Brain Research, 559 (1991) 149-153 © 1991 Elsevier Science Publishers B.V. All fights reserved. 0006-8993/91/$03.50 ADONIS 000689939124841L
149
BRES 24841
Localization of nerve growth factor receptor mRNA in contused rat spinal cord by in situ hybridization Michael E. Reynolds, Nicoletta Brunello*, Italo Mocchetti and Jean R. Wrathall Department of Anatomy and Cell Biology, Georgetown University, Washington, DC 20007 (U.S.A.)
(Accepted 18 June 1991) Key words: Nerve growth factor receptor mRNA; Immunohistochemistry; Spinal cord injury; In situ hybridization; Factor VIII; ED1; Glial fibfillary acidic protein; Adult rat
Northern blot analysis using a probe for the low-affinity nerve growth factor receptor (NGFR) revealed that a mild contusive injury induces the expression of NGFR mRNA in rat spinal cord with a maximal expression at 7 days post-injury. We have now localized this induction using in situ hybridization and found the highest concentration of NGFR mRNA at the lesion epicenter. The location and pattern of autoradiographie grains were compared with that of various cell types at the injury site as determined by immunocytochemicalstudies. The results suggest that cells associated with blood vessels at the epicenter are induced to express NGFR mRNA at 7 days post-injury. Nerve growth factor (NGF) supports the survival and growth of peripheral sympathetic and sensory neurons during development and exhibits neurotrophic properties for specific populations of adult peripheral and central neurons after injury 17'22. It is believed that the neurotrophic activity of N G F is mediated by the receptorligand complex of N G F - N G F R , its internalization and subsequent transport to the cell body 12. In peripheral nerve, the induction of N G F R m R N A in Schwann cells of the distal segment after injury is thought to play an important role in subsequent regeneration 13. Immunohistochemical and in situ hybridization studies have also demonstrated that nerve growth factor receptor (NGFR) protein and m R N A , respectively, are present in several areas of the r a t CNS7'9'26v With respect to the spinal cord, N G F R and its m R N A are strongly expressed during development but to a lesser and much more limited extent in the adult 7"s'26. However, the expression of N G F R in the adult spinal cord can be up-regulated in response to injury. Sciatic nerve lesions lead not only to an increase in local N G F R in the nerve but also to the reexpression of the receptor by ipsilateral ventral motoneurons in the lumbar enlargement s. Further, we have shown that after a mid-thoracic contusive injury of the spinal cord, the levels of N G F R m R N A increase about 7-fold compared to laminectomy controls 2. Northern blot hybridization analysis using a probe against the low-affinity N G F R gene showed that this increase is (1) re-
stricted to the thoracic segments of the cord that contain the injury site and (2) maximal at 7 days after injury. We have now investigated the source of this m R N A by examining thoracic spinal cord tissue from rats at 7 days post-injury. In situ hybridization was used to localize N G F R m R N A and immunocytochemistry to identify cell types present in these locations. A standardized mild contusive injury was produced in female Sprague-Dawley rats (200-220 g) at the T8 vertebral level using a weight drop device 25. Each animal was maintained for 7 days, then decapitated and the spinal cord rapidly removed. A 15 mm segment of the thoracic spinal cord centered at the injury site was frozen at -20 °C and serial horizontal sections (14 a m ) cut and mounted on 3-aminopropyltriethoxysilane (Sigma) coated slides 16. Control sections were prepared from spinal cords of uninjured rats. N G F R m R N A was localized using a ass-labelled cRNA antisense probe encoding NGFR. This probe was generated by the standard transcription method 19 on the linearized plasmid pNico 2, a derivative of the p5B plasmid 3 (a gift from Dr. Moses V. Chao). The 600 base cRNA probe was shortened to approximately 150-200 bases by alkaline hydrolysis 5. Specificity of hybridization was monitored in sections of segments of sciatic nerve 7 days after transection, and contralateral normal nerve, embedded together in liver (Fig. 1A, inset). For in situ hybridization, all tissue sections were fixed with 4% paraformaldehyde in phosphate
* Permanent address: Center for Neuropharmacology, Institute of Pharmacological Sciences, University of Milan, Italy. Correspondence: J.R. Wrathall, Department of Anatomy and Cell Biology, School of Medicine, Georgetown University, 3900 Reservoir Rd. N.W., Washington, DC 20007, U.S.A. Fax: (1) (202) 687-1823.
150 buffer (0.1 M sodium phosphate, pH 7.4) for 5 min at ambient temperature and then hybridized using a standard protocol for riboprobes 24, with slight modifications. (1) Proteinase K pretreatment was avoided according to the recommendation of Gibbs et al.9. (2) After post-hybridization treatment with RNase A, sections were washed with 2x SSC for 60 min at 50 °C, rather than 37 °C. After hybridization, film autoradiographs were prepared using Kodak X-Omat AR film. These were subjected to image analysis using the Zeiss IBAS system with grain density converted to a gray level scale. Slides were also dipped for emulsion autoradiography using Kodak NTB2. Animals for immunohistochemistry were perfused with phosphate-buffered 4% paraformaldehyde and serial cryostat sections, similar to those used for in situ hybridization, were prepared. Slides were post fixed for 1 min in 10% neutral buffered formalin and dehydrated in graded alcohols then rehydrated. Sections were then washed in a high salt buffer (HSB: 29.2 g NaCI, 1.1 g monobasic NaPO 4 and 1.7 g dibasic NaPO 4 per liter), incubated in blocking serum for 30 min and then in primary antibody overnight at 4 °C. Antibodies to Factor VIII (1/50, Bio Genex Labs, Dublin, CA), ED1 (1/50, Chemicon International, Termecula, CA) and glial fibrillary acidic protein (GFAP) (1/100, Biomedical Technologies, Stoughton, MA) were used to detect these immunohistochemical markers for endothelial cells 11, macrophages 6 and astrocytes23, respectively. Antibody-antigen complex was detected with fluorescein conjugated secondary antibody, or alternatively, an immunoperoxidase technique using the Vecta ABC kit (Vector Labs, Burhngame, CA) with diaminobenzidine as chromagen.
Comparisons of sections from control and contused spinal cords after in situ hybridization (Fig. 1) showed the highest grain densities in sections of injured spinal cords at the epicenter (region of maximal damage). Grain densities were also higher than background levels in preserved gray matter rostral and caudal to the epicenter. Emulsion autoradiographs (Fig. 2A) showed a relatively diffuse pattern of grains in the gray matter adjacent to the injury sites. In contrast, at the epicenter, a linear or reticular pattern of grains was observed. As injury resuits in loss of neurons at the epicenter 2°, this pattern was compared to that of non-neuronal cells associated with the injury site. The pattern of distribution of astrocytes at one week after contusion was revealed with antibody to GFAP (Fig. 2B). Overall, GFAP staining was greater than in uninjured spinal cord. Reactive astrocytes with intense GFAP staining were seen at the margins of the epicenter and in adjacent gray and white matter. Specific GFAP staining was not demonstrated within the epicenter, although some non-specific staining was evident even in control sections from which primary antibody was omitted. In contrast, the epicenter was intensely stained with antibody to ED1 (Fig. 2C), a macrophage/monocyte marker 6. Large numbers of these cells were present closely opposed to one another at the epicenter. Fewer ED1 positive ceils were also seen in residual gray and white matter at the margins of the epicenter. In uninjured spinal cord, virtually no ED1 positive cells were found. Antibody to Factor VIII was used to identify endothelial cells u. Positive cells were observed as part of blood vessels in residual gray and white matter, and in sections of uninjured spinal cord. However, the most intense immunoreactivity was present
Fig. I. Distribution of NGFR mRNA in thoracic spinal cord as shown by in situ hybridization with asS-labelled riboprobe. Computer enhanced images of film autoradiographs prepared from longitudinal sections through the ventral spinal cord. A: in uninjured spinal cord, grains representing NGFR mRNA are indistinguishable from background. Inset: section through segments of sciatic nerve embedded in liver used to monitor specificity of hybridization. Fascicles from nerve transected 7 days earlier (upper right) show high levels of NGFR mRNA as compared to those from the contralateral control nerve (lower left), or surrounding liver. B: in contused spinal cord, 7 days after injury, NGFR mRNA is discernable at the epicenter (arrow) and in the gray matter of adjacent, preserved, spinal cord tissue rostral and caudal to the injury site. The highest density of NGFR mRNA is localized at the injury epicenter.
151
Fig. 2. Comparison of the distribution of NGFR mRNA at 7 days after spinal cord contusion with the distribution of various cell types as shown by immunocytochemistry. A-D: same magnification, bar = 1 mm. Arrows mark the junction between the injury epicenter, on the right, with adjacent preserved spinal cord tissue, on the left. A: emulsion autoradiograph showing the diffuse distribution of grains representing NGFR mRNA in gray matter adjacent to the injury site. In contrast, at the epicenter, the NGFR mRNA appears localized in a reticular pattern. B: distribution of astrocytes as revealed by antibody to GFAP. Intense staining is seen at the borders of the epicenter and in adjacent spinal cord tissue. Within the epicenter, the staining is non-specific. C: distribution of maerophage/monocytes as revealed by antibody to ED1. Large numbers of EDl-positive cells are present within the epicenter and at the margins of the lesion site. D: distribution of blood vessels as seen with antibody to the endothelial marker Factor VIII. There is a reticular pattern of staining within the epicenter and at the junction with adjacent spinal cord tissue. E, F: same magnification, bar = 0.2 mm. Comparison of reticular pattern of Factor VIIIpositive cells (E) within the epicenter with the distribution pattern for NGFR mRNA (F).
in a n e t w o r k of b l o o d vessels within the epicenter and especially at its margins (Fig. 2D). These vessels app e a r e d larger than those of uninjured spinal c o r d as well as exhibiting stronger F a c t o r V I I I activity. The reticular p a t t e r n o f these b l o o d vessels (Fig. 2E) was strikingly similar to the pattern of N G F R m R N A (Fig. 2F).
These results suggest that a significant p o r t i o n of the N G F R m R N A expression induced by contusive injury m a y be r e l a t e d to neovaseularization o f the injury site. Expression of N G F R m R N A could be in endothelial cells and/or in closely associated cells. T h e limits o f resolution of the current study preclude distinction b e t w e e n
152 these possibilities. N G F R protein has been previously detected in association with CNS blood/vessels of fetal and neonatal rats 26. Immunocytochemical staining using 192-IgG demonstrated N G F R
was localized in the
smooth muscle and adventitia of the vessels rather than the endothelium. A similar immunocytochemical localization of N G F R has also been reported for CNS vasculature of primates 21. These investigators suggest that the
the lesion epicenter. Activated macrophages appear to be important in inducing N G F m R N A in injured peripheral nerve 1° and can themselves release high amounts of N G F TM in tissue culture. We found large n u m b e r s of E D l - p o s i t i v e macrophages at the epicenter in areas where N G F R m R N A expression is highest. In contrast, few macrophages are present in the preserved spinal cord tissue adjacent to the injury site, where there are
N G F R may be present in sympathetic axons innervating
many blood vessels, but no evidence of enhanced N G F R
the smooth muscle of the blood vessels and/or the Schwann cells ensheathing these axons. High resolution immunocytochemical and ultrastructural studies will be
gene expression. Thus, our results are consistent with the hypothesis that macrophages may play a role in the increase in N G F and, indirectly, N G F R after injury. Our data also suggest that some processes, such as induction of N G F and the gene for its receptor, that support regeneration in the peripheral nervous system ~3,
needed to determine the precise cellular localization of N G F R protein produced in response to spinal cord injury. Further, the gene that appears to code for the highaffinity N G F receptor, and its product, the T R K protein, have recently been identified 1415. The effect of contusion upon this N G F receptor has yet to be determined. O u r results raise the question of potential signals for the induction of N G F R m R N A in the contused spinal cord. O n e possibility is an increased level of N G F itself 4,
also occur in the spinal cord after contusive injury. Whether they play a functional role in recovery from any of the effects of spinal cord injury remains to be determined.
as we have found a dramatic increase in an NGF-like protein at the injury site by one week after contusion ~. The cellular source of N G F in the contused spinal cord is currently u n k n o w n but could involve macrophages at
We thank Dr. Moses Chao for his gift of the p5B plasmid, Dr. Carol Colton for the gift of antibodies to Factor VIII and ED1, and Jim Bouzoukis for excellent technical assistance. This study was supported in part by National Institute of Health BRSG RR 05360 to I.M. and in part by NO1-NS7-2310 and Genetech. Inc., San Francisco (CA) to J.R.W.
l Bakhit, C., Armanini, M., Wong, W.L., Bennett, G.L. and Wrathall, J.R., Increase in nerve growth factor-like immunoreactivity and decrease in choline acetyltransferase following contusive spinal cord injury, Brain Research, in press. 2 Brunello, N., Reynolds, M., Wrathall, J.R. and Mocchetti, I., Increased nerve growth factor receptor mRNA in contused rat spinal cord, Neurosci. Lett., 118 (1990) 238-240. 3 Buck, C.R., Martinez, H.J., Chao, M.V. and Black, I.B., Differential expression of the nerve growth factor receptor gene in multiple brain areas, Dev. Brain Res., 44 (1988) 259-268. 4 Cavicehioli, L., Flanigan, T.P., Vantini, G., Fusco, M., Polato, P., Toffano, G., Walsh, ES. and Leon, A., NGF amplifies expression of NGF receptor messenger RNA in forebrain cholinergic neurons of rats, Eur. J. Neurosci., 1 (1989) 1605-1613. 5 Cox, K.H., DeLeon, D.V., Angerer, L.M. and Angerer, R.C., Detection of mRNA's in sea urchin embryos by in situ hybridization using asymmetric RNA probes, Dev. Biol., 101 (1984) 485-502. 6 Dijkstra, C.D., Dopp, E.A., Joling, P. and Kraal, G., The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies EDI, ED2 and ED3, Immunology, 54 (1985) 589-599. 7 Eckenstein, E, Transient expression of NGF-receptor-like immunoreactivity in postnatal rat brain and spinal cord, Brain Research, 446 (1988) 149-154. 8 Enfors, P., Henschon, A., Olson, L. and Person, H., Expression of nerve growth factor receptor mRNA is developmentally regulated and increased after axotomy in rat spinal cord motoneurons, Neuron, 2 (1989) 1605-1613. 9 Gibbs, R.B., mcCabe, J.T., Buck, C.R., Chao, M.V. and Pfaff, D.W., Expression of NGF receptor in the rat forebrain detected with in situ hybridization and immunohistochemistry, Mol. Brain Res., 6 (1989) 275-287. 10 Heumann, R., Lindholm, D., Bandtlow, C., Meyer, M.,
Radeke, M.J., Misko, T.P., Shooter, E. and Thoenen, H., Differential regulation of mRNA encoding nerve growth factor and its receptor in rat sciatic nerve during development, degeneration, and regeneration: role of macrophages, Proc. Natl. Acad. Sci. U.S.A., 84 (t987) 8735-8739. 11 Jaffe, E.A., Endothelial cells and the biology of factor VIII, N. Engl. J. Med., 296 (1977) 377-383. 12 Johnson Jr., E.M., Taniuchi, M., Clark, H.B., Springer, J.E., Koh, S., Tayrien, M.W. and Loy, R., Demonstration of the retrograde transport of nerve growth factor receptor in the peripheral and central nervous system, J. Neurosci., 7 (1987) 923929. 13 Johnson, E.M., Taniuchi, M. and Stefano, P., Expression and possible function of nerve growth factor receptor on Schwann cells, Trends Neurosci., 11 (1988) 299-304. 14 Kaplan, D.R., Hempstead, B.L., Martin-Zanca, D., Chao, M.V. and Parada, L.E, The trk proto-oncogene product: a signal transducing receptor for nerve growth factor, Science, 252 (1991) 554-558. 15 Klein, R., Jing, S., Nanduri, V., O'Rourke, E. and Barbacid, M., The trk proto-oncogene encodes a receptor for nerve growth factor, Cell, 65 (1991) 189-197. 16 Koo, E.H., Hoffman, P.N. and Price, D.L., Levels of neurotransmitter and cytoskeletal protein mRNA's during nerve regeneration in sympathetic ganglia, Brain Research, 449 (1988) 361-363. 17 Levi-Montalcini, R., The nerve growth factor 35 years later, Science, 236 (1987) 1154-1162. 18 Mallat, M., Houlgatte, R., Brachet, P. and Prochiantz, A., Lipopolysaccharide stimulated rat brain macrophages release NGF in vitro, Dev. Biol., 133 (1989) 309-311. 19 Melton, D.A., Kreig, P.A., Rebagfiati, M.R., Maniatis, T., Zinn, K. and Green, M.R., Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promotor, Nucleic Acid
153 Res., 12 (1984) 5497-5520. 20 Noble, L.J. and Wrathall, J.R., Correlative analyses of lesion development and functional status after graded spinal cord contusion injuries in the rat, Exp. Neurol., 103 (1989) 34-40. 21 Schatteman, G.C., Gibbs, L., Lanahan, A.A., Claude, P. and Bothwell, M., Expression of NGF receptor in the developing and adult primate central nervous system, J. Neurosci., 8 (1988) 860-873. 22 Shooter, E.M., The role of nerve growth factor in neuronal development, maintenance and regeneration. In Fidia Research Foundation Neuroscience Award Lectures, VoL 3, Raven, New York, 1989, pp. 21-51. 23 Uyeda, C.T., Eng, L.E and Bignami, A., Immunological study
of the glial fibrillary acidic protein, Brain Research, 37 (1972) 81-89. 24 Wilcox, J.N., Gee, C.E. and Roberts, J.L., In situ cDNA: mRNA hybridization: development of a technique to measure mRNA levels in individual cells, Methods Enzymol., 124 (1986) 510-533. 25 Wrathall, J.R., Pettegrew, R.K. and Harvey, E, Spinal cord contusion in the rat: production of graded, reproducible, injury groups, Exp. Neurol., 88 (1985) 108-122. 26 Yan, Q. and Johnson, E.M., An immunohistochemical study of nerve growth factor receptor in developing rats, J. Neurosci., 9 (1988) 3481-3498.
149
BRES 24841
Localization of nerve growth factor receptor mRNA in contused rat spinal cord by in situ hybridization Michael E. Reynolds, Nicoletta Brunello*, Italo Mocchetti and Jean R. Wrathall Department of Anatomy and Cell Biology, Georgetown University, Washington, DC 20007 (U.S.A.)
(Accepted 18 June 1991) Key words: Nerve growth factor receptor mRNA; Immunohistochemistry; Spinal cord injury; In situ hybridization; Factor VIII; ED1; Glial fibfillary acidic protein; Adult rat
Northern blot analysis using a probe for the low-affinity nerve growth factor receptor (NGFR) revealed that a mild contusive injury induces the expression of NGFR mRNA in rat spinal cord with a maximal expression at 7 days post-injury. We have now localized this induction using in situ hybridization and found the highest concentration of NGFR mRNA at the lesion epicenter. The location and pattern of autoradiographie grains were compared with that of various cell types at the injury site as determined by immunocytochemicalstudies. The results suggest that cells associated with blood vessels at the epicenter are induced to express NGFR mRNA at 7 days post-injury. Nerve growth factor (NGF) supports the survival and growth of peripheral sympathetic and sensory neurons during development and exhibits neurotrophic properties for specific populations of adult peripheral and central neurons after injury 17'22. It is believed that the neurotrophic activity of N G F is mediated by the receptorligand complex of N G F - N G F R , its internalization and subsequent transport to the cell body 12. In peripheral nerve, the induction of N G F R m R N A in Schwann cells of the distal segment after injury is thought to play an important role in subsequent regeneration 13. Immunohistochemical and in situ hybridization studies have also demonstrated that nerve growth factor receptor (NGFR) protein and m R N A , respectively, are present in several areas of the r a t CNS7'9'26v With respect to the spinal cord, N G F R and its m R N A are strongly expressed during development but to a lesser and much more limited extent in the adult 7"s'26. However, the expression of N G F R in the adult spinal cord can be up-regulated in response to injury. Sciatic nerve lesions lead not only to an increase in local N G F R in the nerve but also to the reexpression of the receptor by ipsilateral ventral motoneurons in the lumbar enlargement s. Further, we have shown that after a mid-thoracic contusive injury of the spinal cord, the levels of N G F R m R N A increase about 7-fold compared to laminectomy controls 2. Northern blot hybridization analysis using a probe against the low-affinity N G F R gene showed that this increase is (1) re-
stricted to the thoracic segments of the cord that contain the injury site and (2) maximal at 7 days after injury. We have now investigated the source of this m R N A by examining thoracic spinal cord tissue from rats at 7 days post-injury. In situ hybridization was used to localize N G F R m R N A and immunocytochemistry to identify cell types present in these locations. A standardized mild contusive injury was produced in female Sprague-Dawley rats (200-220 g) at the T8 vertebral level using a weight drop device 25. Each animal was maintained for 7 days, then decapitated and the spinal cord rapidly removed. A 15 mm segment of the thoracic spinal cord centered at the injury site was frozen at -20 °C and serial horizontal sections (14 a m ) cut and mounted on 3-aminopropyltriethoxysilane (Sigma) coated slides 16. Control sections were prepared from spinal cords of uninjured rats. N G F R m R N A was localized using a ass-labelled cRNA antisense probe encoding NGFR. This probe was generated by the standard transcription method 19 on the linearized plasmid pNico 2, a derivative of the p5B plasmid 3 (a gift from Dr. Moses V. Chao). The 600 base cRNA probe was shortened to approximately 150-200 bases by alkaline hydrolysis 5. Specificity of hybridization was monitored in sections of segments of sciatic nerve 7 days after transection, and contralateral normal nerve, embedded together in liver (Fig. 1A, inset). For in situ hybridization, all tissue sections were fixed with 4% paraformaldehyde in phosphate
* Permanent address: Center for Neuropharmacology, Institute of Pharmacological Sciences, University of Milan, Italy. Correspondence: J.R. Wrathall, Department of Anatomy and Cell Biology, School of Medicine, Georgetown University, 3900 Reservoir Rd. N.W., Washington, DC 20007, U.S.A. Fax: (1) (202) 687-1823.
150 buffer (0.1 M sodium phosphate, pH 7.4) for 5 min at ambient temperature and then hybridized using a standard protocol for riboprobes 24, with slight modifications. (1) Proteinase K pretreatment was avoided according to the recommendation of Gibbs et al.9. (2) After post-hybridization treatment with RNase A, sections were washed with 2x SSC for 60 min at 50 °C, rather than 37 °C. After hybridization, film autoradiographs were prepared using Kodak X-Omat AR film. These were subjected to image analysis using the Zeiss IBAS system with grain density converted to a gray level scale. Slides were also dipped for emulsion autoradiography using Kodak NTB2. Animals for immunohistochemistry were perfused with phosphate-buffered 4% paraformaldehyde and serial cryostat sections, similar to those used for in situ hybridization, were prepared. Slides were post fixed for 1 min in 10% neutral buffered formalin and dehydrated in graded alcohols then rehydrated. Sections were then washed in a high salt buffer (HSB: 29.2 g NaCI, 1.1 g monobasic NaPO 4 and 1.7 g dibasic NaPO 4 per liter), incubated in blocking serum for 30 min and then in primary antibody overnight at 4 °C. Antibodies to Factor VIII (1/50, Bio Genex Labs, Dublin, CA), ED1 (1/50, Chemicon International, Termecula, CA) and glial fibrillary acidic protein (GFAP) (1/100, Biomedical Technologies, Stoughton, MA) were used to detect these immunohistochemical markers for endothelial cells 11, macrophages 6 and astrocytes23, respectively. Antibody-antigen complex was detected with fluorescein conjugated secondary antibody, or alternatively, an immunoperoxidase technique using the Vecta ABC kit (Vector Labs, Burhngame, CA) with diaminobenzidine as chromagen.
Comparisons of sections from control and contused spinal cords after in situ hybridization (Fig. 1) showed the highest grain densities in sections of injured spinal cords at the epicenter (region of maximal damage). Grain densities were also higher than background levels in preserved gray matter rostral and caudal to the epicenter. Emulsion autoradiographs (Fig. 2A) showed a relatively diffuse pattern of grains in the gray matter adjacent to the injury sites. In contrast, at the epicenter, a linear or reticular pattern of grains was observed. As injury resuits in loss of neurons at the epicenter 2°, this pattern was compared to that of non-neuronal cells associated with the injury site. The pattern of distribution of astrocytes at one week after contusion was revealed with antibody to GFAP (Fig. 2B). Overall, GFAP staining was greater than in uninjured spinal cord. Reactive astrocytes with intense GFAP staining were seen at the margins of the epicenter and in adjacent gray and white matter. Specific GFAP staining was not demonstrated within the epicenter, although some non-specific staining was evident even in control sections from which primary antibody was omitted. In contrast, the epicenter was intensely stained with antibody to ED1 (Fig. 2C), a macrophage/monocyte marker 6. Large numbers of these cells were present closely opposed to one another at the epicenter. Fewer ED1 positive ceils were also seen in residual gray and white matter at the margins of the epicenter. In uninjured spinal cord, virtually no ED1 positive cells were found. Antibody to Factor VIII was used to identify endothelial cells u. Positive cells were observed as part of blood vessels in residual gray and white matter, and in sections of uninjured spinal cord. However, the most intense immunoreactivity was present
Fig. I. Distribution of NGFR mRNA in thoracic spinal cord as shown by in situ hybridization with asS-labelled riboprobe. Computer enhanced images of film autoradiographs prepared from longitudinal sections through the ventral spinal cord. A: in uninjured spinal cord, grains representing NGFR mRNA are indistinguishable from background. Inset: section through segments of sciatic nerve embedded in liver used to monitor specificity of hybridization. Fascicles from nerve transected 7 days earlier (upper right) show high levels of NGFR mRNA as compared to those from the contralateral control nerve (lower left), or surrounding liver. B: in contused spinal cord, 7 days after injury, NGFR mRNA is discernable at the epicenter (arrow) and in the gray matter of adjacent, preserved, spinal cord tissue rostral and caudal to the injury site. The highest density of NGFR mRNA is localized at the injury epicenter.
151
Fig. 2. Comparison of the distribution of NGFR mRNA at 7 days after spinal cord contusion with the distribution of various cell types as shown by immunocytochemistry. A-D: same magnification, bar = 1 mm. Arrows mark the junction between the injury epicenter, on the right, with adjacent preserved spinal cord tissue, on the left. A: emulsion autoradiograph showing the diffuse distribution of grains representing NGFR mRNA in gray matter adjacent to the injury site. In contrast, at the epicenter, the NGFR mRNA appears localized in a reticular pattern. B: distribution of astrocytes as revealed by antibody to GFAP. Intense staining is seen at the borders of the epicenter and in adjacent spinal cord tissue. Within the epicenter, the staining is non-specific. C: distribution of maerophage/monocytes as revealed by antibody to ED1. Large numbers of EDl-positive cells are present within the epicenter and at the margins of the lesion site. D: distribution of blood vessels as seen with antibody to the endothelial marker Factor VIII. There is a reticular pattern of staining within the epicenter and at the junction with adjacent spinal cord tissue. E, F: same magnification, bar = 0.2 mm. Comparison of reticular pattern of Factor VIIIpositive cells (E) within the epicenter with the distribution pattern for NGFR mRNA (F).
in a n e t w o r k of b l o o d vessels within the epicenter and especially at its margins (Fig. 2D). These vessels app e a r e d larger than those of uninjured spinal c o r d as well as exhibiting stronger F a c t o r V I I I activity. The reticular p a t t e r n o f these b l o o d vessels (Fig. 2E) was strikingly similar to the pattern of N G F R m R N A (Fig. 2F).
These results suggest that a significant p o r t i o n of the N G F R m R N A expression induced by contusive injury m a y be r e l a t e d to neovaseularization o f the injury site. Expression of N G F R m R N A could be in endothelial cells and/or in closely associated cells. T h e limits o f resolution of the current study preclude distinction b e t w e e n
152 these possibilities. N G F R protein has been previously detected in association with CNS blood/vessels of fetal and neonatal rats 26. Immunocytochemical staining using 192-IgG demonstrated N G F R
was localized in the
smooth muscle and adventitia of the vessels rather than the endothelium. A similar immunocytochemical localization of N G F R has also been reported for CNS vasculature of primates 21. These investigators suggest that the
the lesion epicenter. Activated macrophages appear to be important in inducing N G F m R N A in injured peripheral nerve 1° and can themselves release high amounts of N G F TM in tissue culture. We found large n u m b e r s of E D l - p o s i t i v e macrophages at the epicenter in areas where N G F R m R N A expression is highest. In contrast, few macrophages are present in the preserved spinal cord tissue adjacent to the injury site, where there are
N G F R may be present in sympathetic axons innervating
many blood vessels, but no evidence of enhanced N G F R
the smooth muscle of the blood vessels and/or the Schwann cells ensheathing these axons. High resolution immunocytochemical and ultrastructural studies will be
gene expression. Thus, our results are consistent with the hypothesis that macrophages may play a role in the increase in N G F and, indirectly, N G F R after injury. Our data also suggest that some processes, such as induction of N G F and the gene for its receptor, that support regeneration in the peripheral nervous system ~3,
needed to determine the precise cellular localization of N G F R protein produced in response to spinal cord injury. Further, the gene that appears to code for the highaffinity N G F receptor, and its product, the T R K protein, have recently been identified 1415. The effect of contusion upon this N G F receptor has yet to be determined. O u r results raise the question of potential signals for the induction of N G F R m R N A in the contused spinal cord. O n e possibility is an increased level of N G F itself 4,
also occur in the spinal cord after contusive injury. Whether they play a functional role in recovery from any of the effects of spinal cord injury remains to be determined.
as we have found a dramatic increase in an NGF-like protein at the injury site by one week after contusion ~. The cellular source of N G F in the contused spinal cord is currently u n k n o w n but could involve macrophages at
We thank Dr. Moses Chao for his gift of the p5B plasmid, Dr. Carol Colton for the gift of antibodies to Factor VIII and ED1, and Jim Bouzoukis for excellent technical assistance. This study was supported in part by National Institute of Health BRSG RR 05360 to I.M. and in part by NO1-NS7-2310 and Genetech. Inc., San Francisco (CA) to J.R.W.
l Bakhit, C., Armanini, M., Wong, W.L., Bennett, G.L. and Wrathall, J.R., Increase in nerve growth factor-like immunoreactivity and decrease in choline acetyltransferase following contusive spinal cord injury, Brain Research, in press. 2 Brunello, N., Reynolds, M., Wrathall, J.R. and Mocchetti, I., Increased nerve growth factor receptor mRNA in contused rat spinal cord, Neurosci. Lett., 118 (1990) 238-240. 3 Buck, C.R., Martinez, H.J., Chao, M.V. and Black, I.B., Differential expression of the nerve growth factor receptor gene in multiple brain areas, Dev. Brain Res., 44 (1988) 259-268. 4 Cavicehioli, L., Flanigan, T.P., Vantini, G., Fusco, M., Polato, P., Toffano, G., Walsh, ES. and Leon, A., NGF amplifies expression of NGF receptor messenger RNA in forebrain cholinergic neurons of rats, Eur. J. Neurosci., 1 (1989) 1605-1613. 5 Cox, K.H., DeLeon, D.V., Angerer, L.M. and Angerer, R.C., Detection of mRNA's in sea urchin embryos by in situ hybridization using asymmetric RNA probes, Dev. Biol., 101 (1984) 485-502. 6 Dijkstra, C.D., Dopp, E.A., Joling, P. and Kraal, G., The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies EDI, ED2 and ED3, Immunology, 54 (1985) 589-599. 7 Eckenstein, E, Transient expression of NGF-receptor-like immunoreactivity in postnatal rat brain and spinal cord, Brain Research, 446 (1988) 149-154. 8 Enfors, P., Henschon, A., Olson, L. and Person, H., Expression of nerve growth factor receptor mRNA is developmentally regulated and increased after axotomy in rat spinal cord motoneurons, Neuron, 2 (1989) 1605-1613. 9 Gibbs, R.B., mcCabe, J.T., Buck, C.R., Chao, M.V. and Pfaff, D.W., Expression of NGF receptor in the rat forebrain detected with in situ hybridization and immunohistochemistry, Mol. Brain Res., 6 (1989) 275-287. 10 Heumann, R., Lindholm, D., Bandtlow, C., Meyer, M.,
Radeke, M.J., Misko, T.P., Shooter, E. and Thoenen, H., Differential regulation of mRNA encoding nerve growth factor and its receptor in rat sciatic nerve during development, degeneration, and regeneration: role of macrophages, Proc. Natl. Acad. Sci. U.S.A., 84 (t987) 8735-8739. 11 Jaffe, E.A., Endothelial cells and the biology of factor VIII, N. Engl. J. Med., 296 (1977) 377-383. 12 Johnson Jr., E.M., Taniuchi, M., Clark, H.B., Springer, J.E., Koh, S., Tayrien, M.W. and Loy, R., Demonstration of the retrograde transport of nerve growth factor receptor in the peripheral and central nervous system, J. Neurosci., 7 (1987) 923929. 13 Johnson, E.M., Taniuchi, M. and Stefano, P., Expression and possible function of nerve growth factor receptor on Schwann cells, Trends Neurosci., 11 (1988) 299-304. 14 Kaplan, D.R., Hempstead, B.L., Martin-Zanca, D., Chao, M.V. and Parada, L.E, The trk proto-oncogene product: a signal transducing receptor for nerve growth factor, Science, 252 (1991) 554-558. 15 Klein, R., Jing, S., Nanduri, V., O'Rourke, E. and Barbacid, M., The trk proto-oncogene encodes a receptor for nerve growth factor, Cell, 65 (1991) 189-197. 16 Koo, E.H., Hoffman, P.N. and Price, D.L., Levels of neurotransmitter and cytoskeletal protein mRNA's during nerve regeneration in sympathetic ganglia, Brain Research, 449 (1988) 361-363. 17 Levi-Montalcini, R., The nerve growth factor 35 years later, Science, 236 (1987) 1154-1162. 18 Mallat, M., Houlgatte, R., Brachet, P. and Prochiantz, A., Lipopolysaccharide stimulated rat brain macrophages release NGF in vitro, Dev. Biol., 133 (1989) 309-311. 19 Melton, D.A., Kreig, P.A., Rebagfiati, M.R., Maniatis, T., Zinn, K. and Green, M.R., Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promotor, Nucleic Acid
153 Res., 12 (1984) 5497-5520. 20 Noble, L.J. and Wrathall, J.R., Correlative analyses of lesion development and functional status after graded spinal cord contusion injuries in the rat, Exp. Neurol., 103 (1989) 34-40. 21 Schatteman, G.C., Gibbs, L., Lanahan, A.A., Claude, P. and Bothwell, M., Expression of NGF receptor in the developing and adult primate central nervous system, J. Neurosci., 8 (1988) 860-873. 22 Shooter, E.M., The role of nerve growth factor in neuronal development, maintenance and regeneration. In Fidia Research Foundation Neuroscience Award Lectures, VoL 3, Raven, New York, 1989, pp. 21-51. 23 Uyeda, C.T., Eng, L.E and Bignami, A., Immunological study
of the glial fibrillary acidic protein, Brain Research, 37 (1972) 81-89. 24 Wilcox, J.N., Gee, C.E. and Roberts, J.L., In situ cDNA: mRNA hybridization: development of a technique to measure mRNA levels in individual cells, Methods Enzymol., 124 (1986) 510-533. 25 Wrathall, J.R., Pettegrew, R.K. and Harvey, E, Spinal cord contusion in the rat: production of graded, reproducible, injury groups, Exp. Neurol., 88 (1985) 108-122. 26 Yan, Q. and Johnson, E.M., An immunohistochemical study of nerve growth factor receptor in developing rats, J. Neurosci., 9 (1988) 3481-3498.
