Protracted myelin clearance hinders central primary afferent regeneration following dorsal rhizotomy and delayed neurotrophin-3 treatment

Neuroscience Letters 411 (2007) 206–211

Protracted myelin clearance hinders central primary afferent regeneration following dorsal rhizotomy and delayed neurotrophin-3 treatment Lowell T. McPhail a , Jaimie F. Borisoff a,c , Bonnie Tsang a , Lucy P.-R. Hwi a , Jacek M. Kwiecien b , Matt S. Ramer a,∗ a

ICORD (International Collaboration on Repair Discoveries), The University of British Columbia, 6270 University Boulevard, Vancouver, BC, Canada V6T 1Z4 b Department of Pathology and Molecular Medicine, Central Animal Facility, RM-1U22D, Faculty of Health Sciences, McMaster University, 1200 Main Street West, Hamilton, Ont., Canada L8N 3Z5 c Neil Squire Foundation, Burnaby, 2250 Boundary Road, BC, Canada Received 3 June 2006; received in revised form 8 September 2006; accepted 8 September 2006

Abstract Regeneration within or into the CNS is thwarted by glial inhibition at the site of a spinal cord injury and at the dorsal root entry zone (DREZ), respectively. At the DREZ, injured axons and their distal targets are separated by degenerating myelin and an astrocytic glia limitans. The different glial barriers to regeneration following dorsal rhizotomy are temporally and spatially distinct. The more peripheral astrocytic barrier develops first, and is surmountable by neurotrophin-3 (NT-3) treatment; the more central myelin-derived barrier, which prevents dorsal horn re-innervation by NT-3-treated axons, becomes significant only after the onset of myelin degeneration. Here we test the hypothesis that in the presence of NT-3, axonal regeneration is hindered by myelin degeneration products. To do so, we used the Long Evans Shaker (LES) rat, in which oligodendrocytes do not make CNS myelin, but do produce myelin-derived inhibitory proteins. We show that delaying NT-3 treatment for 1 week in normal (LE) rats, while allowing axonal penetration of the glia limitans and growth within degenerating myelin, results in misdirected regeneration with axons curling around presumptive degenerating myelin ovoids within the CNS compartment of the dorsal root. In contrast, delaying NT-3 treatment in LES rats resulted in straighter, centrally-directed regenerating axons. These results indicate that regeneration may be best optimized through a combination of neurotrophin treatment plus complete clearance of myelin debris. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Astrocytes; Rhizotomy; Glial scar

Injured centrally-projecting axons must overcome multiple glial barriers in order for functional regeneration to occur. In the injured spinal cord, the lesion penumbra consists of an inner astrocytic scar and a surrounding region of degenerating myelin. Direct contusive or penetrating injuries of the CNS are problematic for evaluating the relative contributions that astrocytes and myelin make to regeneration failure since the initial trauma is highly variable and heavily influences the extent and time course of the glial reaction. The retraction of injured axons from the site of injury also confounds these studies [12,22]. Dorsal

∗

Corresponding author. Tel.: +1 604 822 5273/4956; fax: +1 604 827 5169. E-mail address: [email protected] (M.S. Ramer).

0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.09.076

rhizotomy, which does not involve direct CNS damage, is more informative given the spatially and temporally distinct evolution of astrocytic and myelin degeneration-associated barriers to regeneration [9,18] and the absence of axonal retraction. DREZ astrocytes (that separate PNS from CNS) are normally permissive to axonal regeneration, but become regenerationprohibitive soon after dorsal rhizotomy [11]. Using NT-3 expressing adenoviral vectors [28], or if treated directly with neurotrophin-3 (NT-3), primary afferent axons overcome the astrocytic barrier, traverse CNS white matter and functionally reinnervate target neurons [17,20]. However, the ability of white matter to permit regeneration in NT-3-treated animals is transient: the protracted degeneration of CNS myelin [5] renders it prohibitive even to NT-3-treated axons if treatment is delayed

L.T. McPhail et al. / Neuroscience Letters 411 (2007) 206–211

for 1 week [18]. While axons remain able to cross the PNSCNS interface following the onset of CNS myelin degeneration, they fail to elongate unidirectionally. Instead, growing axons remain within the CNS compartment of the dorsal root. These findings led us to hypothesize that degenerating myelin represents a second, more substantial barrier to regeneration. Here we use the myelin mutant Long Evans Shaker (LES) rat to test the hypothesis that products of myelin degeneration hinder axonal regeneration following delayed NT-3 treatment. The LES rat is a dysmyelinated mutant in which none of the splice variants of the myelin basic protein (MBP) gene are functional [14]. LES oligodendrocytes are found in greater number than in their normal counterparts, do not make myelin, and yet produce myelin-derived inhibitory molecules, such as MAG and NogoA [16]. Despite the presence of myelin inhibitory proteins, there is evidence for abnormal ongoing axonal plasticity in the adult CNS of LES rats [16]. We have also recently shown that sensory axons injured in the myelin-deficient dorsal columns of LES rats do not retract/die back from the lesion site, nor is there a typical robust phagocytic response to injury as is found in normally-myelinated LE rats [12]. Here, we report that in the absence of CNS myelin, the morphology of regenerating primary afferent axons following delayed NT-3 treatment resembles that of immediately-treated axons in normally-myelinated rats. These results indicate that the products of myelin degeneration, and not astrocytic derived factors, inhibit the progress of regenerating axons following delayed NT-3 treatment at the dorsal root entry zone. All procedures conformed to the guidelines of the Canadian Council on Animal Care Committee, and at McMaster University and the University of British Columbia. Adult LE (n = 9) and LES (n = 9) rats of either sex were deeply-anesthetized with Isoflurane (McMaster) or Ketamine (70 mg/kg) and Xylazine (10 mg/kg) (UBC). The spinal cords were exposed by laminectomy at the cervical level, and the dorsal roots (C3-T1) were crushed with fine forceps. Animals recovered under observation. Three animals of each genotype received injury only and the NT-3-treated animals were divided into two groups of six (n = 3 animals of each genotype). One group received NT-3 at the time of injury, the other received NT-3 1 week following rhizotomy. Briefly, for NT-3 delivery, osmotic minipumps were prepared as described previously [19]. NT-3 was delivered into the cerebrospinal fluid (6 ␮g/day for 1 week) via a cannula inserted through the atlanto-occipital membrane into the intrathecal space. The tip of the catheter rested between the C5–C7 spinal segments. All groups were killed after 1 week of intrathecal drug treatment. Deeply-anesthetized rats were perfused trans-cardially with phosphate-buffered saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB). Cervical spinal cords were removed, post-fixed for 2–4 h, and cryoprotected overnight in 20% sucrose in 0.1 M PB. The spinal cords were then frozen, cut at 16 ␮m, and processed immunohistochemically to examine large diameter NT-3-sensitive dorsal root axons using neurofilament 200 (clone N52, 1:500, Sigma Oakville ON), PNS basal laminae (rabbit anti-laminin-1, Sigma), CNS astrocytes (rabbit anti-glial fibrillary acidic protein, GFAP,

207

Dako, Denmark), oligodendrocytes (mouse anti-RIP, 1:4000, mouse anti-MBP (myelin basic protein), Chemicon, Temecula, CA), macrophages/microglia (mouse anti-Ox-42 and ED-1, 1:500, Serotec, Hornby, ON) and the chondrotin sulfate proteoglycan, neurocan (mouse anti-IF-6, 1:2000, Developmental Studies Hybridoma Bank, IA). The sections were then incubated with secondary antibodies (1:200, all raised in donkey from Jackson Immunoresearch laboratories Inc., West Grove PA), aminomethylcoumarin (AMCA, anti-rabbit) or fluorescein isothiocyanate (FITC, anti-mouse). Image analysis of axon morphology was carried out on individual rhizotomized segments from each animal. Five images from each root were used. Each image was rotated such that the DREZ was horizontal. To determine the degree of axon curl, the PNS-CNS interface was traced (from the laminin channel), and the resulting trace was overlaid onto the corresponding NF200 channel. NF200-positive axons were selected via image thresholding, and the axon overlay was subjected to two edge-detection filters (Fig. 1). Axon curliness was defined as the ratio of the area occupied by horizontal edges to the area occupied by vertical edges within the CNS compartment of the DREZ. All image analysis was performed using SigmaScan Pro 4 software (SPSS, Chicago, IL) by an individual blind to treatment groups and animal strain. NF200 immunoreactive axons were oriented unidirectionally from the PNS-CNS interface toward the cuneate fasciculus in both LE and LES un-injured rats (Fig. 2a and b). One week post-rhizotomy, NF200 immunoreactive axons did not cross the DREZ in either LE or LES rats (Fig. 2c and d). Injured axons were present on the peripheral side of the PNS-CNS interface, but only granular axonal debris was present within the CNS compartment of the root (through which axons must grow in order to re-innervate spinal grey matter). These results support the notion that astrocytes form the first barrier to regeneration at the DREZ, since axon ingrowth did not occur in the absence of CNS myelin. Following immediate NT-3 treatment, axons penetrated CNS white matter in both LE and LES rats (Fig. 2e and f). Axons elongated relatively unidirectionally from the PNS-CNS interface toward the cuneate fasciculus (Fig. 2e and f). Delayed NT-3 treatment in LE rats resulted in regeneration through the initial astrocytic barrier at the DREZ, however, growing axons were misdirected and curled around ovoids of degenerating CNS myelin (Fig. 2g) rather than growing linearly from PNS to cuneate fasiculus, as was seen in LES rats (Fig. 2h). The regenerating LE axons formed coils within the CNS compartment of the dorsal root, as we have described previously [18]. Delayed NT-3 treatment in LES rats resulted in axon regeneration through the DREZ. However, axonal profiles within the CNS compartment of the root were not curled, but were relatively straight. Axon curliness did not differ between uninjured LE and LES animals (data not shown). Where they crossed the PNS-CNS interface, regenerating axons were significantly curlier in LE rats with delayed NT-3 treatment than in other groups (Fig. 2i). We then sought to determine whether differences in axonal morphology following delayed NT-3 treatment was associated

208

L.T. McPhail et al. / Neuroscience Letters 411 (2007) 206–211

Fig. 1. Quantification of axon curliness central to the PNS:CNS interface. Sample images from LE (a, c, e, g) and LES (b, d, f, h) rats which received intrathecal NT-3 1 week following rhizotomy. Sections were processed to visualize laminin (a, b) and NF200 (c, d). The absence of laminin immunoreactivity defined the CNS portion of the dorsal root (outlines in e–h), in which measurements were made. NF200-stained sections were passed through a horizontal (e, f) and a vertical (g, h) Laplacian edge-detection filter (SigmaScan Pro). Following thresholding, the area occupied by horizonal and vertical edges was measured, and the ratio of horizonal edges (AH ) to vertical edges (AV ) was calculated. Scale bar = 50 ␮m.

with differences in glial reactivity between LE and LES rats (Fig. 3). As we have previously shown [7,12], astrocytes were much more numerous and hypertrophic in LES rats than in their normally-myelinated counterparts (Fig. 3a and b). Similarly, RIP immunoreactivity (specific for cells of the oligodendrocyte lineage) was more abundant in LES rats than LE rats, despite the lack of myelin in LES rats (Fig. 3c and d). Microglia (defined by Ox-42 immunoreactivity on the CNS side of the PNS-CNS interface) was more abundant in LES than LE rats (Fig. 3g and h) and was absent from uninjured roots (not shown). In both LE and LES injured rats, activated macrophages Ox-42 and ED-1 were present in the peripheral compartment of the root (to the left of the arrow in Fig. 3g and h). These had

the typical appearance of thin rims of macrophage cytoplasm surrounding degenerating myelin ovoids. Similar rounded activated ED-1 positive phagocyte processes were present in the CNS compartment of the root in LE rats, but not in LES rats (Fig. 3i and j), in which microglia retained stellate or elongated morphologies (Fig. 3g and h). The appearance of Ox-42 and MBP-positive profiles (Fig. 3e–h) in the CNS compartment of LE rats was identical to the morphology of curly LE axons in the CNS compartment following rhizotomy plus delayed NT-3 treatment. Likewise, the lack of Ox-42 and MBP-rimmed ovoids in LES rats correlated with the relatively unidirectional pattern of axon growth in the LES CNS root compartment. Neurocan (an inhibitory chondroitin sulfate proteoglycan) immunoreac-

L.T. McPhail et al. / Neuroscience Letters 411 (2007) 206–211

Fig. 2. Regenerating axon morphology in LE and LES rats. Regenerating axons (NF200, red) at the interface between PNS (laminin, green) and CNS in LE rats (a, c, e, g) and LES rats (b, d, f, h). In untreated rhizotomized rats, axons failed to penetrate the CNS (c, d). Immediate NT-3 treatment resulted in regeneration into the CNS in both genotypes (e, f). Regenerating axons were apparent following delayed NT-3 treatment in both genotypes (g, h), but their morphologies differed substantially: in LE rats, regenerating axons were curly, whereas in LES rats, they were comparatively straight and oriented latero-medially. Quantification of axon morphology revealed that axons were curliest in rhizotomized LE rats with delayed NT-3 treatment (i). Mann-Whitney U-test p = < 0.05. Scale bar 200 ␮m.

209

Fig. 3. Glial responses to rhizotomy at the DREZ in LE and LES rats. The distribution of astrocyte (a, b), oligodendrocyte (c–f), macrophage/microglial (g–j) markers and neurocan (k, l) following dorsal rhizotomy plus delayed NT-3 treatment. All micrographs are from the ipsilateral DREZ of LE and LES rats. The density of the glial markers GFAP, RIP and Ox-42 were all higher in LES than LE rats. MBP ovids were present the CNS portion of the root in LE rats but not LES rats (e, f). Activated macrophages (evident as round or oval Ox-42positive cytoplasmic rims, presumably surrounding degenerating myelin) and ED-1 positive cells were present in the PNS (to the left of the arrows g, h) of both LE and LES rats, but only in CNS compartment of LE rats. In LES rats, Ox-42-positive cells retained a stellate or elongated morphology. The expression of the CSPG neurocan was more concentrated at the entry zone in the LES rat (k, l). Scale bar 200 ␮m.

210

L.T. McPhail et al. / Neuroscience Letters 411 (2007) 206–211

tivity in the CNS portion of the root was greater in LES rats than LE rats (Fig. 3k and l). Again, the pattern of neurocan expression was reminiscent of the morphology of regenerating axons. Delaying NT-3 treatment for 1 week following dorsal rhizotomy, while still enabling axons to cross the astrocytic glia limitans, results in abortive regeneration within the CNS compartment of the root [18]. Regeneration failure on the CNS side of the PNS-CNS interface is evident in the curling morphology of NT-3-stimulated axons, and in their failure to reach the dorsal horn. We hypothesized that abortive regeneration was due to the onset of myelin degeneration and the exposure of primary afferent axons to myelin-derived inhibitors. We tested this hypothesis by delaying NT-3 treatment in rhizotomized LES rats, which do not make CNS myelin [14]. We found that LES sensory axons treated in this manner do not generate the curly axons typical of abortive regeneration in myelinated LE rats. We found that reduced abortive regeneration was associated with reduced microglial activation in LES rats. However, since erratically-regenerating axons do not necessarily associate with activated phagocytes following delayed NT-3 treatment [18], we reasoned that microglial activation per se does not prevent regeneration. Rather, both the conversion of resting microglia to mature phagocytes and the haphazard regeneration within the degenerating CNS compartment following delayed NT-3 treatment are mediated by degenerating myelin. The relationship between injured axons, degenerating myelin and microglial activation also differs between LE and LES rats following direct trauma to the spinal cord. In this case, microglial activation is interposed between myelin degeneration and axonal retraction and/or dieback. Ascending dorsal column axons in LES rats do not retract from the site of injury, unlike those in LE rats [12]. In addition, injured LE dorsal column axons develop complex dystrophic and ramified endings, whereas LES axons each develop a single, small and simple end-bulb. The retraction/dieback of sensory axons in LE rats was attributable to macrophage-mediated degradation of myelin sheaths caudal to the injury site, since it was absent in LES rats (in which there is little macrophage activation following dorsal column injury), and because it was mitigated by the macrophage/microglial inhibitor minocycline in LE rats [12,22]. What is abundantly clear from this and previous studies is that reactive astrocytes, whether at the DREZ or at the site of a spinal cord injury, form a barrier to regeneration that axons in neither LE nor LES rats can cross unless treated with NT-3. The assertion that reactive (as opposed to quiescent) astrocytes are fundamental to regeneration failure stems from work in the spinal cord, [2] and from work showing that the DREZ of an uninjured dorsal root is capable of supporting elongation of sensory axons from PNS to CNS compartments [11]. The identities of the molecules responsible for hindering axonal regeneration at the DREZ remain poorly defined. Glia limitans astrocytes, like those of the CNS glial scar, generate an extracellular matrix containing a large number of molecules that impede the regeneration of axons [10,20]. In addition, tenascin-R expression has been correlated with the inhibitory environment of the DREZ following injury [29]. Other candidates include families of sulfated proteoglycans (SPGs) including NG2 which is upregulated

within the DREZ as a result of injury [29]. Recent in vitro work has examined the role of chondroitin (C) SPGs in the maintenance of the DREZ [6]. By preventing synthesis of, or causing degradation of CSPGs, these studies indicate that CSPGs may act to maintain the respective peripheral and central distributions of Schwann cells and astrocytes, and hence underlie the inability of injured dorsal root axons to penetrate the glia limitans without NT-3. Following dorsal root injury, astrocytes are the major producers of the CSPG neurocan [1]. Indeed, in the present study, the expression of Neurocan appeared higher in the LES than the LE rats, likely due to the increased numbers of astrocytes in the LES rats. Interestingly, as is the case in the mouse [1], neurocan is expressed in CNS astrocytes but not in the astrocytic processes that extend into the PNS as a result of injury. While NT-3 treatment allows axons to overcome the first (astrocytic) barrier, it falls short of promoting successful regeneration through an environment in which elongating axons are exposed to the second myelin-derived barrier. How does degenerating myelin perturb regeneration when NT-3 treatment is delayed? One hypothesis invokes receptor-mediated interactions involving myelin-derived inhibitory molecules, such as myelinassociated glycoprotein (MAG), NogoA, and oligodendrocytemyelin glycoprotein (OMgp) [4,8,26,27]. These hinder regeneration by binding to an ever-expanding complex of receptors, so far involving the Nogo-66 receptor homologues 1 and 2 (the latter of which binds MAG but not OMgp or Nogo) [3,4,8,23,26], Lingo-1 [13], p75 [25,27] and/or TROY [15,21]. The morphological similarity between curly regenerating axons and Ox-42/MBP-ringed ovoids central to the PNSCNS interface evokes a second hypothesis: that regeneration is thwarted by simple steric hindrance conferred by bulky products of incomplete myelin degeneration. Myelin clearance is a protracted process in the CNS. For example, in the rhizotomized rat, myelin debris is still present between 75 and 90 days post-injury [5,24]. Such inefficient myelin clearance in the CNS is likely to be due to ineffective activation of the complement system [9]. The present study provides new evidence for the role of degenerating myelin in the two tiered inhibition at the dorsal root entry zone following injury. These results emphasize the requirement for a combination of therapies for regeneration within the CNS to become feasible [20]. The present data suggest that the most favorable environment for regeneration is one in which CNS myelin has been completely cleared. In combination with procedures that enhance the cell body response to injury (such as NT-3 administration), treatments that facilitate the elimination of CNS myelin debris are likely to result in the most extensive regeneration, and thus may hold the greatest promise for functional recovery following injury. Acknowledgements The authors wish to thank L.J.J. Soril for technical assistance and W.T. Plunet for critical reading of this manuscript. MSR is a Michael Smith Foundation for Health Research (MSFHR) Scholar. LTM and JFB are supported by Fellowships from MSFHR and the Natural Science and Engineering Research Council of Canada (NSERC).

L.T. McPhail et al. / Neuroscience Letters 411 (2007) 206–211

References [1] A.T. Beggah, M.T. Dours-Zimmermann, F.M. Barras, A. Brosius, D.R. Zimmermann, A.D. Zurn, Lesion-induced differential expression and cell association of Neurocan, Brevican, Versican V1 and V2 in the mouse dorsal root entry zone, Neuroscience 133 (2005) 749–762. [2] S.J. Davies, M.T. Fitch, S.P. Memberg, A.K. Hall, G. Raisman, J. Silver, Regeneration of adult axons in white matter tracts of the central nervous system, Nature 390 (1997) 680–683. [3] M. Domeniconi, Z. Cao, T. Spencer, R. Sivasankaran, K. Wang, E. Nikulina, N. Kimura, H. Cai, K. Deng, Y. Gao, Z. He, M. Filbin, Myelin-associated glycoprotein interacts with the Nogo66 receptor to inhibit neurite outgrowth, Neuron 35 (2002) 283–290. [4] A.E. Fournier, T. GrandPre, S.M. Strittmatter, Identification of a receptor mediating Nogo-66 inhibition of axonal regeneration, Nature 409 (2001) 341–346. [5] R. George, J.W. Griffin, The proximo-distal spread of axonal degeneration in the dorsal columns of the rat, J. Neurocytol. 23 (1994) 657–667. [6] B. Grimpe, Y. Pressman, M.D. Lupa, K.P. Horn, M.B. Bunge, J. Silver, The role of proteoglycans in Schwann cell/astrocyte interactions and in regeneration failure at PNS/CNS interfaces, Mol. Cell Neurosci. 28 (2005) 18–29. [7] J.M. Kwiecien, L.T. O’Connor, B.D. Goetz, K.H. Delaney, A.L. Fletch, I.D. Duncan, Morphological and morphometric studies of the dysmyelinating mutant, the Long Evans shaker rat, J. Neurocytol. 27 (1998) 581–591. [8] B.P. Liu, A. Fournier, T. GrandPre, S.M. Strittmatter, Myelin-associated glycoprotein as a functional ligand for the Nogo-66 receptor, Science 297 (2002) 1190–1193. [9] L. Liu, J.K. Persson, M. Svensson, H. Aldskogius, Glial cell responses, complement, and clusterin in the central nervous system following dorsal root transection, Glia 23 (1998) 221–238. [10] J. McGraw, G.W. Hiebert, J.D. Steeves, Modulating astrogliosis after neurotrauma, J. Neurosci. Res. 63 (2001) 109–115. [11] L.T. McPhail, W.T. Plunet, P. Das, M.S. Ramer, The astrocytic barrier to axonal regeneration at the dorsal root entry zone is induced by rhizotomy, Eur. J. Neurosci. 21 (2005) 267–270. [12] L.T. McPhail, D.P. Stirling, W. Tetzlaff, J.M. Kwiecien, M.S. Ramer, The contribution of activated phagocytes and myelin degeneration to axonal retraction/dieback following spinal cord injury, Eur. J. Neurosci. 20 (2004) 1984–1994. [13] S. Mi, X. Lee, Z. Shao, G. Thill, B. Ji, J. Relton, M. Levesque, N. Allaire, S. Perrin, B. Sands, T. Crowell, R.L. Cate, J.M. McCoy, R.B. Pepinsky, LINGO-1 is a component of the Nogo-66 receptor/p75 signaling complex, Nat. Neurosci. 7 (2004) 221–228, Epub 2004 Feb 15. [14] L.T. O’Connor, B.D. Goetz, J.M. Kwiecien, K.H. Delaney, A.L. Fletch, I.D. Duncan, Insertion of a retrotransposon in Mbp disrupts mRNA splicing and myelination in a new mutant rat, J. Neurosci. 19 (1999) 3404–3413. [15] J.B. Park, G. Yiu, S. Kaneko, J. Wang, J. Chang, Z. He, A TNF receptor family member, TROY, is a coreceptor with Nogo receptor in mediating the inhibitory activity of myelin inhibitors, Neuron 45 (2005) 345–351.

211

[16] V. Phokeo, J.M. Kwiecien, A.K. Ball, Characterization of the optic nerve and retinal ganglion cell layer in the dysmyelinated adult Long Evans Shaker rat: evidence for axonal sprouting, J. Comp. Neurol. 451 (2002) 213–224. [17] M.S. Ramer, T. Bishop, P. Dockery, M.S. Mobarak, D. O’Leary, J.P. Fraher, J.V. Priestley, S.B. McMahon, Neurotrophin-3-mediated regeneration and recovery of proprioception following dorsal rhizotomy, Mol. Cell Neurosci. 19 (2002) 239–249. [18] M.S. Ramer, I. Duraisingam, J.V. Priestley, S.B. McMahon, Two-tiered inhibition of axon regeneration at the dorsal root entry zone, J. Neurosci. 21 (2001) 2651–2660. [19] M.S. Ramer, J.V. Priestley, S.B. McMahon, Functional regeneration of sensory axons into the adult spinal cord, Nature 403 (2000) 312–316. [20] L.M. Ramer, M.S. Ramer, J.D. Steeves, Setting the stage for functional repair of spinal cord injuries: a cast of thousands, Spinal Cord 43 (2005) 134–161. [21] Z. Shao, J.L. Browning, X. Lee, M.L. Scott, S. Shulga-Morskaya, N. Allaire, G. Thill, M. Levesque, D. Sah, J.M. McCoy, B. Murray, V. Jung, R.B. Pepinsky, S. Mi, TAJ/TROY, an orphan TNF receptor family member, binds Nogo-66 receptor 1 and regulates axonal regeneration, Neuron 45 (2005) 353–359. [22] D.P. Stirling, K. Khodarahmi, J. Liu, L.T. McPhail, C.B. McBride, J.D. Steeves, M.S. Ramer, W. Tetzlaff, Minocycline treatment reduces delayed oligodendrocyte death, attenuates axonal dieback and improves behavioral outcome after spinal cord injury, J. Neurosci. 24 (2004) 2182–2190. [23] K. Venkatesh, O. Chivatakarn, H. Lee, P.S. Joshi, D.B. Kantor, B.A. Newman, R. Mage, C. Rader, R.J. Giger, The Nogo-66 receptor homolog NgR2 is a sialic acid-dependent receptor selective for myelin-associated glycoprotein, J. Neurosci. 25 (2005) 808–822. [24] P.D. Wall, S.B. McMahon, Long range afferents in rat spinal cord. III. Failure of impulse transmission in axons and relief of the failure after rhizotomy of dorsal roots, Philos. Trans. R. Soc. Lond. B Biol. Sci. 343 (1994) 211–223. [25] K.C. Wang, J.A. Kim, R. Sivasankaran, R. Segal, Z. He, P75 interacts with the Nogo receptor as a co-receptor for Nogo, MAG and OMgp, Nature 420 (2002) 74–78. [26] K.C. Wang, V. Koprivica, J.A. Kim, R. Sivasankaran, Y. Guo, R.L. Neve, Z. He, Oligodendrocyte-myelin glycoprotein is a Nogo receptor ligand that inhibits neurite outgrowth, Nature 417 (2002) 941–944. [27] S.T. Wong, J.R. Henley, K.C. Kanning, K.H. Huang, M. Bothwell, M.M. Poo, A p75(NTR) and Nogo receptor complex mediates repulsive signaling by myelin-associated glycoprotein, Nat. Neurosci. 5 (2002) 1302–1308. [28] Y. Zhang, P.A. Dijkhuizen, P.N. Anderson, A.R. Lieberman, J. Verhaagen, NT-3 delivered by an adenoviral vector induces injured dorsal root axons to regenerate into the spinal cord of adult rats, J. Neurosci. Res. 54 (1998) 554–562. [29] Y. Zhang, K. Tohyama, J.K. Winterbottom, N.S. Haque, M. Schachner, A.R. Lieberman, P.N. Anderson, Correlation between putative inhibitory molecules at the dorsal root entry zone and failure of dorsal root axonal regeneration, Mol. Cell Neurosci. 17 (2001) 444–459.