- Email: [email protected]
Neurones in autonomic ganglia of normal horses contain phosphorylated neurofilaments
J. Comp. Path. 1993 Vol. 108, 109-112
Neurones in Autonomic Ganglia of N o r m a l H o r s e s Contain Phosphorylated Neurofilaments I. R. GrittSths, S. A. Lusk, E. Kyriakides and S. Smith Applied Neurobiology Group and Department of VeterinarySurgeu, Universityof Glasgow, Bearsden, Glasgow G6I 1QH, U.K.
Summary Neurofilaments (NE) are composed of three polypeptides of differing molecu~ lar size, termed NF-L, NF-M and NF-H. The NF-H and, to a lesser degree, NF-M components are phosphorylated. In the majority of normal neurones, the location of phosphorylated NF is confined to neuronal processes, particularly the axon, and excluded from the perikaryon. CelI bodies ofautonomlc neurones of the rat do not contain phosphorylated NF. In many disease states, phosphorylated NF accumulate in the neuronal cell body and therefore in most circumstances their presence indicates abnormality. This paper reports that in at least two autonomic ganglia of normal horses (stellate and coeliomesenteric) the vast majority of neuronal perikarya immunostain strongly for phosphorylated NF. Pretreatment with alkaline phosphatase abolishes staining.
Introduction Neurofilaments (NF) are a ubiquitous constituent ofneurones forming a major component of their cytoskeleton. Three polypeptides, coded by separate genes, constitute the rod-like filament and associated side arms. The polypeptides of relative molecular weight ~ 68, 155-160 and 200 kDa are commonly referred to as NF-L, N F - M and NF-H, respectively (Fliegner and Liem, 1991). The proteins are synthesized in the perikaryon and transported by slow axonal transport to locations such as the axon. It is still controversial whether only assembled filaments move into the axon or whether assembly can occur locally. It is clear, however, that certain post-translational modifications do relate to the intracellular location. N F - H and, to a lesser degree, N F - M are heavily phosphorylated in the axon, whereas phosphorylated NF are absent or markedly reduced in the majority of neuronal cell bodies (Nixon and Sihag, 1991). Several monoclonal antibodies recognize only phosphorylated epitopes on N F - H and N F - M , so allowing their distribution to be studied (Sternberger and Sternberger, 1983). All or the vast majority of neuronal cell bodies in the normal CNS fail to stain for phosphorylated epitopes. A subpopulation of neurones in dorsal root ganglia of the rat (mainly the large light cells) are Corresponding Author: I. R. Griffiths, Dept, of Veterinary Surgery, University of Glasgow, Bearsden, Glasgow G61 IQH, U.K. 0021-9975/93/010109+04 $08.00/0
(C) 1993 Academic Press Limited
1 10
I.R. Griffiths e t al.
reactive, the remainder being negative (Moss and Lewkowicz, 1983; Rosenfeld, Dorman, Griffin, Sternberger, Sternberger and Price, 1987). Autonomic neurones in the rat superior cervical ganglion are also reported to be negative for phosphorylated epitopes (Shaw, Winialski and Reier, 1988). However, in m a n y disease states (Schlaepfer, 1987) or during the axonal reaction following a x o t o m y (Moss and Lewkowicz, 1983), neuronal perikarya in the CNS and peripheral nervous system, including the autonomic neurones, stain intensely with these antibodies denoting accumulation of phosphorylated NF; therefore their presence in this location indicates an abnormality in most circumstances. As part of our investigations into equine grass sickness (EGS), we commenced an immunocytochemical study of various structural and cytoplasmic proteins, including NF proteins, in equine autonomic neurones. Since, to our knowledge, these have not been described previously, we examined stellate and coelio-mesenteric (C/M) ganglia of neurologically-normal horses destroyed because of various orthopaedic conditions or colic. These ganglia are commonly used for diagnostic purposes in cases of suspected EGS. Since the majority of neurones in the autonomic ganglia of these " n o r m a l " horses stained for phosphorylated NF, we investigated this aspect in more detail. Materials and Methods
Stellate and C/M ganglia were removed within 30 min of death from five neurologically-normal horses and immersion-fixed in buffered neutral formalin. Blocks were either post-fixed in Bouin's fixative and processed for paraffin wax sections or infiltrated with 30 per cent sucrose, frozen and cryosections prepared. Samples of jejunum were prepared in a similar fashion. Paraffin wax sections were immunostained by the peroxidase-antiperoxidase (PAP) technique and cryosecti~ns by indirect immunofluorescence. Two antibodies recognizing phosphorylated epitopes, principally on NF-H and to some degree on NF-M were used; RT 97 (Wood and Anderton, 1981) (donated by Dr B. Anderton) and SMI 31 (Sternberger monoclonals incorporated-Affiniti Research Products Ltd, Ilkeston, Derbyshire, U.K.). They were used at dilutions ofl in 2500 to 1 in 5000 and 1 in 1500, respectively. Sections were also pretreated with E. coli type II1 alkaline phosphatase (Sigma, Poole, Dorset, U.K.) at 150, 450 and 600 I.tg per m! in Tris-HC1/0"005M phenylmethylsulphonylfluoride, pH 8"0 for 2'5, 6 or 12 h at 32°C (Sternberger and Sternberger, 1983; Bignami, Chi and Dahl, 1986).
Results
The vast majority of, b u t not all, neuronal perikarya of both ganglia immunostained strongly with both antibodies, with S M I 31 giving a greater intensity than R T 97 (Fig. 1A to D). Nuclei were unstained. Axonal profiles in the ganglia also reacted intensely. Pretreatment with alkaline phosphatase at 150 or 4.50 l.tg per ml fbr 2"5 h abolished axonal staining and reduced neuronal staining. The neuronal immunostaining was completely abolished by 450 gg per ml tbr 6 h or higher treatments (Fig. 1E and F). Immunopositive neurones and axons were also present in the gut plexi of the jejunum when stained with S M I 31. Neurones in sections of normal equine spinal cord tMled to stain lbr phosphorylatcd epitopes.
Phosphorylated NF of Horse Ganglia
Fig. I.
111
Phase (A, C, E/ and immunofluorescent (B, D, F) images of equine codlo-mescnteric ganglia to show presence ol~ phosphoryl,ltcd neurolilaments in the neuronal perikarya. B and D are immunostained with R_T 97 and .qMI 31, respectively. In addition m the reaction ill the neurona[ cell bodies, axona[ processes are also posltb, c. I? shmvs that pretre,ttmcnt with alkaline phosphatase (450 I-tg per ml; 6 h.) abolished the staining with 8bl[ 31. x 120.
Discussion These results indicate that tile majority of neurones in at least two autononaic ganglia of" normal horses and neurones within the intramural plexi of the gut contain phosphorylated neurofilament proteins; a feature which, in most neurones, is of'ten associated with abnormality. The horse appears to be different from the rat in which similar autonomic neurones are not immunostained. This tkaturc presumably implies a difference in post-translational processing, assembly or transport o[ NF in equine atttonomic ganglia. It seems highly unlikely that the presence ofphosphorylated NF is related itt any way to
112
I.R. Griffiths et at.
the pathogenesis of EGS, but is of relevance when assessing disease states by immunocytochemical profiles.
Acknowledgments We are grateful to Dr B. Anderton for RT 97 and to Mr A. May for photography. This study was supported by the Horseraee Betting Levy Board. References Bignami, A., Chi, N. H. and Dahi, D. (1986). Neurofilament phosphorylation in peripheral nerve regeneration. Brain Research, 375, 73-82. Fliegner, K. H. and Liem, R. K. H. (1991). Cellular and molecular biology of neuronal intermediate filaments. InternationalReview of Cytology, 131, 109-167. Moss, T. H. and Lewkowicz, S.J. (1983). The axonal reaction in motor and sensory neurones of mice studied by a monoclonal antibody marker of neurofilament protein. Journal of the Neurological Sciences, 60, 267-280. Nixon, R. A. and Sihag, R. K. (1991). Neurofilament phosphorylation: A new look at regulation and function. Trends in Neurosciences, 14, 501-506. Rosenfeld, J., Dorman, M. E., Griffin, J. W., Sternberger, L. A., Sternberger, N. H. and Price, D. L. (1987). Distribution of neurofilament antigens after axonal / iniury. Journal of Neuropathology and Experimental Neurology, 46, 269-282. Schlaepfer, W. W. (1987). Neurofilaments: structure, metabolism and implication in disease. Journal of Neuropathology and Experimental Neurology, 46, 117-129. Shaw, G., Winialski, D. and Reier, P. (1988). The effect of axotomy and deafferentation on phosphorylation dependent antigenicity of neurofilaments in rat superior cervical ganglion neurons. Brain Research, 460, 227-234. -Sternberger, L. A. and Sternberger, N. H. (1983). Monoclonal antibodies distinguish phosphorglated and nonphosphorylated forms of neurofilaments in situ. Proceedings of the Natzonal Academy of Sciences of the United States of America, 89, 6126-6130. Wood, J. N. and Anderton, B. H. (1981). Monoclonal antibodies to mammalian neurofilaments. BioscienceReports, 1, 263-268.
I Reeeived, August 21st, 1992 ] Accepted, September lOth, 1992]
Neurones in Autonomic Ganglia of N o r m a l H o r s e s Contain Phosphorylated Neurofilaments I. R. GrittSths, S. A. Lusk, E. Kyriakides and S. Smith Applied Neurobiology Group and Department of VeterinarySurgeu, Universityof Glasgow, Bearsden, Glasgow G6I 1QH, U.K.
Summary Neurofilaments (NE) are composed of three polypeptides of differing molecu~ lar size, termed NF-L, NF-M and NF-H. The NF-H and, to a lesser degree, NF-M components are phosphorylated. In the majority of normal neurones, the location of phosphorylated NF is confined to neuronal processes, particularly the axon, and excluded from the perikaryon. CelI bodies ofautonomlc neurones of the rat do not contain phosphorylated NF. In many disease states, phosphorylated NF accumulate in the neuronal cell body and therefore in most circumstances their presence indicates abnormality. This paper reports that in at least two autonomic ganglia of normal horses (stellate and coeliomesenteric) the vast majority of neuronal perikarya immunostain strongly for phosphorylated NF. Pretreatment with alkaline phosphatase abolishes staining.
Introduction Neurofilaments (NF) are a ubiquitous constituent ofneurones forming a major component of their cytoskeleton. Three polypeptides, coded by separate genes, constitute the rod-like filament and associated side arms. The polypeptides of relative molecular weight ~ 68, 155-160 and 200 kDa are commonly referred to as NF-L, N F - M and NF-H, respectively (Fliegner and Liem, 1991). The proteins are synthesized in the perikaryon and transported by slow axonal transport to locations such as the axon. It is still controversial whether only assembled filaments move into the axon or whether assembly can occur locally. It is clear, however, that certain post-translational modifications do relate to the intracellular location. N F - H and, to a lesser degree, N F - M are heavily phosphorylated in the axon, whereas phosphorylated NF are absent or markedly reduced in the majority of neuronal cell bodies (Nixon and Sihag, 1991). Several monoclonal antibodies recognize only phosphorylated epitopes on N F - H and N F - M , so allowing their distribution to be studied (Sternberger and Sternberger, 1983). All or the vast majority of neuronal cell bodies in the normal CNS fail to stain for phosphorylated epitopes. A subpopulation of neurones in dorsal root ganglia of the rat (mainly the large light cells) are Corresponding Author: I. R. Griffiths, Dept, of Veterinary Surgery, University of Glasgow, Bearsden, Glasgow G61 IQH, U.K. 0021-9975/93/010109+04 $08.00/0
(C) 1993 Academic Press Limited
1 10
I.R. Griffiths e t al.
reactive, the remainder being negative (Moss and Lewkowicz, 1983; Rosenfeld, Dorman, Griffin, Sternberger, Sternberger and Price, 1987). Autonomic neurones in the rat superior cervical ganglion are also reported to be negative for phosphorylated epitopes (Shaw, Winialski and Reier, 1988). However, in m a n y disease states (Schlaepfer, 1987) or during the axonal reaction following a x o t o m y (Moss and Lewkowicz, 1983), neuronal perikarya in the CNS and peripheral nervous system, including the autonomic neurones, stain intensely with these antibodies denoting accumulation of phosphorylated NF; therefore their presence in this location indicates an abnormality in most circumstances. As part of our investigations into equine grass sickness (EGS), we commenced an immunocytochemical study of various structural and cytoplasmic proteins, including NF proteins, in equine autonomic neurones. Since, to our knowledge, these have not been described previously, we examined stellate and coelio-mesenteric (C/M) ganglia of neurologically-normal horses destroyed because of various orthopaedic conditions or colic. These ganglia are commonly used for diagnostic purposes in cases of suspected EGS. Since the majority of neurones in the autonomic ganglia of these " n o r m a l " horses stained for phosphorylated NF, we investigated this aspect in more detail. Materials and Methods
Stellate and C/M ganglia were removed within 30 min of death from five neurologically-normal horses and immersion-fixed in buffered neutral formalin. Blocks were either post-fixed in Bouin's fixative and processed for paraffin wax sections or infiltrated with 30 per cent sucrose, frozen and cryosections prepared. Samples of jejunum were prepared in a similar fashion. Paraffin wax sections were immunostained by the peroxidase-antiperoxidase (PAP) technique and cryosecti~ns by indirect immunofluorescence. Two antibodies recognizing phosphorylated epitopes, principally on NF-H and to some degree on NF-M were used; RT 97 (Wood and Anderton, 1981) (donated by Dr B. Anderton) and SMI 31 (Sternberger monoclonals incorporated-Affiniti Research Products Ltd, Ilkeston, Derbyshire, U.K.). They were used at dilutions ofl in 2500 to 1 in 5000 and 1 in 1500, respectively. Sections were also pretreated with E. coli type II1 alkaline phosphatase (Sigma, Poole, Dorset, U.K.) at 150, 450 and 600 I.tg per m! in Tris-HC1/0"005M phenylmethylsulphonylfluoride, pH 8"0 for 2'5, 6 or 12 h at 32°C (Sternberger and Sternberger, 1983; Bignami, Chi and Dahl, 1986).
Results
The vast majority of, b u t not all, neuronal perikarya of both ganglia immunostained strongly with both antibodies, with S M I 31 giving a greater intensity than R T 97 (Fig. 1A to D). Nuclei were unstained. Axonal profiles in the ganglia also reacted intensely. Pretreatment with alkaline phosphatase at 150 or 4.50 l.tg per ml fbr 2"5 h abolished axonal staining and reduced neuronal staining. The neuronal immunostaining was completely abolished by 450 gg per ml tbr 6 h or higher treatments (Fig. 1E and F). Immunopositive neurones and axons were also present in the gut plexi of the jejunum when stained with S M I 31. Neurones in sections of normal equine spinal cord tMled to stain lbr phosphorylatcd epitopes.
Phosphorylated NF of Horse Ganglia
Fig. I.
111
Phase (A, C, E/ and immunofluorescent (B, D, F) images of equine codlo-mescnteric ganglia to show presence ol~ phosphoryl,ltcd neurolilaments in the neuronal perikarya. B and D are immunostained with R_T 97 and .qMI 31, respectively. In addition m the reaction ill the neurona[ cell bodies, axona[ processes are also posltb, c. I? shmvs that pretre,ttmcnt with alkaline phosphatase (450 I-tg per ml; 6 h.) abolished the staining with 8bl[ 31. x 120.
Discussion These results indicate that tile majority of neurones in at least two autononaic ganglia of" normal horses and neurones within the intramural plexi of the gut contain phosphorylated neurofilament proteins; a feature which, in most neurones, is of'ten associated with abnormality. The horse appears to be different from the rat in which similar autonomic neurones are not immunostained. This tkaturc presumably implies a difference in post-translational processing, assembly or transport o[ NF in equine atttonomic ganglia. It seems highly unlikely that the presence ofphosphorylated NF is related itt any way to
112
I.R. Griffiths et at.
the pathogenesis of EGS, but is of relevance when assessing disease states by immunocytochemical profiles.
Acknowledgments We are grateful to Dr B. Anderton for RT 97 and to Mr A. May for photography. This study was supported by the Horseraee Betting Levy Board. References Bignami, A., Chi, N. H. and Dahi, D. (1986). Neurofilament phosphorylation in peripheral nerve regeneration. Brain Research, 375, 73-82. Fliegner, K. H. and Liem, R. K. H. (1991). Cellular and molecular biology of neuronal intermediate filaments. InternationalReview of Cytology, 131, 109-167. Moss, T. H. and Lewkowicz, S.J. (1983). The axonal reaction in motor and sensory neurones of mice studied by a monoclonal antibody marker of neurofilament protein. Journal of the Neurological Sciences, 60, 267-280. Nixon, R. A. and Sihag, R. K. (1991). Neurofilament phosphorylation: A new look at regulation and function. Trends in Neurosciences, 14, 501-506. Rosenfeld, J., Dorman, M. E., Griffin, J. W., Sternberger, L. A., Sternberger, N. H. and Price, D. L. (1987). Distribution of neurofilament antigens after axonal / iniury. Journal of Neuropathology and Experimental Neurology, 46, 269-282. Schlaepfer, W. W. (1987). Neurofilaments: structure, metabolism and implication in disease. Journal of Neuropathology and Experimental Neurology, 46, 117-129. Shaw, G., Winialski, D. and Reier, P. (1988). The effect of axotomy and deafferentation on phosphorylation dependent antigenicity of neurofilaments in rat superior cervical ganglion neurons. Brain Research, 460, 227-234. -Sternberger, L. A. and Sternberger, N. H. (1983). Monoclonal antibodies distinguish phosphorglated and nonphosphorylated forms of neurofilaments in situ. Proceedings of the Natzonal Academy of Sciences of the United States of America, 89, 6126-6130. Wood, J. N. and Anderton, B. H. (1981). Monoclonal antibodies to mammalian neurofilaments. BioscienceReports, 1, 263-268.
I Reeeived, August 21st, 1992 ] Accepted, September lOth, 1992]