Further characterization of the effects of brain-derived neurotrophic factor and ciliary neurotrophic factor on axotomized neonatal and adult mammalian motor neurons

Download Further characterization of the effects of brain-derived neurotrophic factor and ciliary neurotrophic factor on axotomized neonatal and adult mammalian motor neurons

Post on 11-Jun-2016




0 download

Embed Size (px)


<ul><li><p>THE JOURNAL OF COMPARATIVE NEUROLOGY 342~45-56 (1994) </p><p>Further Characterization of the Effects of Brain-Derived Neurotrophic Factor and </p><p>Ciliary Neurotrophic Factor on Axotomized Neonatal and Adult </p><p>Mammalian Motor Neurons </p><p>RICHARD E. CLATTERBUCK, DONALD L. PRICE, AND VASSILIS E. KOLIATSOS Departments of Pathology, Neurology (D.L.P., V.E.K.), and Neuroscience and the </p><p>Neuropathology Laboratory (R.E.C., D.L.P., V.E.K.), The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205-2196 </p><p>ABSTRACT Neurotrophins and neural cytokines are two broad classes of neurotrophic factors. It has </p><p>been reported that ciliary neurotrophic factor (CNTF) and brain-derived neurotrophic factor (BDNF) prevent the degeneration of axotomized neonatal motor neurons. In addition, BDNF is transported retrogradely to a-motor neurons following injection into the muscle, and patterns of BDNF expressed in spinal cord and muscle suggest a physiological role for this factor in motor neurons. In the present study, we characterize the effects of BDNF on axotomized neonatal facial motor neurons and extend these observations to adult models of motor neuron injury (axotomy-induced phenotypic injury of lumbar motor neurons). BDNF reduces axotomy- induced degeneration of neonatal neurons by 55% as determined by Nissl staining (percentage of surviving neurons in vehicle-treated cases, 25%; in BDNF-treated cases, 80%). Rescued neurons have an intact organelle structure but appear smaller and slightly chromatolytic on electron microscopic analysis. As demonstrated by intense retrograde labeling with horseradish peroxidase (HRP) applied to the proximal stump of the facial nerve, neurons rescued by BDNF have intact mechanisms of fast axonal transport. CNTF did not appear to have significant effects on neonatal motor neurons, but the lack of efficacy of this factor may be caused by its rapid degradation at the application site. BDNF is not capable of reversing the axotomy-induced reduction in transmitter markers [i.e., the acetylcholine-synthesizing enzyme choline acetyl- transferase ( C U T ) or the degrading enzyme acetylcholinesterase (AChE)] in neonatal or adult animals or the axotomy-induced up-regulation of the low-affinity neurotrophin receptor p75NGFR (nerve growth factor receptor) in adult motor neurons. However, BDNF appears to promote the expression of p7ijNGFR in injured neonatal motor neurons. In concert, the findings of the present study suggest that BDNF can significantly prevent cell death in injured motor neurons. However, this neurotrophin may not be a retrograde signal associated with the induction and/or maintenance of some mature features of motor neurons, particularly their transmitter phenotype. </p><p>Key words: cell death, cytokines, facial nucleus, nerve growth factor receptor, neurotrophins </p><p>D 1994 Wiley-Liss, Inc. </p><p>Neurotrophic factors are classified broadly into neuro- trophins and neural cytokines. Neurotrophins, a family of peptides with &gt; 60% amino acid homology, are transported retrogradely from the central nervous system or peripheral targets to support the survival and induce or maintain the mature phenotype of certain populations of neurons express- ing specific receptors. Nerve growth factor (NGF) is the prototypical member of this family (Barde et al., 1982; Levi-Montalcini, 1987; Lo, 1992) that includes also brain- </p><p>derived neurotrophic factor (BDNF) (Leibrock et al., 1989; Rosenthal et al., 1991), neurotrophin-3 (NT-3; Ernfors et al., 1990; Hohn et al., 1990; Maisonpierre et al., 1990; Rosenthal et al., 1990), and NT-415 (Berkemeier et al., </p><p>Accepted October 12,1993. Address reprint requests to Vassilis E. Koliatsos, MD, Nenropathology </p><p>Laboratory, The Johns Hopkins University School of Medicine, 558 Ross Research Building, 720 Rutland Avenue, Baltimore, MD 21205-2196. </p><p>0 1994 WILEY-LISS, INC. </p></li><li><p>46 R.E. CLATTERBUCK ET AL. </p><p>1991; Hallbook et al., 1991; Ip et al., 1992). Neural cyto- kines are a group of heterogeneous pleiotropic peptides (fibroblast growth factors, interleukins, transforming growth factors-fi, leukemia inhibitory factor, and CNTF), many of which (e.g., fibroblast growth factors) have been known previously for their effects on nonneural cells (Hil- ton, 1992; Unsicker et al., 1992). Neural cytokines and some of their receptors are expressed ubiquitously and may share some of the properties of neurotrophins in preventing neuronal cell death (Sendtner et al., 1990; Clatterbuck et al., 1993) or maintaining phenotype, but their functions are still poorly understood. </p><p>Various trophic factors and muscle preparations, puta- tively enriched in the appropriate factors, have been used to promote the survival of motor neurons in vivo and in vitro. Skeletal muscle extracts enhance the survival of motor neurons in dissociated cultures from the embryonic chick and rat and in spinal cord explants (Dohrmann et al., 1986, 1987; Smith et al., 1986; Appel et al., 1989). These effects are augmented when extracts are prepared from denervated muscle (Nurcombe et al., 1984). The neural cytokine CNTF has been shown to prevent axotomy-induced cell death in rat neonatal motor neurons (Sendtner et al., 1990) and developmental cell death in avian motor neurons (Oppen- heim et al., 1991). However, its effects on axotomized neonatal motor neurons have not been replicated (Snider et al., 1992; Unsicker et al., 19921, and its observed effects in a mouse mutant with progressive motor axonopathy have not been characterized adequately (Sendtner et al., 1992a). </p><p>Earlier studies on the role of NGF in the normal develop- ment and maintenance of motor neurons were inconclusive (Wayne and Heaton, 1988; Yan et al., 1988). Although motor neurons of early chick embryos and neonatal rats were shown to transport NGF from their target fields (Yan et al., 1988), NGF did not prevent developmental motor neuron death in the chick (Oppenheim et al., 1982) or axotomy-induced death of mammalian neonatal motor neu- rons (Miyata et al., 1986; Yan et a]., 1988). However, very recent investigations employing non-NGF neurotrophins have had a very positive outcome. First, it has been shown that the neurotrophins BDNF and NT-3 are transported to lumbar motor neurons following injections into the dam- aged sciatic nerve of the adult (DiStefano et al., 1992a) and the hindlimb or footpad of the neonatal rat (Yan et al., 1992; Koliatsos et al., 1993). Second, four independent groups have shown that BDNF prevents retrograde cell death of neonatal motor neurons in the facial nucleus and lumbar cord (Yan et al., 1992; Sendtner et al., 199213; Koliatsos et al., 1993) as well as developmental cell death of lumbar motor neurons in the chick (Oppenheim et al., 1992). </p><p>The present study was designed to characterize the effects of BDNF on degenerating neonatal motor neurons at the ultrastructural and functional levels and to compare these effects with those of CNTF. We also assessed whether BDNF can prevent axotomy-induced phenotypic injury in adult motor neurons by employing transections of the sciatic nerve. Our findings indicate that BDNF, but not CNTF, prevents retrograde degeneration of motor neurons; motor neurons rescued by BDNF have near-normal ultra- structure and maintain active retrograde transport. </p><p>MATERIALS AND METHODS Neonatal facial nerve transection: </p><p>Histological studies PO male Sprague-Dawley rats were anesthetized by hypo- </p><p>thermia. Using aseptic procedures, the left facial nerve, </p><p>including its posterior auricular branch, was transected 1-1.5 mm distal to the stylomastoid foramen. A small piece of gelfoam (2 x 2 mm), immersed in human recombinant BDNF (10-20 pg, n = 5) or human recombinant CNTF (10-20 pg, n = 5) in phosphate-buffered saline (PBS), was apposed to the proximal stump, and the overlying skin was sutured with 5-0 silk. Control pups were treated with vehicle solution (n = 3). Wounds were inspected daily for signs of dehiscence or infection. </p><p>One week after surgery, animals were deeply anesthe- tized with chloral hydrate (400 mg/kg) and perfused trans- cardially with 0.1 M PBS (1-2 minute flush), followed by 100-200 ml of 4% freshly depolymerized paraformalde- hyde. The brains were removed, blocked, and postfixed in 4% paraformaldehyde for at least 48 hours. Blocks contain- ing the caudal pons/medulla were dehydrated in graded alcohols and embedded in paraffin. Serial coronal sections (7 pm) were stained with cresyl violet. </p><p>Three sections through the facial nucleus of each animal were taken for neuronal counts (BDNF and CNTF, five cases each; vehicle, three cases). One section corresponded to the middle of the facial nucleus, and the two other sections corresponded to planes at equal anterior and posterior distances from the middle level on each side of the brain. Nucleolated profiles belonging to a-motor neurons were counted at x 40 magnification, using a computerized image analysis system; counts were corrected for split cells with the Abercrombie formula (Abercrombie, 1946). Cell numbers from the lesioned facial nuclei were expressed as percentages of cell numbers in the contralateral unlesioned nuclei. For statistical evaluation of the effects of various treatments on axotomized motor neurons, an analysis of variance was computed by using the average percentage value for each of the three groups. The Newman-Keuls multiple range posthoc test was used to test for simple effects. </p><p>Ultrastructural studies For electron microscopic analysis of axotomized, BDNF- </p><p>treated neurons of the neonatal facial nucleus, perfusion fixation followed a protocol adapted for developing neural tissues (n = 3). Blood was cleared with a brief rinse with PBS (30 seconds, 37C) that was bubbled with a mixture of 02/C02 (95%:5%) for 30 minutes prior to the perfusion. Perfusion fixation was performed using a neutral-buffered solution of 1% paraformaldehyde and 1% glutaraldehyde for 5-8 minutes (37"C), followed by 4% glutaraldehyde for 20 minutes (4C). Brains were removed and postfixed in the former fixative overnight (4C). Small tissue slabs at the pontomedullary junction were subdissected from brainstem blocks, stained with 2% osmium tetroxide for 1 hour, and dehydrated in graded alcohols. Blocks were then embedded in an Araldite-based plastic for thin sectioning. Thin sec- tions counterstained with uranyl acetateilead citrate were studied with a Hitachi 600 electron microscope. </p><p>Studies of axonal transport Animals were treated with BDNF (n = 3) or vehicle (n = </p><p>31, as described above (under Neonatal Facial Nerve Tran- section). On day 6, gelfoam pieces immersed in BDNF or vehicle were replaced by identical pieces immersed in 10% horseradish peroxidase (HRPI-wheat germ agglutinin, fol- lowing visualization of the nerve stump. On day 7, animals were perfused essentially as in the histology group (see above), with 3% neutral-buffered paraformaldehyde. Brains were postfixed overnight in 2% neutral-buffered paraform- </p></li><li><p>BDNF EFFECTS ON MOTOR NEURONS 47 </p><p>aldehyde, immersed for 24 hours in 20% glycerol in 0.1 M phosphate buffer, frozen in isopentane, and sectioned on a sliding microtome (40 pm). Sections through the pons/ medulla were processed in series for cresyl violet and HRP histochemistry, using the tetramethylbenzidine (TMB) pro- cedure according to standard protocols (Mesulam, 1982). </p><p>CNTF bioactivity studies Gelfoam pieces containing 10 pg CNTF (n = 26) or PBS </p><p>(n = 2) were implanted subcutaneously in the postauricular region in PO male Sprague-Dawley rats. These gelfoams were recovered at days 1 (n = 6), 3 (n = 6 ) , and 7 (n = 6) postimplantation frozen on dry ice, and stored at -70C. Some gelfoam pieces were immersed in CNTF (n = 8), and all gelfoams containing PBS were frozen immediately follow- ing impregnation with the appropriate solution. Subse- quently, gelfoams were thawed and incubated in culture media for 2-3 hours. Media were then assayed for the ability to promote ciliary ganglion neuron survival, as described previously (Gupta et al., 1992). </p><p>Phenotypic studies in neonatal and adult animals: Immunocytochemistry for ChAT and </p><p>p75NGFR and histochemistry for AChE PO pups were treated with BDNF (n = 2) or vehicle (n = </p><p>21, as reported above. One week later, animals were per- fused transcardially with 3% paraformaldehyde (see above). Tissue blocks that contained the pontomedullary region were postfixed overnight in 2% paraformaldehyde and then immersed in 20% glycerol in 0.1 M phosphate buffer for 24 hours and frozen in isopentane. Transverse sections were cut on a sliding microtome (40 pm) and processed in series for Nissl, AChE histochemistry, and p75NGFR immunocyto- chemistry using the monoclonal antibody 192-IgG. </p><p>Six adult male Sprague-Dawley rats (250-350 g) were subjected to a unilateral transection of the left sciatic nerve at a level corresponding to the middle of the sacroiliac joint ( - 50 mm from the origin of motor axons) in the L4 and L5 spinal segments, as reported previously (Koliatsos et al., 1991a). Immediately following the transection, the catheter of a ventricular access device (Model 44100; Connell Neuro- surgical, Exton, PA) was lowered at the transection through a passage between the sacral and iliac bones constructed by removing a small piece of bone at symmetrical points across the sacroiliac joint. The reservoir was placed on the superfi- cial fascia of the gluteus maximus, and the reservoir/ cannula was secured in place with sutures that passed through the flange of the reservoir and the fascia; other sutures were used to affix the genu of the cannula to the muscle. The gluteus medius and maximus were sutured in layers with Dexon, and the skin was sutured with silk (vehicle, n = 3; BDNF, n = 3). Subjects treated with BDNF were injected immediately after surgery and then every second day for a total of four applications with 10 pg of BDNF in 300 pl of acidified PBS, followed by washing of the reservoir with an additional 300 p1 of PBS. Controls were treated with 300 p1 of acidified PBS. On the day following the last treatment, all animals were perfused with 3% paraformaldehyde, and L4-L5 tissue blocks were frozen and processed in series for Nissl, C U T , and ~ 7 5 ~ ~ ~ ~ . ChAT and p75NGFR immunocytochemistries were performed on floating sections. Sections were incubated for 30 minutes in 0.4% Triton-X 100 in 0.05 M Tris-buffered saline (TBS); for 1 hour in 3% normal goat serum (NGS), including 0.1% Triton in TBS; and for 48 hours in primary antibodies (the monoclonal antibody 192-IgG for p75NGFR and a commer- </p><p>TABLE 1. Effects of Trophic Factors on Numbers of Facial Motor Neurons </p><p>Average number of neurons in facial Treatment nucleus ir S.E.M. as percentage of aouu n contralateral. unlesioned side </p><p>BDNF 5 CNTF 5 Vehicle 3 </p><p>79.02 2 5.85* 30.38 ? 3.83 24.14 i 2.44 </p><p>*One-way ANOVA revealed a significant variance among the three groups [F(2,10) = 39.82, P &lt; 0.0011. A Newman-Keuls multiple range test found the B...</p></li></ul>


View more >