A multifunctional neurotrophin with reduced affinity to p75NTR enhances transplanted Schwann cell survival and axon growth after spinal cord injury

Experimental Neurology 248 (2013) 170–182

Contents lists available at ScienceDirect

Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr

A multifunctional neurotrophin with reduced affinity to p75NTR enhances transplanted Schwann cell survival and axon growth after spinal cord injury Mitsuhiro Enomoto a,1, Mary Bartlett Bunge a,b,c, Pantelis Tsoulfas a,b,c,⁎ a b c

The Miami Project to Cure Paralysis, University of Miami School of Medicine, Miami, FL 33136, USA Department of Neurological Surgery, University of Miami School of Medicine, Miami, FL 33136, USA Department of Cell Biology, University of Miami School of Medicine, Miami, FL 33136, USA

a r t i c l e

i n f o

Article history: Received 19 March 2013 Revised 24 May 2013 Accepted 13 June 2013 Available online 20 June 2013 Keywords: Neurotrophin Multineurotrophin (MNT) p75NTR Schwann cells (SCs) Supraspinal fibers Sensory axons 5-HT

a b s t r a c t The lack of regeneration of axonal pathways after SCI is associated with the presence of inhibitory molecules within the glial scar, the loss of the neuron's intrinsic capacity to grow and the absence of growth factors. The NGF family of neurotrophins is a potent growth factor for several types of supraspinal and sensory axons. It is unclear, however, whether the neurotrophin's axon growth-promoting activities after central nervous system (CNS) injuries are mediated through the Trk receptors or p75 neurotrophin receptor (p75NTR) or both. To investigate the role of these receptors in the re-growth of specific fiber tracts after SCI, we created a series of neurotrophins that preferentially bind to either TrkB/C or p75NTR receptors. All the mutations were made on the NT-3/D15A backbone, a multifunctional neurotrophin that can bind TrkB, TrkC and p75NTR. To test the mutants' axon growth-promoting activity after rat contusion SCI, we examined several spinal cord fiber projections after transplanting Schwann cells (SCs) expressing the different multi-neurotrophins. Grafts expressing the NT-3/D15A with reduced binding affinity to p75NTR contained more surviving SCs, and sensory as well as supra-spinal fibers, within the transplant than the NT-3/D15A neurotrophin-SC grafts. These data support the idea that neurotrophins lacking p75 activity can be more effective in promoting axon growth after CNS injury. © 2013 Elsevier Inc. All rights reserved.

Introduction Grafting rat Schwann cells (SCs) into injured adult rat spinal cord promotes the regeneration of spinal and sensory axons (Xu et al., 1995a, 1997). Even though SC transplantation after SCI provides a favorable environment for axonal regeneration and myelination, it is not enough to facilitate axonal growth beyond the graft and promote robust functional recovery (Fortun et al., 2009; Oudega and Xu, 2006). To overcome this limitation, several groups have devised combination strategies that combine SC transplantation and delivery of neurotrophins (NTs) by external infusion or transplantation of NT-producing SCs and other cell types (Fortun et al., 2009; Lu and Tuszynski, 2008; Oudega and Xu, 2006; Tetzlaff et al., 2012). This approach has been shown to enhance axonal regeneration of supraspinal as well as spinal and sensory fibers (Bregman et al., 1997; Tuszynski et al., 1994; Xu et al., 1995b). Specific NTs stimulate axonal growth after injury through a variety of cellular

⁎ Corresponding author at: The Miami Project to Cure Paralysis, Department of Neurological Surgery, and Department of Cell Biology, University of Miami School of Medicine, Miami, FL 33136, USA. Fax: + 1 305 243 3923. E-mail address: [email protected] (P. Tsoulfas). 1 Present address: Department of Orthopedic Surgery, Graduate School, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, 113-8519, Japan. 0014-4886/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.expneurol.2013.06.013

mechanisms. Brain derived neurotrophic factor (BDNF) can reverse the atrophy of rubrospinal neurons after axotomy (Kobayashi et al., 1997). NT-3 can prevent the cell death and reduce atrophy of spinal cord projection neurons (Bradbury et al., 1999). BDNF and NT-3 can reduce axonal degeneration and induce sprouting of cortico-spinal axons (Grill et al., 1997; Hiebert et al., 2002; Sayer et al., 2002). These observations suggest that the use of multiple NTs might be necessary for restoring coordinated aspects of locomotion. Structure-function studies investigating the interaction between the NGF family of NTs and their receptors showed that new NTs could be developed to bind more than one Trk receptor (Ibáñez et al., 1993; Urfer et al., 1994, 1997). One of these multineurotrophins (MNTs), NT-3/D15A, has been found to bind and activate TrkB, TrkC and p75NTR (Urfer et al., 1994). The NT-3/D15A has been used in several models of SCI (Cao et al., 2005; Golden et al., 2007). Neurotrophins bind Trk receptors and signal through the activation of specific downstream serine threonine kinases to mediate both neurite/axon growth and survival of neurons (Huang and Reichardt, 2003). But the NTs can also bind to p75NTR and thereby trigger apoptosis during development and following injury (Dechant and Barde, 2002; Teng et al., 2010). The proapoptotic effect is thought to be mediated by the proneurotrophin (proNT) form of these molecules (Teng et al., 2010). Since p75NTR and proNTs have been implicated in apoptosis,

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

we set out to design MNTs that have reduced affinity for this receptor and therefore transduce mainly growth and survival signals. Here we report the generation of an MNT that does not bind to the p75NTR and testing its effects on the survival of SCs and growth of axons in a SCI model. Where the biological properties of NTs have adapted to specific physiological roles in various animal cells and organs, such attributes often are incompatible with their use as theurapeutic drugs. Therefore, the generation of new NT mutants with modified receptor specificity may be more successful than the naturally occurring proteins for salutary clinical outcomes. Material and methods

171

and stained for neuronal tubulin (TuJ1, Covance, Emeryville, CA). We quantified Tuj1 positive areas by capturing pictures using a 4× objective. The intensity was calculated in one-fourth of the area of the explants as in Fig. 2A with image analysis software 3i (Intelligent Imaging, Denver, CO). DRG dissociated cultures were prepared as described previously (Kleitman et al., 1998). The dissociated cells were plated in neurobasal medium with B-27 supplement and L-Glutamine (Invitrogen). Purified human NT-3, human BDNF, conditioned medium (50 ng/ml), or control medium without transfection was added at days 0 and 3. Cells were maintained for 5 days and then fixed and stained for TuJ1 and DAPI. The quantification was done by capturing pictures of five random areas and counting Tuj1 positive cells using a 4x objective.

Recombinant DNA techniques and transfections The mutations were performed on, the human bi-functional neurotrophin, NT-3/D15A, a cDNA previously described (Urfer et al., 1994). For purposes of clarity and to avoid long naming schemes of the various mutants, we will refer to this bi-functional neurotrophin with the general term of multineurotrophin (MNT). All the residues were mutated to amino acid alanine using standard procedures. The cDNAs were cloned into the lentiviral vector pRRLsin.PPT.Th.CMV.MCS.Wpre that carries the CMV promoter (Dull et al., 1998; Follenzi and Naldini, 2002). We constructed the following mutant neurotrophins: 1) NT-3/ D15A/R103A, 2) p75-2 (NT-3/D15A/p75-2 denoting the mutations R114A/K115A), 3) p75-22 (NT-3/D15A/p75-22 denoting mutations on Y11A/R68A/R87A/R114A/K115A), 4) p75-23 (NT-3/D15A/p75-23 denoting mutations on R68A/R87A/R114A/K115A), 5) p75-24 (NT-3/ D15A/p75-24 denoting mutations on Y11/R87A/R114A/K115A), 6) p7525 (NT-3/D15A/p75-25 denoting mutations on R87A/R114A/K115A), 7) p75-26 (NT-3/D15A/p75-26 denoting mutations on Y11A/R114A/ K115A), 8) NT-3/D15A/R103A,Y51A and 9) NT-3/D15A/R103A/p75-2.

Culture of Schwann cells and infection with lentiviruses The generation and characterization of SC cultures from adult Fischer rat sciatic nerves were described previously (Morrissey et al., 1991). These cells were cultured in the presence of mitogenic factors (20 mg/ml bovine pituitary extract, 2 mM forskolin, and 2.5nM heregulin). The purity of SCs by this method is between 95 and 98%. At passage 1 (P1), cells were infected with the lentiviral particles carrying the different mutants and GFP, grown for 5 days and frozen. For the grafting experiments, cells were thawed (P2) and grown for 4–5 days. The cells were trypsinized to dislodge them from the dishes and collected at the concentration of 2 × 106 cells in 6 μl of DMEM/ F12 medium for the injection into the contused spinal cord. Supernatants were collected at P2 and an ELISA assay determined the concentration of secreted neurotrophins. To determine the percentage of infected cells, SCs were fixed and immunostained with an anti-NT-3 antibody (1:200, Peprotech, Oak Park, CA).

Generation of lentiviral particles

Animals and surgical procedures

For lentiviral production we used the four plasmid method as described previously (Dull et al., 1998; Follenzi and Naldini, 2002). The virus was concentrated by ultracentrifugation and stored at − 80 °C in the presence of PBS and 0.5% bovine albumin. The titers, shown as transducing units (TU) from 108 to 109 TU/ml, were derived using ELISA for the HIV Gag capsid protein p24.

Adult female Fischer rats (n = total 56, 180–200 g, Harlan, Frederick, MD) were housed according to National Institutes of Health and United States Department of Agriculture guidelines. The Institutional Animal Care and Use Committee of the University of Miami approved all animal experiments. The rats were anesthetized with a mixture of 1.5% isoflurane, 70% nitrous oxide, and 30% oxygen, and then placed on a surgical table on a heating pad (37 ± 0.5 °C). Lacrilube ophthalmic ointment (Allergan Pharmaceuticals, Irvine, CA) was applied to the eyes to prevent drying. All surgeries were performed using aseptic techniques. The spinal column was exposed and a laminectomy performed at T9 and the cord moderately contused using the NYU impactor (10 g, 12.5 mm). Gentamicin (5 mg/kg, intramuscular; Abbott Laboratories, North Chicago, IL) was administered immediately postsurgery once a day for 7 days. The analgesic, Buprenex (0.01 mg/kg of 0.3 mg/ml, subcutaneous; Reckitt Benckiser, Richmond, VA) was delivered postsurgery twice a day for 2 days. The rats were maintained for 7 weeks after injury. One week after injury, the injury site was exposed. A total of 2 × 106 cells in 6 μl of DMEM/F12 medium was injected into the contused area using a 10 ml Hamilton syringe with a pulled glass micropipette (tip diameter 100 μm) held in a micromanipulator at the depth of 1 mm. The glass micropipette was held in place for 3 min and withdrawn slowly. After the injection of SCs, the muscle layers and skin were closed separately. Rats were transplanted with SCs, infected with LV-eGFP (n = 10), LV-NT-3/D15A (n = 12), LV-NT-3/D15A/R103A (n = 11), LV-NT-3/ D15A/p75-2 (n = 13) and NT-3/D15A/R103A/p75-2 (n = 10). The rats in each experimental group were randomly divided into subgroups for 1 μm-plastic sections and thin sections for EM (LV-GFP, n = 3; LV-NT-3/D15A, n = 4; LV-NT-3/D15A/p75-2, n = 4). We did not include a group without transplanting SC because previous results have shown multiple times that SC cells alone improve the locomotor behavior (Takami et al., 2002).

Western blot and ELISA To generate neurotrophin supernatants, 293-HEK cells were transfected using Lipofectamine 2000 (Invitrogen, Grand Island, NY). Twenty-four hours post-transfection the medium was replaced with DMEM/F12 containing 10 mg/l insulin and 1 mg/l human transferrin. The supernatants and cells were collected two days later. NT-3 protein was analyzed by the NT-3 Emax ImmunoAssay system (Promega, Madison, WI). Lysates of transfected cells were prepared as described previously (Tsoulfas et al., 1996). Western blotting was performed using a rabbit polyclonal anti-NT-3 antibody (1:500, Peprotech, Oak Park, CA). The enhanced chemiluminescence (ECL) western blotting detection system was used for the visualization of the proteins (GE Healthcare Life Sciences, Piscataway, NJ). Dorsal root ganglion (DRG) explant neurite outgrowth and survival assays Cervical and lumbar DRGs were dissected from E15 Sprague Dawley rats. For evaluation of neurite outgrowth, we plated DRG explants on dishes coated with poly L-ornithine (Sigma-Aldrich, St. Louis, MO) and 1 mg/ml fibronectin (Invitrogen) in the presence of purified human NT-3 (Peprotech, Oak Park, CA), human BDNF (Peprotech), conditioned medium with NT-3 mutants (1 ng/ml), and control medium without transfection for two days. Explants were fixed

172

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

Histological and immunohistochemical procedures Six weeks after transplantation, animals were intracardially perfused with paraformaldehyde under terminal anesthesia. The spinal cords were removed and 15 mm-long thoracic segments, including the injury site, were embedded in gelatin. For 1 μm-plastic and thin sections, we dissected a 1 mm-thick slice of the injury epicenter as previously described (Xu et al., 1995a, 1995b). Every 6th frozen section was stained with neuronal markers, monoclonal anti-p75 (1:2, 192 IgG, DHSB, Iowa City, IW), neurofilament-M (1:1000, 3H11, EnCor, Gainesville, FL), neurofilament-H (1:600, SM32 Covance, Emeryville, CA), polyclonal neurofilament-M (1:200. EMD Millipore, Billerica, MA), CGRP (1:1000, Peninsula laboratories, San Carlos, CA), 5-HT (1:1000, ImmunoStar, Hudson, WI), Parvalbumin (1:500, Swant, Switzerland), or GFAP (1:1500, Dako, Carpinteria, CA) antibodies. Fluorescent isolectin IB4 is from Invitrogen. The secondary antibodies, Alexa488, Alexa594 and Alexa647, are from Invitrogen. Quantification Estimation of cavity volume of spinal cords We estimated the volumes of cavities of the spinal cord using an unbiased method employing computer-assisted microscopy and Neurolucida software (MBF Bioscience, Williston, VT). One series of horizontal sections was mounted with Citifluor containing DAPI (Vector Laboratories, Burligame, CA). We traced the contour of the cavities in each section for 15 mm-lengths. We used serial section manager of the Neurolucida software to make 3D reconstructions. The software calculated the total estimated volume of injury. Estimation of the number of transplanted cells GFP positive cells were counted in one series of mounted sections using an unbiased method employing computer-assisted microscopy and StereoInvestigator software (MBF Bioscience, Williston, VT). Briefly, we traced the boundary of eGFP positive areas, and set the counting frame as 50 × 50 μm and used serial section manager of the StereoInvestigator software to make 3D reconstructions. We counted both GFP and DAPI positive cells using a 63 × objective with oil immersion in the transplanted area of each section. The software calculated the total estimated number of GFP positive cells in the transplant. Estimation of the number of myelinated axons in transverse sections We estimated the total number of myelinated axons in 1 μm transverse plastic sections stained with toluidine blue using the microscope and StereoInvestigator software as described (Takami et al., 2002). We analyzed the sections of LV-eGFP (n = 3), LV-D15A (n = 4), and LV-D15A/p75-2 (n = 4 in 4). We randomly took 22 pictures in each section, and counted the number of myelinated axons. Estimation of the number of parvalbumin and 5HT-positive axons 4 sections from the serial sectioning were incubated with parvalbumin or 5-HT antibody, respectively, overnight at 4 °C. Parvalbuminpositive fibers were located in dorsal spinal cord; 5HT positive fibers were located from mid-dorsal to ventral areas. After washing, biotinylated anti-rabbit IgG (Vector laboratories, Burligame, CA) as the secondary antibody was incubated for 1 h, and subsequently reacted with the avidin–biotin complex (ABC Elite; Vector laboratories, Burligame, CA). The sections were visualized by using the diaminobenzine (DAB) substrate kit (Vector laboratories). We estimated the length and density of immunoreactive axons using an unbiased method employing computer-assisted microscopy and StereoInvestigator software. For analysis of parvalbumin positive fibers, we set the contour as the GFP positive area, and counted inside the area. The areas for 5HT positive fibers were divided into four regions; GFP positive areas, 1 mm rostral, and 1 mm caudal from the edge of GFP positive areas and the area around the transplant. We set

4 contours in the section, and counted the axons in each section using Isotropic Virtual Planes of the software (Larsen et al., 1998). Behavioral test for hindlimb function Hindlimb performance was scored using the open field locomotor BBB test (Basso et al., 1995, 1996). Two observers, unaware of the treatment procedures, performed BBB scoring once a week for 7 weeks. Results Engineering of MNTs that selectively bind to TrkB/C and/or p75NTR receptors Previous work demonstrated that several supraspinal fiber groups respond to different NTs (Fortun et al., 2009; Lu and Tuszynski, 2008; Tetzlaff et al., 2012). Therefore, we wanted to use a single molecule instead of several neurotrophins to target multiple fiber tracts. To accomplish this we used the human NT-3 gene, with a single residue mutation on position 15 (D15A) of the mature form of NT-3, NT-3/ D15A. This neurotrophin can activate TrkB as well as the cognate ligands BDNF and NT4/5 (Urfer et al., 1994). Since the p75NTR receptor often antagonizes the trophic functions of NTs, we mutated specific conserved NT-3 residues that have been shown to be important binding determinants for p75NTR (Gong et al., 2008; Urfer et al., 1994). These amino acid residues are Tyr-11 (Y11), Arg-68 (R68), Arg-87 (R87), Arg-114 (R114) and Lys-115 (K115) (Fig. 1). Equivalent amino acid residues in NGF contribute to the interaction with p75NTR (He and Garcia, 2004). Previously it has been shown that for NT-3, the residues R114 and K115 are the most critical binding determinants in the interaction with p75NTR (Urfer et al., 1994). The importance of other conserved residues, K49 and Arg87, in site 1 (Gong et al., 2008) to the overall binding affinity to p75NTR has not yet been determined. Therefore, we used the MNT NT3/D15A/R114A/K115A (p75-2) backbone to mutate amino acid residues in positions Y11, Tyr-51 (Y51), R68 and R87 to alanine in various combinations. The reason for this approach was to further reduce the binding affinity of p75-2 to p75NTR. For control purposes we made two more neurotrophin mutants. The first was an MNT that binds mainly to the p75NTR receptor, and the second had reduced affinity for both Trk and p75NTR receptors. We created the first control MNT by mutating residue R103 of the human NT-3/D15A to alanine. Residue Arg-103 (R103) is highly conserved in all NTs (NGF, BDNF, NT-3 and NT-4/5) and is the most important binding determinant for the interaction between NT-3 and TrkC (Urfer et al., 1994; Wiesmann and de Vos, 2001). This MNT will allow testing whether the p75NTR activity alone has any effect on the survival of SCs and axon growth. The second control MNT is NT-3/D15A/R103A/p75-2. This MNT should have reduced affinity to both the Trk and p75NTR receptors. Requirement of MNT p75NTR binding for optimal survival and neurite growth of fetal DRG neurons in vitro To select for the MNT with reduced affinity to p75NTR we measured the secretion efficiency and bioactivity of various MNT supernatants on DRG neurons. As shown in Fig. 2A, NT-3 ELISA assays, performed on supernatants from transfected HEK293 cells containing the various mutant MNTs, did not detect any mutant neurotrophins with alanine residues at positions Y51, R68 and R87 (p75-21, p75-25 and R103/Y51A). This suggests that perhaps these residues are critical for processing and secretion. All other neurotrophins were secreted at various levels compared to the NT-3/D15A. Similar to previous data (Urfer et al., 1994), the neurotrophin with the R103A mutation was secreted more efficiently than the NT3/D15A. Western blot analysis (Fig. 2B), using antisera that recognize NT-3, revealed the presence of mature neurotrophin in supernatants from NT-3/D15A, NT-3/D15A/

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

173

Fig. 1. Contacts between NT-3 and p75NTR. (A) Ribbon view of the overall structure of rat p75NTR:human NT-3 complex. The unit comprises one homodimer of NT-3 bound to two monomers of p75NTR. Yellow and green each represent monomers of the NT-3 homodimer. The p75NTR monomers are colored in orange and light blue. Squares indicate the interface at sites 1 and 2 between NT-3 homodimers and one monomer of p75NTR. Close-up views are presented in B and C. (B) Site 1 interface shows the salt bridge (dashed lines) between residue R87 from the green NT-3 monomer and residue D41 from the p75NTR monomer. Residue Y51 from the same NT-3 monomer also is shown. (C) Site 2 interface shows the salt bridges between residue R114 from the yellow NT-3 monomer to residue E119 of p75NTR, and residue 68 of the green NT-3 monomer to residue D134 of p75NTR. Also are shown the hydrogen bonds between residue Y11 of the green NT-3 monomer to residue L106 of the P75NTR monomer and residue R114 of the yellow NT-3 monomer to residue C136 of the p75NTR. Bridges are indicated in dashed lines. (D) Sequence alignment of the mature form of human NT-3, NGF, BDNF, and NT4/5. Numbering of residues is based on the mature human NT-3. The mutated amino acids in this study are indicated in asterisks for residues that form salt bridges and hydrogen bonds and inverted triangles for residues not involved in intermolecular interactions. The renderings of the ribbon model structures were done using Molsoft Browser Pro.

R103A, p75-2, p75-26 and NT-3/D15A/R103A/p75-2 transfected cells. This is consistent with the ELISA assay results. With the exemption of p75-23, all other p75 mutant MNT supernatants contained NT-3

immunopositive signals at varying molecular mass and ratios. To control for the efficiency of transfection among all the constructs, we performed Western blot analysis on cell lysates of transfected HEK293

174

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

Fig. 2. Characterization of the different NT-3/D15A MNT variants. (A) ELISA analysis with an anti-NT-3 antibody of conditioned medium from 293 HEK cells transfected with the different NT-3/D15A mutants. Data points represent mean values ± SEM from three independent experiments. Western blot analysis of conditioned media (B) and whole cell lysates (C) of the same NT-3/D15A mutant proteins as in (A). Pu.NT-3 is purified recombinant human NT-3; Mock is HEK-293 cells transfected with an empty vector. Arrows indicate the mature and immature forms of NT-3 in B and C, respectively. (D) Immunocytochemical analysis of E15 rat DRG explants with the neuronal marker Tuj1 (green) and DAPI (blue). The DRG explants were grown for 4 days in the presence of mock transfected conditioned medium (control), recombinant human BDNF, recombinant human NT-3 and conditioned medium from NT-3/D15A, NT-3/D15A/R103A or p75-2. The concentration of neurotrophin in all conditions was 1 ng/ml. (E) Quantitative analysis of neurite Tuj1 fluorescent signal intensity with various NT-3/D15A mutant proteins. Data points represent mean values ± SEM from N9 DRG explants analyzed in three different experiments. (F) Analysis of E15 neuron survival exposed to 50 ng/ml of various neurotrophins. Data points represent mean values ± SEM from N300 neurons analyzed in three different experiments. Dunnett's Multiple comparison test was used.*** = p b 0.0001, * = p b 0.05. Scale bar, μm.

cells with antisera that recognize NT-3. All lysates except for the mocktransfected cells showed two intense immunopositive bands at higher molecular mass than the expected mature form (Fig. 2C). These two bands correspond to the proMNT forms and have a similar pattern to the proNT-3 as described earlier (Yano et al., 2009). Based on these results we continued the testing of the biological activity of MNT variants detected with the ELISA assay. We assessed the biological activity of various MNT mutants in two bioassays, embryonic DRG explants for neurite outgrowth and dissociated embryonic peripheral sensory neurons for survival. Past studies have shown that DRGs contain neurons that are dependent on

both NT-3 and BDNF and express p75NTR (Buchman and Davies, 1993; Chao, 1994). For most sensory neurons including DRGs, a protein concentration of 1 ng/ml enhances survival of half the maximum number of neurons in response to the saturating concentration. This high sensitivity of DRG neurons to neurotrophins allows us to detect small differences in NT activity. Embryonic DRG explants were incubated with 1 ng/ml of purified NT-3, BDNF, or 1 ng/ml of supernatants containing the MNT (Figs. 2D,E). The neurite outgrowth of NT-3/D15A was as extensive as the in NT-3 treated explants. As expected, BDNF showed only a small increase in neurite outgrowth compared to control because only a small percentage of DRG neurons are dependent on BDNF. The

Fig. 3. Increased of GFP-expressing SCs in implants with reduced affinity to p75NTR.Representative horizontal sections 6 weeks after transplantation of SCs infected with eGFP (A), NT-3/D15A-eGFP (B), NT-3/D15A/R103A-eGFP (C), or p75-2-eGFP (D). Transplanted cells expressing GFP were observed in all treatment groups. Asterisks indicate the presence of cavities in the spinal cord. (E) Quantification of surviving SCs expressing GFP after grafting. Data points represent mean values ± SEM, analyzed from at least 7 animals for each group. (F) Estimation of cavity volumes in the different groups of injured animals. Data points represent mean values ± SEM, analyzed from at least 7 animals for each group. Tukey's multiple comparison test was used. * = p b 0.05. Scale bar, 1 mm (A).

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

175

176

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

MNT, NT-3/D15A/R103A, showed no significant increase compared to the explant without NTs (control) (Fig. 2D). Interestingly, p75-2 and p75-6, showed a 40% decrease in the signal intensity, suggesting that in embryonic explants the p75NTR activity is required for optimal neurite outgrowth. As shown in Fig. 2F, in dissociated DRG cultures, neuron survival with NT3/D15A was equal to the sum of neurons surviving with NT-3 or BDNF alone. In contrast, survival with the MNT with the R103 mutation was reduced by 80%. Unexpectedly, the neuronal survival was also reduced with the MNTs, p75-2 and p75-26. Also it is important to note that the MNTs with R103A and R103A/p75-2 mutations did not abolish completely the survival activity of the MNT, suggesting that these molecules still retain some binding properties to the Trk and p75NTR receptors. Together these data indicate that the presence in fetal DRG neurons of both Trks (TrkB, TrkC) and p75NTR receptors is important for optimal neurite outgrowth and survival in vitro. Enhanced presence of SCs and increased myelinated fibers in SC implants expressing the MTN with reduced affinity to p75NTR P75NTR can act as a co-receptor for different ligands that reflect distinct biological functions (Dechant and Barde, 2002; Teng et al., 2010). ProNTs have been found to induce apoptosis during development and following injury through the p75NTR:sortilin receptor complex (Teng et al., 2010). Therefore, we expect that the MNT with reduced affinity to p75NTR will support better engraftment of SCs and promote greater axon growth after SCI. We transplanted adult rat SCs secreting MNTs that selectively bind either the Trk or the p75NTR receptors. Because the cellular responses to NTs are highly dependent on NT concentration, we selected SCs that secreted similar amounts of the different MNTs (Supplementary Fig. 1). To identify the transplanted SCs we co-transduced with lentiviruses expressing eGFP. EGFP expression in horizontal sections revealed the presence of numerous SCs in all conditions at six weeks after transplantation (Figs. 3A–D). The cells were distributed from the injection site in a rostro-caudal direction. The majority of eGFP positive cells expressed p75NTR, a marker for non myelinating SCs (data not shown). A small portion of the cells in the transplant periphery was p75NTR positive and eGFP negative, implying the presence of migrating endogenous SCs (data not shown). But by far most of the SCs in the grafts were transplanted cells. We estimated the number of eGFP positive cells in each experimental condition, using unbiased methods (Larsen et al., 1998). On average, we found the following values; SCs infected with eGFP only, 25894 ± 5105; infected with NT-3/D15A, 135,957 ± 12749; infected with NT-3/D15A/R103A, 62,269 ± 17923; infected with p75-2, 263,988 ± 25034; infected with NT-3/D15A/R103A/p75-2, 92,410 ± 13613 (Fig. 3E). The differences among these values were statistically significant. In contrast to the embryonic DRG explants, these results suggest that the survival effects exerted by the MNTs on SCs are mediated mostly through the Trk receptors. Furthermore the cavity volume was reduced (Fig. 3F) for both NT-3/D15A and p75-2 MNTs as compared to the eGFP implants. Interestingly the transplants secreting the MNT, NT-3/D15A/R103A with reduced affinity to Trk receptors and intact sites for p75NTR binding had significantly higher cavity volumes. Taken together these results suggest that MNTs that lack p75NTR binding are more effective than the non mutated forms in supporting the survival of transplanted SCs after SCI. Transplanted SCs myelinate growing axons in the contused spinal cord (Takami et al, 2002). Neurotrophins through p75NTR activity have been implicated in inducing myelination in the PNS during development (Cosgaya et al., 2002). To determine whether the number of SC-myelinated axons differs among the various SC transplants, we counted the axons surrounded by peripheral myelin. The SC transplants were recognized by the presence of peripheral myelin in the center of the sections (dotted areas in Fig. 4A). Higher magnification

views showed the presence of a signet ring configuration of the myelinated axon and diagnostic of SC myelin (Bates et al., 2011) in contrast to oligodendrocyte myelin. The estimated numbers of myelinated axons were 4148 ± 926 for GFP, 10,711 ± 355 for NT-3/D15A, and 26,400 ± 4707 for p75-2 SC transplants. Consistent with previous results (Golden et al., 2007), the NT-3/D15A SC implants contain more myelinated fibers than the non-MNT-secreting ones. The very large p75-2 transplants contained significantly more axons with peripheral myelin than the NT3/D15A transplants (Fig. 4C). These results demonstrate that in our experimental conditions, the MNT with reduced p75NTR binding affinity significantly enhances the number of SC myelinated axons in the injury site. TrkC and not TrkA sensory fibers in MNT-secreting SC transplants NT-3, BDNF, and NGF can support the survival of DRG neurons and function as axonal attractants (Davies, 2000; Gallo and Letourneau, 1998; Huang and Reichardt, 2003). In the adult DRG, TrkA, the receptor for NGF, is expressed in nociceptive and thermoceptive sensory neurons (Liu and Ma, 2011); TrkC, the receptor for NT-3, is expressed in proprioceptive and specific types of mechanoreceptor neurons (Airaksinen et al., 1996; de Nooij et al, 2013); and TrkB, the receptor for BDNF and NT-4/5, is expressed in specific mechanosensory neurons (Li et al., 2011). To identify some of the types of sensory fibers within the SC transplants, we used antibodies that recognize parvalbumin, a marker for TrkC expressing sensory neurons (Celio, 1990; Copray et al., 1994), calcitonin gene-related peptide (CGRP) to identify TrkA expressing neurons (Averill et al., 1995) and isolectin B4 (IB4) conjugated to fluorescein to stain for non-peptidergic nociceptive fibers that express Ret (Bennett et al., 1996; Molliver et al., 1997). To date there are no specific antibody markers to identify the central axonal projections of TrkB sensory neurons, thus precluding the possibility to determine the influence of MNTs (NT-3/D15A and p75-2) on the growth of these sensory fibers. Parvalbumin-positive fibers were located in white matter around the transplants and within the transplants (Fig. 5D). Several parvalbumin positive fibers in the p75-2 transplants (Fig. 5F) appear to be singly enveloped by eGFP-labeled SCs, suggesting that several of these parvalbumin positive fibers are myelinated (Figs. 5C, E and G). When we estimated the length densities of parvalbumin positive fibers using stereological unbiased methods, we observed that p75-2 transplants contained a higher density of parvalbumin fibers within the transplant than among the SCs expressing NT-3/D15A (Fig. 5H). Contrary to the parvalbumin positive fibers within the SC transplants, most of the nociceptive CGRP and IB4 positive fibers were located in the dorsal horn, circumventing the transplants. A few CGRP positive fibers entered the periphery of the p75-2 transplants (Fig. 6B) whereas IB4 positive fibers did not enter the transplants (Figs. 6C,D). Overall the pattern of sensory fiber distribution in the grafted spinal cords showed that the p75-2 transplants enhance and attract the growth of proprioceptive afferents but not those of nociceptive fibers that depend on growth factors other than NT-3. p75-2 SC transplants enhance the growth of 5-HT supraspinal fibers BDNF and NT-3 increase the growth of several descending supraspinal pathways involved in locomotor activity such as the corticospinal and brainstem raphespinal tracts (Coumans et al., 2001; Grill et al., 1997; Jin et al., 2002; Schnell et al., 1994). Serotonergic axons descending from the brainstem raphespinal neurons are essential for modulation and locomotor rhythm generation in the spinal cord (Jordan et al., 2008; Rossignol and Dubuc, 1994; Schmidt and Jordan, 2000). To examine the growth and sparing of these fibers within and below the grafts, we quantified the 5-HT immunoreactive (IR) fibers. 5HT-IR fibers were mainly recognized in the dorso-lateral gray and white matter in the transplanted contused spinal cord. Many

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

177

Fig. 4. More myelinated fibers in transplants expressing MNT without p75NTR activity. SC myelinated axons, evident as dark rings, were present in all groups in stained plastic sections. Panels A and B show spinal cords from LV-p75-2 grafts. The dotted areas in A delineate the area of transplanted cells. Boxed area in A is shown at higher magnification in B. In panel B, the implant can be seen to be full of SC myelinated axons that are not fasciculated. (C) Quantification of myelinated axons in the transplants. Data points represent mean values ± SEM, analyzed from at least 4 animals for each group. Tukey's multiple comparison test was used. * = p b 0.05. Scale bar, 500 μm (A), 20 μm (B).

5-HT-IR fibers were close to the SC transplants (Figs. 7A,B and C) and several entered the transplants (Figs. 7a,b and c). Quantitatively, the length densities for 5-HT positive fibers were analyzed using an unbiased method in serial horizontal sections in the GFP, NT-3/D15A, and NT-3/D15A/p75-2 transplants. We divided the horizontal sections into 3 regions: contused area around the transplant, inside the SC grafts, and the caudal region next to the edge of the grafted cells. The length density for NT-3/D15A/p75-2 transplants in the contused area around the transplant was two-fold higher relative to the other

Fig. 5. Increased parvalbumin fibers in transplants expressing the MNT with reduced p75NTR affinity. Horizontal sections of spinal cords with p75-2 SC transplants expressing GFP (B, C) were labeled with an antibody against parvalbumin (D, E). The same section was labeled with DAPI (A). Merged images for B, C, and D, E are shown in color (F) and (G), respectively. Confocal microscope high power magnification image of a z-stack projection shows SC eGFP expressing cells enveloping an axon labeled with parvalbumin (red, arrow in C, E and G). (H) Quantitative analysis, within the transplant, shows that the axon length density in p75-2 transplants was significantly higher than in eGFP or NT-3/D15A grafts. Data points represent mean values ± SEM, analyzed from at least 8 animals for each group. Tukey's multiple comparison test was used. * = p b 0.05. Scale bar, 100 μm (A, B, D and F), 20 μm (C, E and G).

178

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

Fig. 6. Most sensory nociceptive fibers remain outside the p75-2 transplants. Horizontal sections of spinal cords with p75-2 SC transplants expressing eGFP (A and green in D) were labeled with an antibody against CGRP (B and red in D) and IB4 (C and blue in D). The merged image (D) shows differential distribution between CGRP and IB4 positive fibers. Arrows indicate the presence of CGRP fibers in the periphery of the transplant. Images are representative of at least 8 animals per group. Scale bar, 100 μm.

conditions (p b 0.05; Tukey's multiple comparison test) (Fig. 7D). Similarly, inside the transplant, a significant increase of fibers was found in the p75-2 versus the NT-3/D15A treated animals (p b 0.01; Mann–Whitney test) (Fig. 7E). The length density of axons in the caudal region was higher in the p75-2 transplants, but there was no statistically significant difference between the length densities in all three conditions (Fig. 7F). Similar to the results obtained with the growth of sensory fibers, transplants that secrete an MNT with reduced p75NTR binding affinity led to the presence of more descending raphespinal axons. Thus, MNTs lacking the p75NTR activity are more effective than those that bind both Trk and p75NTR, in increasing 5-HT fibers in and around the transplant.

Lack of locomotor improvement in rats transplanted with SCs expressing the MNTs In an earlier transplantation study of NT3/D15A-secreting SCs, it was observed that, despite large numbers of axons in the implant, the BBB score was not different from the eGFP-expressing SCs (Golden et al., 2007). Therefore, we asked whether the new mutant NT-3/ D15A neurotrophin delivery could improve locomotion. Functional assessments using the Basso, Beattie, and Bresnahan (BBB) Locomotor Rating Scale began on post-operative day 2 and every week after SC transplantation (Basso et al., 1995, 1996). Consistent with the Golden et al.'s (2007) report, our data showed that there were no differences in BBB scores between the transplants expressing NT-3/D15A and the other MNTs (Fig. 8).

Discussion Here we report the generation of an MNT with reduced affinity for the p75NTR receptor and testing of its effects on the survival of SCs and growth of axons in a SCI model. In our studies we used the NT-3/D15A, which carries a mutation to residue D15 and can activate TrkB just as well as its cognate ligand BDNF (Urfer et al., 1994). This specific amino acid residue does not have any structural contacts to p75NTR (Fig. 1) and therefore any mutations of NT-3/D15A residues involved in intermolecular interactions with p75NTR should be similar to those reported for NT-3. Insights into the interaction of neurotrophins with p75NTR can be gleaned by mutagenesis studies on NT-3 and crystal structures of human NT-3:rat p75NTR, human NGF:rat p75NTR and mouse proNGF: rat p75NTR (Feng et al., 2010; Gong et al., 2008; He and Garcia, 2004). Despite the differences in stoichiometry of the different crystal structure complexes among NGF, proNGF and NT-3 for the p75NTR receptor, the sites of intermolecular interactions have some common features. Both NGF and NT-3 have two main sites of interactions, sites 1 and 2, and an additional smaller site, site 3, between NT-3 and p75NTR (Gong et al., 2008; He and Garcia, 2004). Mutations of NT-3 specific residues that interact with p75NTR resulted in proteins with reduced affinity and biological activity (Urfer et al., 1994). The C-terminal residues, Arg114, Lys115 and Arg68, located in site 2, seem to be the most important for binding. In contrast, several other conserved residues in NT-3, including Arg31 from site 1, Y11 from site 2 and K73 from site 3, are of minor importance (Urfer et al.,

Fig. 7. Increase in 5-HT positive fibers in p75-2 transplants.Horizontal sections of spinal cords with p75-2 transplants expressing GFP (A and green in C) were labeled with an antibody against 5-HT; color merged images of GFP expressing cells (A) and labeled 5-HT (B) are shown in (C). Higher power magnifications of the boxed areas in A–C are shown in panels a–c. Quantitative analysis of the axon density was estimated in specific areas within and around the lesion: the contused area (D), inside the transplant (E) and 1 mm caudal to the transplant (F). The axon density of serotonergic fibers in p75-2 transplants was significantly higher than in animals with control SCs or NT-3/D15A SCs in the contused area around the transplant. Data points represent mean values ± SEM, analyzed from at least 6 animals (D). Tukey's multiple comparison test, * = p b 0.05. (E) Inside the transplants the p75-2 group contained more fibers compared to the NT-3/D15A transplant group. Data points represent mean values ± SEM, analyzed from at least 6 animals. Mann–Whitney test, ** = p b 0.01. (F) Higher axon density also was found caudal to the p75-2 transplants although the difference was not statistically significant. Tukey's multiple comparison test, p b 0.22. Scale bar, 100 μm (A, B and C), 20 μm (a, b and c).

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

1994). The importance of conserved residues K49 and Arg87, in site 1, to the overall binding affinity to p75NTR has not been determined yet. Since all the MNT variants we constructed were on an NT-3 backbone,

179

we expect the affinity of all the MNT mutants generated in this study to be the same as those of NT-3 described in our earlier study (Urfer et al., 1994).

180

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

Fig. 8. Lack of locomotor recovery with MNTs as assessed by open-field BBB scoring. Transplantation of SCs infected with GFP, NT-3/D15A, or p75-2 after moderate contusion did not improve hindlimb performance. The graph shows the average open-field BBB scores at the indicated time intervals. The arrowhead on the abscissa indicates the time of transplantation. The results are expressed as the mean ± SEM.

The p75-2 MNT revealed, using embryonic sensory neurons, that the binding to the p75NTR was essential for full biological activity. Earlier work has shown p75NTR cooperating to transduce NT signals in neurons in the developing PNS (Dechant and Barde, 2002). For example, the sensitivity to NGF and NT-3 is altered in p75-deficient DRG neurons (Buchman and Davies, 1993) and ablation of p75NTR in mice causes severe defects in both the nervous and vascular systems (Von Schack et al., 2001). When additional mutations were performed on the p75-2 backbone, the biological activity was greatly reduced (Fig. 2. and data not shown). However, the ELISA assay failed to detect the accumulation of these MNTs in the media, perhaps suggesting that conformational changes affected their antigenic epitopes. This is supported by the Western blot data revealing that p75 mutant MNT supernatants containing immunopositive bands at varying molecular masses do not accumulate in stable forms. These results indicate that changing multiple amino acids affects the processing of these MNTs through the secretory pathway and that these conserved NT residues might be important determinants for intracellular processing and stability of the precursor proteins (Mowla et al., 2001; Seidah et al., 1996). A significant effect of p75-2 MNT on SC number was discovered in the transplanted SCI animals. The number of SCs was double compared to NT-3/D15A SC implants and 4- to 5-fold higher than in SCs infected with eGFP alone. There are several mechanisms by which p75-2 could support the survival of more SCs after transplantation. First, it is known that SCs have the ability to sustain their own survival by growth factor secretion through autocrine mechanisms (Mirsky and Jessen, 1999). For example, insulin-like growth factor 2 (IGF2), PDGF and NT-3 act synergistically to block SC death (Meier et al., 1999). Second, although we did not demonstrate that the infected SCs produced the proNT form of p75-2 MNT, the data suggest that decreased affinity of the proMNT form to p75NTR might be in part responsible for more SC survival in the injured spinal cord. Nonmyelinating SCs express high levels of p75NTR (Jessen and Mirsky, 1991) and both proBDNF and proNT-3 can induce proapoptotic signals in both glia and neurons (Teng et al., 2010). Further, p75NTR is involved in SC death in vitro and after axotomy (Syroid et al., 2000). Support of the idea that a proMNT form is produced by SCs is that the cavity volume is increased by 1- to 2 fold in animals receiving SCs secreting NT-3/D15A/R103A; because this MNT binds preferentially to the p75NTR perhaps apoptosis is induced in the grafted spinal cords.

We do not yet know to what extent proliferation played a role in the increase in SC number. Because neurotrophins exert a potent neurotropic effect on sensory neurons, the presence of more sensory fibers in the graft may have stimulated the proliferation of SCs. Immunocytochemical methods to detect SCs in the G1 phase of the cell cycle at 6 weeks failed to reveal any differences between the various groups (data not shown). However, a proliferative effect of the axons on SCs could be exerted before the onset of myelination; earlier time points need to be examined to assess the extent of proliferation. Analyses in vitro and in vivo of the temporal induction of the MNT forms in SCs might be necessary to clarify their roles in SC survival and proliferation. Previous studies have shown that NT-3 signaling through the TrkC inhibits myelination of SCs whereas BDNF acting on p75NTR promotes myelination (Chan et al., 2001; Cosgaya et al., 2002). Further, peripheral nerves in p75NTR deficient mice are hypomyelinated and do not respond to the myelin-promoting effects of BDNF. Because the p75-2 MNT binds mainly to the TrkB and TrkC but not p75NTR receptors, we would have expected that, in the p72-2 MNT transplanted animals, the balance of signaling should be in favor of non-myelinating signals, resulting in fewer myelinated axons within the transplant. Thus, our data suggest that the dichotomy between BDNF promoting and NT-3 inhibiting myelination in developing peripheral nerves may not apply to adult rat SC transplantation. The combinatorial treatment of SCs with the p75-2 MNT promoted a considerable amount of fiber growth into the SC graft from sensory proprioceptive afferents and the serotonergic pathway. Previous studies have shown that adult sensory axons respond to NT-3 and that these axons can invade the injured spinal cord (Bradbury et al., 1999; Ramer et al., 2000). Similar to the large peripheral NF200 positive fibers that are myelinated by SCs, several of these parvalbumin positive large axons also were myelinated by SCs. These findings are in agreement with previous studies showing a tropic effect of exogenously provided NT-3 on sensory fibers after SCI (Lu et al., 2003; Taylor et al., 2006). This ability has been exploited to restore sensory connectivity to the correct CNS targets (Alto et al., 2009). Furthermore, the absence of CGRP and IB-4 nociceptive fibers inside the graft suggests that the p72-2 MNT tropic effects are specific to the neurons expressing TrkC. An important finding of the present work is that 5-HT positive axons in p75-2 MNT-secreting transplants were doubled compared to NT-3/D15A MNT-secreting SC implants. These results are consistent with previous experiments using similar approaches showing that the growth of 5-HT positive fibers can be stimulated by NTsecreting SCs (Golden et al., 2007). Most of the 5-HT fibers grew in and around the transplant indicating a permissive environment for supraspinal fibers. Even though we found a large number of fibers caudal to the graft, it is unclear whether these fibers enter and emerge from the graft. Taken together these results clearly show that the MNT with reduced affinity to p75NTR is a more potent inducer of axon growth and/or sparing for sensory and supraspinal fibers in SCI than the NT-3/D15A SCs. This might be a direct reflection on the number of surviving SCs or the lack of competitive neurotrophin binding by SCs expressing p75NTR. Because myelinating SCs express high levels of p75NTR, it is significant that an MNT with lower affinity to p75NTR could be more abundantly available for axon growth especially at early stages after SC implantation. The question remains whether the p75NTR presence in axons plays a role in the more robust growth that we see in the p75-2 secreting SCs. Previous reports showed that in p75NTR knockout mice the CNS axons fail to regenerate after injury (Boyd and Gordon, 2001; Ferri et al., 1998). Therefore these earlier reports argue that the axonal expression of p75NTR might not be as important for the increased growth of axons. A recent report has asserted that the p75NTR in SCs prevents spontaneous sensory reinnervation (Scott and Ramer, 2010), thus indicating that reducing the activity of p75NTR in SCs might be necessary to enable more robust growth of regenerating axons.

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

Despite the very significant increase in SC number and myelinated axons and more robust axonal outgrowth of sensory and supraspinal fibers, the locomotor recovery was not improved compared to GFP-SC grafts. However these results provide a better opportunity for restoring function once this stratagem of NT promise is combined with other newly developing interventions to enable fiber growth out of the graft into the spinal cord (Williams and Bunge, 2013). Also, our novel results using newly created MNTs demonstrate the feasibility of presenting one modified MNT instead of multiple NTs for spinal cord repair, an advance that could simplify clinical applicability.

Conclusions Results presented in a previous report (Golden et al., 2007) and in the present study show that neurotrophins that can bind more than one Trk receptor at once are more effective in supporting the survival of SCs and the growth of axons into the SC implant. Also, in this manuscript we show that a neurotrophin lacking the capacity to bind to the p75NTR, p75-2, leads to a higher number of transplanted adult SCs in a SCI rat model than does the unmodified neurotrophin. Also, p75-2 supports the regeneration/sparing of more supraspinal and sensory axons. We suggest, therefore, that for more effective therapeutic outcomes, modified neurotrophins that lack p75NTR activity should be used in combination with other interventions such as modifying the scar that surrounds the implant and rehabilitation strategies. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.expneurol.2013.06.013.

Acknowledgments This work was supported in part by the National Institutes of Health grant FORE-SCI N01-NS-3-2351, the Miami Project to Cure Paralysis and the Buoniconti Fund to Cure Paralysis. We are grateful to Dr. Dalton Dietrich for his advice and generous support. We thank Dr. Jae Lee for suggestions to improve the manuscript, the Miami Project Animal Core for contusions and post-operative animal care, Dr. Andres Hurtado for technical advice on spinal cord surgery, Dr. Beata Frydel for advice regarding stereological methods and, Margaret Bates and Yungfang Wang for histological expertise. The authors declare no competing financial interests. References Airaksinen, M.S., Koltzenburg, M., Lewin, G.R., Masu, Y., Helbig, C., Wolf, E., Brem, G., Toyka, K.V., Thoenen, H., Meyer, M., 1996. Specific subtypes of cutaneous mechanoreceptors requireneurotrophin-3 following peripheral target innervation. Neuron 16, 287–295. Alto, L.T., Havton, L.A., Conner, J.M., Hollis Ii, E.R., Blesch, A., Tuszynski, M.H., 2009. Chemotropic guidance facilitates axonal regeneration and synapse formation after spinal cord injury. Nat. Neurosci. 12, 1106–1113. Averill, S., McMahon, S.B., Clary, D.O., Priestley, J.V., Reichardt, L.F., 1995. Immunocytochemical localization of TrkA receptors in chemically identified subgroups of adult rat sensory neurons. Eur. J. Neurosci. 7, 1484–1494. Basso, D.M., Beattie, M.S., Bresnahan, J.C., 1995. A sensitive and reliable locomotor rating scale for open field testing in rats. J. Neurotrauma 12, 1–21. Basso, D.M., Beattie, M.S., Bresnahan, J.C., 1996. Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Exp. Neurol. 139, 244–256. Bates, M.L., Puzis, R., Bunge, M.B., 2011. Preparation of spinal cord injured tissue for light and electron microscopy including preparation for immunostaining. In: Lane, E.L., Dunnett, S.B. (Eds.), Conventional Animal Models of Movement Disorders. Neuromethods, 62. Springer Humana, N.Y., pp. 381–399. Bennett, D.L., Averill, S., Clary, D.O., Priestley, J.V., McMahon, S.B., 1996. Postnatal changes in the expression of the trkA high-affinity NGF receptor in primary sensory neurons. Eur. J. Neurosci. 8, 2204–2208. Boyd, J.G., Gordon, T., 2001. The neurotrophin receptors, trkB and p75, differentially regulate motor axonal regeneration. J. Neurobiol. 49, 314–325. Bradbury, E.J., Khemani, S., King, V.R., Priestley, J.V., McMahon, S.B., 1999. NT-3 promotes growth of lesioned adult rat sensory axons ascending in the dorsal columns of the spinal cord. Eur. J. Neurosci. 11, 3873–3883.

181

Bregman, B.S., McAtee, M., Dai, H.N., Kuhn, P.L., 1997. Neurotrophic factors increase axonal growth after spinal cord injury and transplantation in the adult rat. Exp. Neurol. 148, 475–494. Buchman, V.L., Davies, A.M., 1993. Different neurotrophins are expressed and act in a developmental sequence to promote the survival of embryonic sensory neurons. Development 118, 989–1001. Cao, Q., Xu, X.M., Devries, W.H., Enzmann, G.U., Ping, P., Tsoulfas, P., Wood, P.M., Bunge, M.B., Whittemore, S.R., 2005. Functional recovery in traumatic spinal cord injury after transplantation of multineurotrophin-expressing glial-restricted precursor cells. J. Neurosci. 25, 6947–6957. Celio, M.R., 1990. Calbindin D-28k and parvalbumin in the rat nervous system. Neuroscience 35, 375–475. Chan, J.R., Cosgaya, J.M., Wu, Y.J., Shooter, E.M., 2001. Neurotrophins are key mediators of the myelination program in the peripheral nervous system. Proc. Natl. Acad. Sci. U. S. A. 98, 14661–14668. Chao, M.V., 1994. The p75 neurotrophin receptor. J. Neurobiol. 25, 1373–1385. Copray, J.C., Mantingh-Otter, I.J., Brouwer, N., 1994. Expression of calcium-binding proteins in the neurotrophin-3-dependent subpopulation of rat embryonic dorsal root ganglion cells in culture. Brain Res. Dev. Brain Res. 81, 57–65. Cosgaya, J.M., Chan, J.R., Shooter, E.M., 2002. The neurotrophin receptor p75NTR as a positive modulator of myelination. Science 298, 1245–1248. Coumans, J.V., Lin, T.T., Dai, H.N., MacArthur, L., McAtee, M., Nash, C., Bregman, B.S., 2001. Axonal regeneration and functional recovery after complete spinal cord transection in rats by delayed treatment with transplants and neurotrophins. J. Neurosci. 21, 9334–9344. Davies, A.M., 2000. Neurotrophins: more to NGF than just survival. Curr. Biol. 10, R374–R376. Dechant, G., Barde, Y.A., 2002. The neurotrophin receptor p75(NTR): novel functions and implications for diseases of the nervous system. Nat. Neurosci. 5, 1131–1136. de Nooij, J.C., Doobar, S., Jessell, T.M., 2013. Etv1 inactivation reveals proprioceptor subclasses that reflect the level of NT3 expression in muscle targets. Neuron 77, 1055–1068. Dull, T., Zufferey, R., Kelly, M., Mandel, R.J., Nguyen, M., Trono, D., Naldini, L., 1998. A third-generation lentivirus vector with a conditional packaging system. J. Virol. 72, 8463–8471. Feng, D., Kim, T., Ozkan, E., Light, M., Torkin, R., Teng, K.K., Hempstead, B.L., Garcia, K.C., 2010. Molecular and structural insight into proNGF engagement of p75NTR and sortilin. J. Mol. Biol. 396, 967–984. Ferri, C.C., Moore, F.A., Bisby, M.A., 1998. Effects of facial nerve injury on mouse motoneurons lacking the p75 low-affinity neurotrophin receptor. Brain Res. Rev. 57, 183–191. Follenzi, A., Naldini, L., 2002. Generation of HIV-1 derived lentiviral vectors. In: Phillips, M.I. (Ed.), Methods in Enzymology: Gene Therapy Methods. Academic Press, San Diego, London, pp. 454–465. Fortun, J., Hill, C.E., Bartlett Bunge, M., 2009. Combinatorial strategies with Schwann cell transplantation to improve repair of the injured spinal cord. Neurosci. Lett. 456, 124–132. Gallo, G., Letourneau, P.C., 1998. Localized sources of neurotrophins initiate axon collateral sprouting. J. Neurosci. 18, 5403–5414. Golden, K.L., Pearse, D., Blits, B., Garg, M.S., Oudega, M., Wood, P.W., Bartlett Bunge, M., 2007. Transduced Schwann cells promote axon growth and myelination after spinal cord injury. Exp. Neurol. 207, 203–217. Gong, Y., Cao, P., Yu, H.J., Jiang, T., 2008. Crystal structure of the neurotrophin-3 and p75NTR symmetrical complex. Nature 454, 789–793. Grill, R., Murai, K., Blesch, A., Gage, F.H., Tuszynski, M.H., 1997. Cellular delivery of neurotrophin-3 promotes corticospinal axonal growth and partial functional recovery after spinal cord injury. J. Neurosci. 17, 5560–5572 (PW). He, X.L., Garcia, K.C., 2004. Structure of nerve growth factor complexed with the shared neurotrophin receptor p75. Science 304, 870–875. Hiebert, G.W., Khodarahmi, K., McGraw, J., Steeves, J.D., Tetzlaff, W., 2002. Brainderived neurotrophic factor applied to the motor cortex promotes sprouting of corticospinal fibers but not regeneration into a peripheral nerve transplant. J. Neurosci. Res. 69, 160–168. Huang, E.J., Reichardt, L.F., 2003. Trk receptors: roles in neuronal signal transduction. Annu. Rev. Biochem. 72, 609–642. Ibáñez, C.F., Ilag LL, L.L., Murray-Rust, J., Persson, H., 1993. An extended surface of binding to Trktyrosine kinase receptors in NGF and BDNF allows the engineering of a multifunctional pan-neurotrophin. EMBO J. 12, 2281–2293. Jessen, K.R., Mirsky, R., 1991. Schwann cell precursors and their development. Glia 4, 185–194. Jin, Y., Fischer, I., Tessler, A., Houle, J.D., 2002. Transplants of fibroblasts genetically modified to express BDNF promote axonal regeneration from supraspinal neurons following chronic spinal cord injury. Exp. Neurol. 177, 265–275. Jordan, L.M., Liu, Jun, Hedlund, Peter B., Akay, Turgay, Keir, G., Pearson, K.G., 2008. Descending command systems for the initiation of locomotion in mammals. Brain Res. Rev. 57, 183–191. Kleitman, N., Wood, P., Johnson, M.I., Bunge, R.P., 1998. Schwann cell surfaces but not extracellular matrix organized by Schwann cells support neurite outgrowth from embryonic rat retina. J. Neurosci. 8, 653–666. Kobayashi, N.R., Fan, D.P., Giehl, K.M., Bedard, A.M., Wiegand, S.J., Tetzlaff, W., 1997. BDNF and NT-4/5 prevent atrophy of rat rubrospinal neurons after cervical axotomy, stimulate GAP-43 and Talpha1-tubulin mRNA expression, and promote axonal regeneration. J. Neurosci. 17, 9583–9595. Larsen, J., Gundersen, H., Nielsen, J., 1998. Global spatial sampling with isotropic virtual planes: estimators of length density and total length in thick, arbitrarily orientated sections. J. Microsc. 191, 238–248.

182

M. Enomoto et al. / Experimental Neurology 248 (2013) 170–182

Li, L., Rutlin, M., Abraira, V.E., Cassidy, C., Kus, L., Gong, S., Jankowski, M.P., Luo, W., Heintz, N., Koerber, R.H., Woodbury, J.C., Ginty, D.D., 2011. The functional organization of cutaneous low-threshold mechanosensory neurons. Cell 147, 1615–1627. Liu, Y., Ma, Q., 2011. Generation of somatic sensory neuron diversity and implications on sensory coding. Curr. Opin. Neurobiol. 21, 52–60. Lu, P., Tuszynski, M.H., 2008. Growth factors and combinatorial therapies for CNS. Exp. Neurol. 209, 313–320. Lu, P., Jones, L.L., Snyder, E.Y., Tuszynski, M.H., 2003. Neural stem cells constitutively secrete neurotrophic factors and promote extensive host axonal growth after spinal cord injury. Exp. Neurol. 181, 115–129. Meier, C., Parmantier, E., Brennan, A., Mirsky, R., Jessen, K.R., 1999. Developing Schwann cells acquire the ability to survive without axons by establishing an autocrine circuit involving insulin-like growth factor, neurotrophin-3, and plateletderived growth factor-BB. J. Neurosci. 19, 3847–3859. Mirsky, R., Jessen, K.R., 1999. The neurobiology of Schwann cells. Brain Pathol. 9, 293–311. Molliver, D.C., Wright, D.E., Leitner, M.L., Parsadanian, A.S., Doster, K., Wen, D., Yan, Q., Snider, W.D., 1997. IB4-binding DRG neurons switch from NGF to GDNF dependence in early postnatal life. Neuron 19, 849–861. Morrissey, T.K., Kleitman, N., Bunge RP, R.P., 1991. Isolation and functional characterization of Schwann cells derived from adult peripheral nerve. J. Neurosci. 11, 2433–2442. Mowla, S.J., Farhadi, H.F., Pareek, S., Atwal, J.K., Morris, S.J., Seidah, N.G., Murphy, R.A., 2001. Biosynthesis and post-translational processing of the precursor to brainderived neurotrophic factor. J. Biol. Chem. 276, 12660–12666. Oudega, M., Xu, X.M., 2006. Schwann cell transplantation for repair of the adult spinal cord. J. Neurotrauma 23, 453–467. Ramer, M.S., Priestley, J.V., McMahon, S.B., 2000. Functional regeneration of sensory axons into the adult spinal cord. Nature 403, 312–316. Rossignol, S., Dubuc, R., 1994. Spinal pattern generation. Curr. Opin. Neurobiol. 4, 894–902. Sayer, F.T., Oudega, M., Hagg, T., 2002. Neurotrophins reduce degeneration of injured ascending sensory and corticospinal motor axons in adult rat spinal cord. Exp. Neurol. 175, 282–296. Schmidt, B.J., Jordan, L.M., 2000. The role of serotonin in reflex modulation and locomotor rhythm production in the mammalian spinal cord. Brain Res. Bull. 53, 689–710. Schnell, L., Schneider, R., Kolbeck, R., Barde, Y.A., Schwab, M.E., 1994. Neurotrophin-3 enhances sprouting of corticospinal tract during development and after adult spinal cord lesion. Nature 367, 170–173. Scott, A.L.M., Ramer, M.S., 2010. Schwann cell p75NTR prevents spontaneous sensory reinnervation of the adult spinal cord. Brain 28, 1611–1682. Seidah, N.G., Benjannet, S., Pareek, S., Chrétien, M., Murphy, R.A., 1996. Cellular processing of the neurotrophin precursors of NT3 and BDNF by the mammalian proprotein convertases. FEBS Lett. 379, 247–250. Syroid, D.E., Maycox, P.J., Soilu-Hänninen, M., Petratos, S., Bucci, T., Burrola, P., Murray, S., Cheema, S., Lee, K.F., Lemke, G., Kilpatrick, T.J., 2000. Induction of postnatal Schwann cell death by the low-affinity neurotrophin receptor in vitro and after axotomy. J. Neurosci. 20, 5741–5747.

Takami, T., Oudega, M., Bates, M.L., Wood, P.W., Kleitman, N., Bunge, M.B., 2002. Schwann cell but not olfactory ensheathing glia transplants improve hindlimb locomotor performance in the moderately contused adult rat thoracic spinal cord. J. Neurosci. 22, 6670–6681. Taylor, L., Jones, L., Tuszynski, M.H., Armin Blesch, A., 2006. Neurotrophin-3 gradients established by lentiviral gene delivery promote short-distance axonal bridging beyond cellular grafts in the injured spinal cord. J. Neurosci. 26, 9713–9721. Teng, K.K., Felice, S., Kim, T., Hempstead, B.L., 2010. Understanding proneurotrophin actions: recent advances and challenges. Dev. Neurobiol. 70, 350–359. Tetzlaff, W., Okon, E.B., Karimi-Abdolrezaee, S., Hill, C.E., Sparling, J.S., Plemel, J.R., Plunet, W.T., Tsai, E.C., Baptiste, D., Smithson, L.J., Kawaja, M.D., Fehlings, M.G., Kwon, B.K., 2012. A systematic review of cellular transplantation therapies for spinal cord injury. J. Neurotrauma 28, 1611–1682. Tsoulfas, P., Stephens, R.M., Kaplan DR, D.R., Parada, L.F., 1996. TrkC isoforms with inserts in the kinase domain show impaired signaling responses. J. Biol. Chem. 271, 5691–5697. Tuszynski, M.H., Peterson, D.A., Ray, J., Baird, A., Nakahara, Y., Gage, F.H., 1994. Fibroblasts genetically modified to produce nerve growth factor induce robust neuritic ingrowth after grafting to the spinal cord. Exp. Neurol. 126, 1–14. Urfer, R., Tsoulfas, P., Soppet, D., Escandón, E., Parada, L.F., Presta, L.G., 1994. The bindingepitopes of neurotrophin-3 to its receptors trkC and gp75 and the design of amultifunctional human neurotrophin. EMBO J. 13, 5896–5909. Urfer, R., Tsoulfas, P., O'Connell, L., Presta, L.G., 1997. Specificity determinants in neurotrophin-3 and design of nerve growth factor-based trkC agonists by changing central beta-strand bundle residues to their neurotrophin-3 analogs. Biochemistry 36, 4775–4781. von Schack, D., Casademunt, E., Schweigreiter, R., Meyer, M., Bibel, M., Dechant, G., 2001. Complete ablation of the neurotrophin acceptor p75NTR causes both in the nervous and the vascular system. Nat. Neurosci. 10, 977–988. Wiesmann, C., de Vos, A.M., 2001. Nerve growth factor: structure and function. Cell. Mol. Life Sci. 201, 748–759. Williams, R.R., Bunge, M.B., 2013. Schwann cell transplantation: a repair strategy for spinal cord injury? In Functional Neural Transplantation III. In: Dunnett, S.B., Bjorklund, A. (Eds.), Primary and Stem Cell Transplantation for Brain Repair Part 2 (in press). Xu, X.M., Guénard, V., Kleitman, N., Bunge, M.B., 1995a. Axonal regeneration into Schwann cell seeded guidance channels grafted into transected adult rat spinal cord. J. Comp. Neurol. 351, 145–160. Xu, X.M., Guénard, V., Kleitman, N., Aebischer, P., Bartlett Bunge, M., 1995b. A combination of BDNF and NT-3 promotes supraspinal axonal regeneration into Schwann cell grafts in adult rat thoracic spinal cord. Exp. Neurol. 134, 261–272. Xu, X.M., Chen, A., Guenard, V., Kleitman, N., Bartlett Bunge, M., 1997. Bridging Schwann cell transplants promote axonal regeneration from both the rostral and caudal stumps of transected adult rat spinal cord. J. Neurocytol. 26, 1–16. Yano, H., Torkin, R., Andrés Martin, L., Chao, M.V., Teng, K.K., 2009. Proneurotrophin-3 is a neuronal apoptotic ligand: evidence for retrograde-directed cell killing. J. Neurosci. 29, 14790–14802.