injury, CNTF treated subgroup of experimental group 2, because of their common use of CNTF as a trophic agent after a second injury. The cross-sectional area (CSA) of surviving neurons at different post-injury periods (experimental group 1) or after a second injury (experimental group 2) was measured from fluorescent photomicrographs with the NIH Image Program. All TB labeled neurons on three sections of the RN (from rostral, medial and caudal portions) from each animal were measured. Neurons of the RN were classified as small (,600 mm 2), medium (600–900 mm 2) or large (.900 mm 2), based upon the size distribution of TB-labeled RN neurons from a previous study. 15 Neurons of the LVN were classified as small (,400 mm 2), medium (400–800 mm 2) or large (.800 mm 2). Significant differences in the percentage of neurons within each size category were determined by an analysis of variance, with individual post hoc comparisons made with the Bonferroni t-test (P , 0.05).
Trophic factors promote long-term neuron survival
931 RESULTS
Change in neuron number and size one to eight weeks after spinal cord injury Neurons within the red nucleus RN and lateral vestibular nucleus that had been injured during preparation of a C3Hx lesion contained the fluorescent dye True Blue (TB) within their cell soma (Figs 1A, 3A). Based upon cell size and spread of the dye that labeled short portions of contiguous cell processes, injured neurons easily were distinguished from surrounding non-neuronal cells. Because the aim of this study was to examine long-term cell survival and the effects of neurotrophic or growth factors on injured neurons only, we included in our examination only those neurons that contained TB. One week after injury, TB labeled neurons were observed throughout the caudal to rostral regions of the RN and LVN that were chosen for study. The number of labeled cells at one week after injury represents those neurons that survived the initial trauma inflicted by SCI and is the maximum number of cells that might be present at any longer post-injury interval or after trophic factor treatment. In the 10 tissue sections examined from each nuclear region, there were 381 ^ 78 RN neurons and 267 ^ 37 LVN neurons labeled with TB (Table 1). Four weeks after injury there was no change in the number of neurons in the RN (Fig. 1B) or LVN compared to those present at one week. There was a significant decrease (to 75% of the number at four weeks post-injury, P , 0.05) in the number of RN neurons that survived for eight weeks after SCI (Fig. 1C, Table 1), whereas, there was no significant change in the number of surviving LVN neurons, compared to those at four weeks post-injury. Effects of a second injury and trophic factor treatment
Fig. 1. Fluorescence micrographs of True Blue-labeled neurons in the red nucleus following SCI at (A) one, (B) four or (C) eight weeks prior to sacrifice. Most labeled neurons are located in the ventral-lateral portion of the magnocellular nucleus of the RN, with no detectable decrease in number between one and four weeks after injury, compared to an obvious decrease in survival of labeled neurons at eight weeks after injury. Scale bar 200 mm. Table 1. Changes in neuron survival in the red nucleus and lateral vestibular nucleus between one and eight weeks after spinal cord injury Group
Red nucleus
Lateral vestibular nucleus
One week post C3Hx Four weeks post C3Hx Eight weeks post C3Hx
381 ^ 78 358 ^ 48 269 ^ 43*
267 ^ 37 239 ^ 15 283 ^ 41
Values are mean cell number ^ S.D. *Indicates significant difference from four weeks post C3Hx at P , 0.05.
Following the removal of scar tissue from the rostral cavity wall and expansion of the cavity in this direction, animals were allowed to survive for four weeks, which was eight weeks after the initial injury and comparable to the longest post-injury period in the previous group of animals. Without a second injury, 75% of the RN neurons survived for eight weeks as did 100% of the injured LVN neurons (see above). Following a second injury and treatment of the lesion site with PBS, there was a significant decrease in survival in both the RN (44% of those present eight weeks after injury) and LVN (48% of those present eight weeks after injury) (Fig. 2A, Table 2). When BDNF was provided after the second injury, over 73% of previously injured RN neurons survived, which was a significant increase over the PBS-treated group (Table 2). In contrast, BDNF treatment did not have a noticeable effect upon the survival of LVN neurons, with 52% survival, compared to 48% after PBS treatment. CNTF treatment after a second injury provided even greater protection to RN neurons, as the same number survived a second injury as did those surviving a single injury (Fig. 2C, D; Table 2). Survival was significantly increased compared to PBS treatment and was improved by 37% compared to BDNF treatment after a second injury. In the LVN, neuron survival after a second injury and CNTF treatment was 65% of that found eight weeks after a single injury, which was not a significant decrease from the single injury, eight week survival group. Providing FGF2 to the lesion site after a second injury had no effect on neuron survival in either the RN or LVN (Fig. 2B),
932
J. D. Houle and J.-H. Ye
Fig. 2. Fluorescent micrograph depicting changes in neuron survival in the RN, after (A) a second injury at four weeks after the initial injury and treatment with PBS or after treatment with (B) FGF2 or (C, D) CNTF. Treatment with PBS or FGF2 failed to rescue RN neurons, compared to the effects of CNTF, where significantly more chronically-injured neurons remained at four weeks after a second injury. There is no obvious atrophy of True Blue-labeled neurons treated with CNTF after a second injury. RN neurons of all sizes are present in this region. Scale bar 200 mm.
as the number of TB-labeled cells was comparable to that found after PBS treatment, which was significantly lower than the number that had survived for eight weeks after only a single injury. Effects of ciliary neurotrophic factor at longer post-injury periods Because only CNTF had a significant effect on neuron survival after a delayed second injury to RN and LVN neurons, only CNTF treatment at longer post-injury periods was examined. When the second injury occurred eight weeks after the initial injury and the lesion site was treated with CNTF, neuron survival four weeks later (12 weeks after the initial injury) in both the RN and LVN was comparable to that recorded following CNTF treatment of a second lesion four weeks after the initial injury (Fig. 3A, B; Table 3). That is, Table 2. Changes in neuron survival in the red nucleus and lateral vestibular nucleus after a second injury and trophic factor treatment Group
Eight weeks post C3Hx PBS four weeks post C3Hx, killed four weeks later BDNF four weeks post C3Hx, killed four weeks later CNTF four weeks post C3Hx, killed four weeks later FGF2 four weeks post C3Hx, killed four weeks later
Red nucleus
Lateral vestibular nucleus
269 ^ 43 118 ^ 14*
283 ^ 41 136 ^ 45*
197 ^ 55 1
146 ^ 36*
270 ^ 34 1
183 ^ 57
108 ^ 12*
103 ^ 25*
Values are mean cell number ^ S.D. *Indicates significant difference from eight weeks post C3Hx at P , 0.05. 1Indicates significant difference from PBS-treatment after a second injury at P , 0.05.
Fig. 3. Fluorescent micrograph of changes in cell survival in the LVN after a second injury at (A) four weeks or (B) eight weeks after injury and treatment with CNTF in each case. There is no significant difference in the number and distribution of neurons in the LVN at either time-point. These labeled neurons survive for at least four weeks after re-injury and CNTF treatment. Scale bar 200 mm.
Trophic factors promote long-term neuron survival Table 3. Changes in the ability to rescue neurons in the red nucleus and lateral vestibular nucleus by ciliary neurotrophic factor treatment at long post-injury periods Group
CNTF four weeks post C3Hx, killed four weeks later CNTF eight weeks post C3Hx, killed four weeks later CNTF 14 weeks post C3Hx, killed four weeks later CNTF 22 weeks post C3Hx, killed four weeks later
Red nucleus
Lateral vestibular nucleus
270 ^ 34
183 ^ 57
262 ^ 61
175 ^ 68
89 ^ 24*
103 ^ 8*
57 ^ 8*
57 ^ 6*
Values are mean cell number ^ S.D. *Indicates significant difference from a second injury and CNTF treatment at eight weeks post C3Hx, P , 0.05.
CNTF treatment significantly reduced the extent of neuron loss resulting from a second spinal cord injury, whether this second injury occurred at four or eight weeks after the initial injury. When the second injury was delayed for 14 or 22 weeks, treatment with CNTF was not effective in reducing neuronal loss in either the RN or LVN (Table 3), as there was a significant decrease in the number of neurons that survived, compared to those re-injured at eight weeks after the initial injury. Neurons of the LVN, however, did demonstrate a more robust nature at these longer post-injury periods than did the RN neurons. Over 60% of LVN neurons survived when CNTF was provided after a second injury 14 weeks after the initial injury, compared to 34% survival of RN neurons. Changes in cell size after spinal cord injury Following SCI there was no change in the CSA of the overall population of injured neurons within the RN between one and four weeks post-injury, although the percentage of large neurons was slightly increased at four weeks post-injury (Table 4). In the period between four and eight weeks postinjury there was a significant increase in the mean CSA for all injured RN neurons, the result of a decrease in the presence (survival) of small neurons and an increase in the representation of medium and large neurons (Fig. 2D, Table 4). This change in survival of different sized RN neurons between four and eight weeks post-injury was not influenced by a second injury and PBS treatment, although treatment with CNTF in this situation profoundly affected survival of different RN neurons. The mean CSA after a second injury and CNTF treatment decreased by 25% compared to PBS-treated neurons, as the percentage of surviving small neurons was
933
significantly increased (Fig. 2D, Table 4), approaching the percentage of small neurons present at four weeks post-injury. The effect of SCI on the size of LVN neurons was different from the RN neurons in that there was a significant decrease in overall CSA between one and four weeks post-injury (Table 5), with no further change between weeks 4 and 8. This decrease resulted from a sharp decline in the number of large neurons present at this interval. A second injury at four weeks after the initial injury, followed by PBS treatment had no detectable effect on overall LVN neuron size, although the distribution of small- and medium-sized neurons was reversed following reinjury, compared to the percentage found at eight weeks after a single injury (Table 5). Following a second injury, CNTF treatment led to a significant increase in mean cell size (Table 5) compared to PBS treatment. As a population, LVN neurons were restored to their size recorded at one week post-injury as a result of the increase in neurons with a CSA greater than 800 mm 2. DISCUSSION
This study addressed issues related to the chronicallyinjured spinal cord and the determination of appropriate post-injury intervals for the implementation of interventions designed to affect a structural repair. Our previous reports on the effects of growth and trophic factors on axonal regeneration by chronically-injured neurons indicated that consideration be given to the possibility that certain features of injured neurons continue to change during the post-injury period. Several necessary manipulations of the spinal cord which affect brainstem neurons are made during the procedures designed to promote regeneration, and it was not clear how such manipulations might influence the overall regenerative effort of chronically-injured neurons. Accordingly, this study examined: (i) the survival of two different populations of brainstem neurons at post-injury periods up to eight weeks, (ii) the effects of a second injury at different times after the initial injury on neuron survival and the potential for neurotrophic or growth factors to rescue re-injured neurons, and (iii) whether chronically-injured neurons could sustain the ability to respond to neurotrophic factors for post-injury periods as long as 26 weeks. Neuron survival after spinal cord injury The study of neuron survival at sub-chronic and chronic post-injury time-points was carried out to determine how stable the injured cell population might be after SCI. No change in the number of LVN neurons from one to eight
Table 4. Changes in red nucleus neuron size after spinal cord injury Group
One week post C3Hx Four weeks post C3Hx Eight weeks post C3Hx PBS four weeks post C3Hx, killed four weeks later CNTF four weeks post C3Hx, killed four weeks later
Mean cell size (mm 2)
Soma size distribution (%) , 600 mm 2
600–900 mm 2
. 900 mm 2
445 ^ 57 468 ^ 37 651 ^ 104* 637 ^ 30*
82 ^ 14 83 ^ 9 51 ^ 8* 48 ^ 11*
17 ^ 13 12 ^ 7 27 ^ 5 30 ^ 8*
2^1 5^3 21 ^ 13* 22 ^ 10*
483 ^ 106
78 ^ 12
12 ^ 5
10 ^ 7
Values are mean ^ S.D. *Indicates significant difference from one week post C3Hx at P , 0.05.
934
J. D. Houle and J.-H. Ye Table 5. Changes in lateral vestibular nucleus neuron size after spinal cord injury
Group
One week post C3Hx Four weeks post C3Hx Eight weeks post C3Hx PBS four weeks post C3Hx, killed four weeks later CNTF four weeks post C3Hx, killed four weeks later
Mean cell (mm 2)
Soma size distribution (%) , 400 mm 2
400–800 mm 2
. 800 mm 2
654 ^ 74 475 ^ 42* 509 ^ 95 520 ^ 45
24 ^ 9 44 ^ 15 43 ^ 11 30 ^ 2
46 ^ 3 48 ^ 14 40 ^ 7 53 ^ 10
30 ^ 10 8 ^ 1* 17 ^ 16 17 ^ 9
651 ^ 28**
15 ^ 5
60 ^ 3
25 ^ 2
Values are mean ^ S.D. *Indicates significant difference from one week post C3Hx at P , 0.05. **Indicates significant difference from PBS treatment at P , 0.05.
weeks after injury indicated that a fairly uniform population of neurons would be available for therapeutic intervention at any time during this period. Our data agree with the observations of Jin et al. 18 that LVN neuron survival does not change significantly between two and six weeks after a cervical lesion. In contrast, the number of RN neurons remained stable for four weeks, but there was a significant decline in survival when examined eight weeks after SCI. This observation suggests a delayed secondary response to injury that is severe enough to lead to cell death and further, that an effective intervention applied at four weeks after injury may not be as effective if applied at a later post-injury period. Other studies have examined the effects of SCI at fixed intervals after injury, without giving consideration to the possibility that injured neurons may be unstable after these short or long intervals. Reports of RN neuron response to injury indicate that most neurons survive for four to five weeks 9,37 and that there is little difference in the survival of RN neurons at eight or 16 weeks after a C3Hx. 24 It is possible that the second wave of cell death that we have detected between four and eight weeks after SCI subsides, such that no further decrease in survival would be observed, although we do not have data to support this suggestion. One important aspect to be aware of in many of these studies involving the neuronal response to injury is that the level of spinal cord injury greatly influences the severity of the response. Clearly, a mid-thoracic injury results in a less dramatic increase in expression of regeneration-associated genes and relatively little cell death of RN neurons compared to an upper cervical injury. 30,31 One of the more important findings of this study was that the additional loss of neurons following re-injury of the spinal cord after long post-injury periods (at least eight weeks), could be prevented by providing CNTF (or BDNF in the case of RN neurons) to the lesion site. Furthermore, the effects of these factors were not transient, in that neuronal survival was evident for at least four weeks and survival was observed after factors were provided at the lesion site over a short timeperiod (involving several changes of factor-saturated gel foam over 1 h). Although it is not known how long these particular factors may remain present and active at the lesion site, NT-3 applied on saturated gel foam to the injured spinal cord will retain biological activity for at least five days. 29 The relatively short period of application of fresh factor in our study is in contrast to studies in which trophic factors have been infused for several weeks into the nuclear region of axotomized neurons. 21 Extended survival of spinal motoneurons has been shown following a four-week infusion of BDNF into
the subarachnoid space. 26 Whether survival of chronicallyinjured neurons can be enhanced and extended for longer periods of time by increasing the time of exposure to trophic factors remains to be determined. Because some neurotrophic factors appear to have only a transient effect on neuron survival 7,8 after axotomy, there is the possibility that the capacity to respond to certain environmental conditions diminishes as the post-injury interval increases. Several investigations have shown an enhanced and prolonged neuron survival and phenotypic expression after exposure to combinations of various neurotrophic factors. 2,5,13,19,25,34 The ability to reduce neuronal death after a second injury is significant because of the necessity to remove scar tissue from a lesion site in regeneration studies, for optimal apposition of a neural tissue transplant to the injured spinal cord tissue. While this second injury appears to be harmful to neurons in the absence of exogenous trophic factor support, it may be that the overall regenerative effort is enhanced by the combination of a second injury with trophic factor support. Injury to the peripheral process of sensory neurons is known to prime these cells so that their efforts to regrow after a second injury are more vigorous. 23 Whether CNS neurons are primed to exhibit a similar heightened response after SCI is not known, however, it is clear that chronically-injured neurons subjected to a second lesion require trophic factor support, not only for axonal regeneration, 12,17,36 but also for neuron survival as indicated in the present study. Re-injured neurons of the RN could be rescued by two different trophic factors, while the LVN were responsive to just CNTF. It is interesting that FGF2 was not effective in promoting cell survival in either region, given the fact that it did enhance axonal regeneration from both LVN and RN neurons when provided four weeks after injury. 17 This observation indicates separate mechanisms for regulation of the neuronal response to injury, i.e. cell survival likely is controlled differently than is axonal regrowth and different growth or trophic factors could have different effects when applied at different times after injury. It is possible that a change in axonal transport after injury 33 plays a role in the effectiveness of these different factors. There was a close correlation between the number of surviving neurons and the number of regenerating neurons in both RN and LVN. While this is not a surprising finding, it is interesting that the proportion of surviving chronicallyinjured neurons that exhibited regeneration did not change at any of the post-injury periods or after factor treatment. Regardless of the number of RN neurons present, only 8–10%
Trophic factors promote long-term neuron survival
were found to regenerate into a peripheral nerve (PN) graft, while in the LVN approximately 35–40% of surviving neurons regenerated their axon after treatment with CNTF. 36 This suggests a mechanism for regulating axonal regrowth that is not related directly to the number of neurons available or to the physical constraints of the PN graft (i.e. being too small to accommodate more axons). Up-regulation of mRNA for regeneration-associated genes, such as Ta1tubulin, actin and growth-associated protein 43, has been demonstrated following the administration of BDNF to the RN within one week after a spinal cord injury 31 or as long as one year after injury. 21 The number of RN neurons that regenerate their axon into a PN graft also is increased following infusion of BDNF into the RN. 21
935
by treatment with CNTF. The sensitivity of small neurons after a second injury resulted in an increase in overall CSA of RN neurons, which was not observed after CNTF treatment. Several studies have shown significant atrophy of axotomized RN neurons at short post-injury periods, which can be reversed or prevented by infusion of BDNF into the RN, 21,31 but not following intraspinal transplantation of fetal spinal cord tissue. 24 Because only TB-labeled cells were included in the present analysis, we are confident that our measurements accurately reflect the status of only injured neurons. The effect of CNTF on re-injured LVN neurons was similar to the results obtained with BDNF infusion into the RN. These results suggest that both cell survival and cell shrinkage should be considered in future studies of the neuronal response to injury and to trophic factor treatment.
Neuronal atrophy or death after spinal cord injury The quantification of neuron number and CSA provide a measure of the overall neuronal response to SCI and to exogenous trophic factors provided after a second injury. Changes in the CSA can be indicative of how different types of neurons (based upon size) within a defined nucleus respond to these different interventions. A decrease in overall CSA may reflect atrophy of injured neurons or death of a selective population of neurons, as suggested by the current study. No change in mean CSA of injured RN neurons was detected between one and four weeks post C3Hx, however, between four and eight weeks after injury there was a significant increase in CSA. Analysis of neurons within different size categories indicated that the percentage of small neurons decreased by 40% during this time period, with consequent increases in the percentage of medium and large neurons. Interestingly, small RN neurons appeared to be more sensitive to a second injury, the effects of which could be ameliorated
CONCLUSIONS
Overall, this study indicates that brainstem neurons injured by a spinal cord trauma may be reduced in number gradually during long post-injury periods. These chronically-injured neurons are very sensitive to a second injury, but their survival can be enhanced by exposure to specific neurotrophic factors, namely CNTF and BDNF. These observations provide an indication of the potential for interventive treatment with trophic factors long after the initial injury, such that the number of neurons available for regeneration can be increased, which is directly correlated with the number of regenerated axons detected within a peripheral nerve graft. Acknowledgements—We thank Regeneron/Amgen Partners for providing BDNF and CNTF for this study. Charles Perrin, Ginger Brown and Kelly Ball provided excellent technical assistance. This work was supported by NIH grant NS26380.
REFERENCES
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
Agarwala S. and Kalil R. E. (1998) Long-term protection of axotomized neurons in the dorsal lateral geniculate nucleus in the rat following a single administration of basic fibroblast growth factor or ciliary neurotrophic factor. J. comp. Neurol. 392, 264–272. Arakawa Y., Sendtner M. and Thoenen H. (1990) Survival effect of ciliary neurotrophic factor (CNTF) on chick embryonic motoneurons in culture: comparison with other neurotrophic factors and cytokines. J. Neurosci. 10, 3507–3515. Barron K. D., Banerjee M., Dentinger M. P., Scheibly M. E. and Mankes R. (1989) Cytological and cytochemical (RNA) studies on rubral neurons after unilateral rubrospinal tractotomy: the impact of GM1 ganglioside administration. J. Neurosci. Res. 22, 331–337. Bregman B. S. and Reier P. J. (1986) Neural tissue transplants rescue axotomized rubrospinal cells from retrograde death. J. comp. Neurol. 244, 86–95. Bregman B. S., McAtee M., Dai H. N. and Kuhn P. L. (1997) Neurotrophic factors increase axonal growth after spinal cord injury and transplantation in the adult rat. Expl Neurol. 148, 475–494. Dale S. M., Kuang R. Z., Wei X. and Varon S. (1995) Corticospinal motor neurons in the adult rat: degeneration after intracortical axotomy and protection by ciliary neurotrophic factor (CNTF). Expl Neurol. 135, 667–673. Diener P. S. and Bregman B. S. (1994) Neurotrophic factors prevent the death of CNS neurons after spinal cord lesions in newborn rats. NeuroReport 5, 1913–1917. Eriksson N. P., Lindsay R. M. and Aldskogius H. (1994) BDNF and NT-3 rescue sensory but not motoneurones following axotomy in the neonate. NeuroReport 5, 1445–1448. Feringa E. R., McBride R. L. and Pruitt J. N. (1988) Loss of neurons in the red nucleus after spinal cord transection. Expl Neurol. 100, 112–120. Giehl K. M., Schacht C. M., Yan Q. and Mestres P. (1997) GDNF is a trophic factor for adult rat corticospinal neurons and promotes their long-term survival after axotomy in vivo. Eur. J. Neurol. 9, 2479–2489. Goshgarian H. G., Koistinen J. M. and Schmidt E. R. (1983) Cell death and changes in the retrograde transport of horseradish peroxidase in rubrospinal neurons following spinal cord hemisection in the adult rat. J. comp. Neurol. 214, 251–257. Grill R., Blesch A. and Tuszynski M. H. (1997) Robust growth of chronically injured spinal cord axons induced by grafts of genetically modified NGFsecreting cells. Expl Neurol. 148, 444–452. Himes B. T., Goldberger M. E. and Tessler A. (1994) Grafts of fetal central nervous system tissue rescue axotomized Clarke’s nucleus neurons in adult and neonatal operates. J. comp. Neurol. 339, 117–131. Houle J. D. and Reier P. J. (1988) Transplantation of fetal spinal cord tissue into the chronically injured rat spinal cord. J. comp. Neurol. 269, 535–547. Houle J. D. (1991) Demonstration of the potential for chronically injured neurons to regenerate axons into intraspinal peripheral nerve grafts. Expl Neurol. 113, 1–9. Houle J. D. and Ye J. H. (1997) Changes occur in the ability to promote axonal regeneration as the post-injury period increases. NeuroReport 8, 751–755. Houle J. D., Ye J. H. and Kane C. J. M. (1997) Axonal regeneration by chronically injured supraspinal neurons can be enhanced by exposure to insulinlike growth factor, basic fibroblast growth factor, or transforming growth factor beta. Restor. Neurol. Neurosci. 10, 205–215. Jin Y., Tay D. and So K.-F. (1997) The effect of distance of axotomy in the spinal cord on NOS expression and neuronal survival in the lateral vestibular nucleus of rats. Soc. Neurosci. Abstr. 23, 899.
936
J. D. Houle and J.-H. Ye
19. Kato A. C. and Lindsay R. M. (1994) Overlapping and additive effects of neurotrophins and CNTF on cultured human spinal cord neurons. Expl Neurol. 130, 196–201. 20. Klo¨cker N., Bra¨unling F., Isenmann S. and Ba¨hr M. (1997) In vivo neurotrophic effects of GDNF on axotomized retinal ganglion cells. NeuroReport 8, 3439–3442. 21. Kobayashi N. R., Fan D.-P., Giehl K. M., Bedard A. M., Wiegand S. J. and Tetzlaff W. (1997) BDNF and NT-4/5 prevent atrophy of rat rubrospinal neurons after cervical axotomy, stimulate GAP-43 and Ta1-tubulin mRNA expression, and promote axonal regeneration. J. Neurosci. 17, 9583–9595. 22. Mansour-Robaey S., Clarke D. B., Wang Y.-C., Bray G. M. and Aguayo A. J. (1994) Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells. Proc. natn. Acad. Sci. U.S.A. 91, 1632–1636. 23. McQuarrie I. G. and Grafstein B. (1981) Effect of a conditioning lesion in optic nerve regeneration in goldfish. Brain Res. 216, 253–264. 24. Mori F., Himes B. T., Kowada M., Murray M. and Tessler A. (1997) Fetal spinal cord transplants rescue some axotomized rubrospinal neurons from retrograde cell death in adult rats. Expl Neurol. 143, 45–60. 25. Nishi R. (1994) Neurotrophic factors: two are better than one. Science 265, 1052–1053. 26. Novikov L., Novikova L. and Kellerth J.-O. (1997) Brain-derived neurotrophic factor promotes axonal regeneration and long-term survival of adult rat spinal motoneurons in vivo. Neuroscience 79, 765–774. 27. Peinado-Ramon P., Salvador M., Villegas-Perez M. P. and Vidal-Sanz M. (1996) Effects of axotomy and intraocular administration of NT-4, NT-3 and brain-derived neurotrophic factor on the survival of adult rat retinal ganglion cells. Invest. Ophthal. vis. Sci. 37, 489–500. 28. Prendergast J. and Stelzner D. J. (1976) Changes in the magnocellular portion of the red nucleus following thoracic hemisection in the neonatal and adult rat. J. comp. Neurol. 166, 163–172. 29. Shibiyama M., Hattori S., Himes B. T., Murray M. and Tessler A. (1998) Neurotrophin-3 prevents death of axotomized Clarke’s nucleus neurons in adult rat. J. comp. Neurol. 390, 102–111. 30. Tetzlaff W., Alexander S. W., Miller F. D. and Busby M. A. (1991) Response of facial and rubrospinal neurons to axotomy: changes in mRNA expression for cytoskeletal proteins and GAP-43. J. Neurosci. 8, 3181–3189. 31. Tetzlaff W., Kobayashi N. R., Giehl K. M. G., Tsui B. J., Cassar S. L. and Bedard A. M. (1994) Response of rubrospinal and corticospinal neurons to injury and neurotrophins. Prog. Brain Res. 103, 271–286. 32. Theriault E. and Tator C. H. (1994) Persistence of rubrospinal projections following spinal cord injury in the rat. J. comp. Neurol. 342, 249–258. 33. Tseng G.-F., Shu J., Huang S.-J. and Wang Y.-J. (1995) A time dependent loss of retrograde transport ability in distally axotomized rubrospinal neurons. Anat. Embryol. 191, 243–249. 34. Vejsada R., Sagot Y. and Kato A. C. (1995) Quantitative comparison of the transient rescue effects of neurotrophic factors on axotomized motoneurons in vivo. Eur. J. Neurosci. 7, 108–115. 35. Villegas-Perez M. P., Vidal-Sanz M., Rasminsky M., Bray G. M. and Aguayo A. J. (1993) Rapid and protracted phases of retinal ganglion cell loss follow axotomy in the optic nerve of adult rats. J. Neurobiol. 24, 23–36. 36. Ye J. H. and Houle J. D. (1997) Treatment of the chronically injured spinal cord with neurotrophic factors can promote axonal regeneration from supraspinal neurons. Expl Neurol. 143, 70–81. 37. Xu X. M. and Martin G. F. (1990) The response of rubrospinal neurons to axotomy in the adult opossum Didelphis virginiana. Expl Neurol. 108, 46–54. (Accepted 29 June 1999)