- Email: [email protected]
First appearance of type II neurons during ontogenesis in the spiral ganglion of the rat. An immunocytochemical study
Developmental Brain Research, 48 (1989) 143-149 Elsevier
143
BRD 50935
First appearance of Type II neurons during ontogenesis in the spiral ganglion of the rat. An immunocytochemical study A. Hafidi and R. Romand Laboratoire de Neurobiologie, Universitd Blaise Pascal, Ensemble scientifique des Cdzeauz, Aubidre (France) (Accepted 7 February 1989) Key words: Neurofilament subunit; Spiral ganglion; Ontogenesis; Type II neuron; Immunocytochemistry; Rat
Ontogenesis of spiral ganglion in the rat was studied using antibodies to three subunits of neurofilaments (NFs): NF 68 KDa, NF 160 KDa and NF 200 KDa. The expression of immunoreactivity was examined with 3 immunocytochemical methods: indirect immunofluorescence, peroxidase-antiperoxidase and avidin-biotin complex. Aim of the study was to detect the time of differentiation of the spiral ganglion type II neurons. At 16 and 18 days of gestation, most neuron cell bodies express immunoreactivity to only two NF subunits: NF 68 and NF 160, but at birth they react with the antibodies to all 3 subunits albeit weakly. Nevertheless, a small population (about 7%) of nerve cells that strongly reacts against all 3 NF subunits emerges in the basal turn, already at 20 days of gestation. Two to 3 days after birth, the strongly stained cells are dispersed throughout the entire ganglion. The intensity of their reaction to the NF antibodies is similar to that seen in the adult animal. The strong immunoreactivity of this selective neuronal population suggest, that they correspond to the type II spiral ganglion neurons. Our results imply that the differentiation between the type I and the type II of spiral neurons in the rat occurs perinatally.
INTRODUCTION Spiral ganglion cells are afferent neurons o f the organ of Corti; their peripheral and central processes connect auditory receptors with the cochlear nucleus. Spiral ganglion contains at least two clearly distinct populations of neurons. These two types of cells are well characterized by their structural, ultrastructura124'27'3a and by biochemical features 3 and by their connections with receptors 14. Type I neurons (T I) represent 90-95% of ganglion cells; they have a large perikaryon rich in rough endoplasmic reticulum and ribosomes. Type II neurons (T II) represent 5 - 1 0 % of ganglion cells; they have a small perikaryon and show a characteristic accumulation of neurofilaments (NFs) 3,24,26,33. In the rat, T II neurons can be differentiated ultrastructurally only around the end of the first postnatal week 25'29. Since the accumulation of NFs is the main ultrastructural criterion distinguishing the
two types of neurons, their emergence is not very easy to detect in the electron microscope. In the present study, therefore we have used the antibodies to the 3 subunits of NFs. NFs are the neuron-specific intermediate filaments15,37; in mammals they are composed of 3 subunits with apparent molecular weights of 200 kDa (NF 200), 160 k D a (NF 160) and 68 k D a (NF 68) 11'18'28. The 3 subunits are immunochemically distinct 16,31,34. It has been shown previously that adult T II neurons strongly react with the 3 NF subunits s'26. The present study was undertaken to evaluate the presence of NF subunits in the developing ganglion cells and to characterize the first appearance of T II neurons. MATERIALS AND METHODS We used S p r a g u e - D a w l e y rat fetuses of 16, 18 and 20 days of gestation, newborn and 2-6, 8 and 10
Correspondence: R. Romand, Laboratoire de Neurobiologie, Universit6 Blaise Pascal, Ensemble scientifique des C6zeaux, 63177 Aubi6re C6dex, France. 0165-3806/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
144 day old pups and adult rats. At least 6-10 cochleae per age were studied. After decapitation, the cochleae were removed and immediately put into 4 °C Carnoy fixative (absolute alcohol/chloroform/glacial acetic acid, 6:3:1). Dissection was performed in the same fixative. Cochleae were kept in the fixative for 2-3 h, and were then rinsed repeatedly in phosphate saline buffer (PBS) for several hours. Cochleae from 5 days on were decalcified in 10% EDTA, in pH 7.4 and rinsed in PBS. Subsequently they were put into a gradient of sucrose and frozen for cryostat sections of 6-8 btm. Cryostat sections were used for indirect immunofluorescence. For peroxidase-antiperoxidase (PAP) and avidin-biotin complex (ABC) peroxidase methods we used Carnoy fixed sections embedded in plastic medium. We used monoclonal antibodies raised against mouse lgG (Boehringer-Mannheim) against the 3 NF subunits. For indirect immunofluorescence, the second antibody was goat IgG conjugated with fluorescein isothiocyanate (FITC) (Boehringer). For the PAP method, the PAP complex was a mouse product (Jacksori Lab.). The bridging antibody was a sheep immunoglobulin product which recognizes mouse IgG. All the reactions for immunofluorescence and PAP were carried out at room temperature, in a humid chamber. For indirect immunofluorescence, sections were incubated with fetal calf serum (20%) for 30 min, to block unspecific binding. Sections were drained without washing and incubated with the primary antibody diluted 1/5 (4 /xg/ml), for 60-90 min. Sections were rinsed and covered with the second antibody coupled to FITC for 60-90 min, then rinsed and mounted for microscopic observations. For the PAP method, the first antibody was applied as for the immunofluorescence study. The second antibody was used in a 1/120 dilution (10 pg/ml), for 1 h; after rinsing, the PAP complex was
used in a 1/120 dilution, for 90 min, rinsed in Tris-HCl buffer and covered with a solution of 0.05 diaminobenzidine (DAB)/0,015 H20 2 in the same buffer. The reaction was stopped by removing DAB solution and flooding the section with distilled water. Sections were dehydrated and mounted for microscopic observations. ABC peroxidase (Vector Lab.) sections were first incubated with debiotinylated blocking serum for 20 min, drained without washing, and incubated with the first antibody as for the PAP method. Sections were washed in PBS for 15 min and incubated with a diluted biotinylated antibody solution for 30 min. After washing, sections were incubated with the Vectastain ABC reagent. The horseradish peroxidase reaction was then developed as for the PAP method. The number of reacting versus non-reacting neurons was evaluated from sections spaced 10/~m apart for at least two cochleae for each stage. All measurements were restricted to spiral ganglia from the basal turn. RESULTS
At all stages, from the 16th gestational day to the 10th postnatal day, peripheral and central processes of spiral neurons, as well as the fibers in the intraganglionic bundle react with the three antibodies to NF subunits, but the reaction is stronger with the NF 160 subunit in younger stages (Figs. 1A,B). The soma of prenatal spiral ganglion neurons, during 16, 18 and 20 days of gestation, present an immunoreactivity with only two antibodies, to the NF 68 and the NF 160 subunits. The reaction with the antibody NF 160 was the easiest to obtain (Fig. 1A,B). Around birth the spiral ganglion cell bodies react weakly against all 3 antibodies. During the early development, the microscopical
Fig. 1. Immunohistochemical reaction of spiral ganglion in neonatal stages. A,B: 16 day old rat fetus treated with monoclonal antibody against NF 160 subunit and visualized with ABC method. A: a reaction is visible throughout all the auditory nerve (large arrows) and spiral ganglion (small arrows). A higher magnification of the basal turn (B) shows a strong reaction either in the spiral ganglion neurons and in preganglionic fibers going towards the receptors (arrow) and postganglionic fibers (large arrow). A, x90; B, x360. C,D: 20 day old rat fetus treated with a monoclonal antibody against the NF 160 subunit. C: PAP method. D: indirect immunofluorescence. Many neurons still present a reaction, although some show stronger labelling (arrowheads) that is restricted to the perikaryon, x990. E,F: newborn rat pups treated with monoclonal antibody against NF I60 subunit (E) and NF 68 subunit (F) and revealed by ABC technique. Although several neurons still present a slight reaction, one clearly shows a stronger reaction product (arrow) in the perikaryon. E, ×450; F, x990.
145
146 detection of immunoreactivity in neuronal soma of spiral ganglion cells was difficult due to the small volume of their cytoplasm. Between 16 and 20 days of gestation, a narrow ring of immunoreactivity could be detected using immunofluorescence. At the
20th day of gestation, in the basal turn, a few neurons began to show a distinctly stronger immunoreactivity (Fig. 1C,D). These cells are easier to identify using PAP and ABC methods than with immunofluorescence. In the newborn animal, this
147 population of cells presents 3.7-9.5% (with a mean of 6.7%) and it is still restricted to the basal turn only (Fig. 1E,F). But already in the 2-3 day old cochlea, some spiral neurons display strong immunoreactivity within the entire ganglion. The reactivity for each NF subunit was restricted to the cytoplasm (Fig. 2A-F). The percentage of reacting neurons, measured in the basal turn at 6 days, equals 2.9-7.2 (with a mean of 5.2). This value remained similar for all ages studied including the adult (Fig. 2G-I). In the organ of Corti, labelling with the antibodies against all 3 subunits of NFs gives a strong reaction below both types of sensory cells, but particularly below the outer hair cells (OHCs). This distribution prevails up to about the 8th postnatal day; after which time the labelling intensity below the OHCs decreases. In the 10 day old animal, as in the adult, the immunoreactivity against all 3 subunits is still vivid below the inner hair cells (IHCs) but very weak below the OHCs. DISCUSSION
The cytoplasmic accumulation of NFs in the neuronal subpopulation is characteristic to many ganglia. This feature has been revealed by immunocytochemical methods in the trigeminal ganglion 2°, the dorsal root ganglion 17'2x and in sympathetic and parasympathetic neurons 36. Antibodies against NFs have been successfully
used in immunohistochemical methods to differentiate neurons in the spiral ganglion of the adult rat 3"1°'26. We used the same methods in the developing animals since it is difficult to distinguish the earliest appearance of T II neurons in electron microscopic observations 25"29. It has been established from the electron microscopic observations in the rat that the T II neurons were first observed around 6-8 days after birth in the rat. Using 3 different immunohistochemical methods, we found a subpopulation of spiral ganglion neurons presenting an accumulation of NFs in their perikarya in the perinatal stage. Their percentage corresponds closely to that of T II neurons in the adult. It has been shown previously26 that in the adult rat the percentage of neurons reacting against NF antibodies corresponds closely to the percentage of T II neurons found by more conventional methods 13. According to Anniko et al. ~, in the newborn mouse the immunoreactivity to a NF antibody is present only in the neuronal processes. Using the same technique (indirect immunofluorescence) but different antibodies we were able to detect in the rat reactive neurons already at the 16th day of gestation. We consider, however, the difference between our data as inconclusive because of the difference in experimental procedures. In the human fetus Yehoash et al. 38 found the immunoreactivity against NF in the spiral ganglion very early during development. In addition Anniko et al. 2 found in the human fetus of 14 weeks of gestation a discrete subpopulation of
Fig. 2. Immunohistochemical characterization of T II spiral ganglion neurons in rat pups and adult rats. A: 3 day old rat pup treated with a monoclonal antibody against the NF 160 subunit and revealed with the PAP method. Only a few neurons present a strong reaction in the ganglion (arrows) as the fibers from the intraganglionic spiral bundle (open arrowheads). ×450. B: 3 day old rat pup incubated with a monoclonal antibody against the NF 200 subunits and treated with the ABC method. At this magnification the reaction product is clearly visible in the neuron (large arrow). This neuron is surrounded by numerous non-reacting cells (small arrows), x 1350. C: 3 day old rat pup treated with a monoclonal antibody against the NF 160 subunit and revealed with the PAP technique. As for the previous figure, the reaction products are well restricted to the perikaryon. The nucleus presents an excentric position. × 1350. D-F: 6 day old rat pups treated with monoclonal antibodies against various subunits; (D, NF 200; E, NF 160; F, NF 68) and revealed with the indirect immunofluorescence method. High magnification of the few neurons that present a strong immunofluorescence to the 3 antibodies. For each antibody the reaction is restricted to the perikaryon, while the black round area in the center of T II cells corresponds to the nucleus. Non-responding neurons in the background represent T I neurons. × 1080. G: 10 day old rat pup incubated with monoclonal antibody against NF 200 subunit and revealed by indirect immunofluorescence method. A few neurons in the entire spiral ganglion present strong reactivity, while most preganglionic (small arrowheads) and postganglionic fibers (large arrows) show strong immunofluorescence. At this stage there is no difference in size between the two types of neurons, x360. H,I: adult rats treated with monoclonal antibody against NF 200 subunit and revealed with indirect immunofluorescence (H) and (I). H: in this section only 5 neurons present a strong reaction (arrowheads). Compare the size of the reacting neurons with those in the background which are larger. Small bright dots correspond to fibers cut transversally. Many come from the intraganglionic spiral bundle (large arrow), x360. I: higher magnification of a darkly labelled neuron (large arrow) that corresponds to T II neuron with a neighboring neuron presenting a slight reaction in its perikaryon (small arrow) that corresponds to T I neurons, x1080.
148 neurons that reacted strongly to a N F antibody. N F proteins are expressed in the neurons at the onset of their differentiation 4,7,s,23. It is therefore not surprising that most of the spiral ganglion cells reacted to N F antibodies. During early d e v e l o p m e n t all neurons synthesize a large quantity of N F necessary for the growth of their processes. In our observations, neuronal processes reacted with the 3 N F subunits, supporting the view that all N F subunits are present before birth 4"35. But perinatal neuronal soma as well as processes reacted stronger against the N F 160 subunit than against the two o t h e r subunits. The composition of NFs changes during d e v e l o p m e n t , with the highest molecular weight being either absent or u n p h o s p h o r y l a t e d in fetal or early postnatal nervous s y s t e m 4-6'22'3°. It has been shown that our antibody against the N F 200 subunit reacts only with the p h o s p h o r y l a t e d form of N F protein 9'31 and fails to stain p e r i k a r y a of pyramidal cells, m o t o n e u r o n s and Purkinje cells 31. In this respect, spiral ganglion neurons are exceptional, although recently, reacting cell bodies were observed from fetal neurons of the spinal cord s . The immunoreactivity of most neuronal soma subsides a r o u n d birth which may correspond to a
stabilization of neuronal growth 12. Conversely, the T II neurons still show evidence of strong i m m u n o r e activity due to either a strong synthesis or an accumulation of NFs in their cell bodies. The early detection of T II neurons by immunohistochemical methods is due to the fact that they can detect N F subunits before those form dense packs of filaments visualized ultrastructurally. The latency of N F subunits assembly into NFs seems to be very short 32. The late detection of T II neurons in the electron microscope may be due to the delay in aggregation of NFs into thick bundles. The strong immunohistochemical reactions observed below I H C s and O H C s must be related to peripheral processes of spiral ganglion neurons or efferent fibers. The modification of reactivity below O H C s during d e v e l o p m e n t may be related to variations of innervation in the organ of Corti 19. ACKNOWLEDGEMENTS This study was s u p p o r t e d by I N S E R M , R e s e a r c h G r a n t 86434. The authors would like to thank Dr. H. Sobkowicz for her helpful c o m m e n t s on the manuscript.
REFERENCES 1 Anniko, M., Thornell, L.E., Gustavsson, H. and Virtanen, I, Intermediate filaments in newborn inner ear of the mouse, O.R.L., 48 (1986) 98-106. 2 Anniko, M., Thornell, L.E. and Virtanen, I., Cytoskeletal organization of the human inner ear. II Characterization of intermediate filaments in the cochlea, Acta Otolaryngol., 437 (1987) 29-54. 3 Berglund, A.M. and Ryugo, D.K., A monoclonal antibody label type II neurons of the spiral ganglion, Brain Res., 383 (1986) 327-332. 4 Cochard, P. and Paulin, D., Initial expression of neurofilament and vimentine in the central and peripheral nervous system of the mouse embryo in vivo J. Neurosci., 8 (1984) 2080-2094. 5 Dahl, D., Crosby, C.J., Gardner, E.E. and Bignami, A., Delayed phosphorylation of the largest neurofilament subunit in rat optic nerve development, J. Neurosci. Res., 15 (1986) 513-519. 6 Dahl, D., Maturation of the large neurofilament protein NF 150 in rat postnatal development, J. Neurosci. Res., 17 (1987) 367-374. 7 Dahl, D., Gardner, E.E. and Crosby, C.J., Axonal maturation in development - - I. Characterization of monoclonal antibodies reacting with axon-specific neurofilament epitopes, Int. J. Dev. Neurosci., 5 (1987) 17-27. 8 Dahl, D., Early and late appearance of neurofilament phosphorylated epitopes in rat nervous system develop-
9
10 11
12 13 14 15 16
ment: in vivo and in vitro study with monoclonal antibodies, J. Neurosci., 20 (1988) 431-441. Debus, E., Weber, K. and Osborn, M., Monoclonal antibodies specific for glial fibriUary acidic (GFA) protein and for each of the neurofilament triplet polypeptides, Differentiation, 25 (1983) 193-202. Hafidi, A., Romand, R. and Ontoniente, B., Immunohistochemical characterization of T II cells in the spiral ganglion of the rat, Soc. Neurosci. Abstr., 13 (1987) 149. Hoffman, P.N. and Lasek, R.J., The slow component of axonal transport: Identification of major structural polypeptides of the axon of their generality among mammalian neurons, J. Cell Biol., 66 (1975) 351-360. Hoffman, EN., Griffin, J.W. and Price, D.L., Slowing of neurofilament transport and the radial growth of developing nerve fibers, J. Neurosci., 5 (1985) 2920-2929. Keithley, E.M. and Feidman, M.L., Spiral ganglion cell counts in an age graded series of rat cochleas, J. Comp. Neurol., 188 (1979) 429-442. Kiang, N.Y.S., Rho, J.M., Northrup, C.C., Liberman, M.C. and Ryugo, D.K., Hair cell innervation by spiral ganglion cells in adult cats, Science. 217 (1982) 175-177. Lazarides, E., Intermediate filaments: A chemically heterogenous, developmentally regulated class of proteins, Annu. Rev. Biochem., 51 (1982) 219-250. Lee, V.M.-Y., Wu, H.L. and Schlaepfer, W.W., Monoclonal antibodies recognize individual neurofilament triplet proteins, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 60896092.
149 17 Lee, V.M.-Y., Carden, M.J., and Schlaepfer, W.W., Monoclonal antibodies distinghuish several differentially phosphorylated states of the two largest rat neurofilament subunits (NF-H and NF-M) and demonstrate their existence in the normal nervous system of adult rats, J. Neurosci., 7 (1987) 3474-3488. 18 Liem, R.H.K., Yen, S.H., Solomon, G.E. and Shelanski, M.L., Intermediate filament in nervous tissues, J. Cell. Biol., 79 (1978) 637-645. 19 Lenoir, M., Shnerson, A. and Pujol, R., Cochlear receptor development in the rat with emphasis on synaptogenesis, Anat. Embryol., 160 (1980) 253-262. 20 Maeda, T., Sensory innervation of the periodontal ligament in the incisor and molar of the monkey Macaca fuscata. An immunohistochemical study for neurofilament protein, glial specific S-100 protein, Arch. Histol. Jpn., 50 (1987) 437-454. 21 Oblinger, M.M., Characterization of posttranslational processing of the mammalian high molecular weight neurofilament protein in vivo, J. Neurosci., 7 (1987) 2510-2521. 22 Patcher, J.S. and Liem, R.H.K., The differential appearance of neurofilament triplet polypeptides in the developing rat optic nerve, Dev. Biol., 103 (1984) 200-210. 23 Raju, T., Bignami, A. and Dahl, D., In vivo and in vitro differentiation of neurons and astrocytes in the rat embryo, Dev. Biol., 85 (1981) 344-357. 24 Romand, R. and Romand, M.R., The spiral ganglion. In I. Friedmann and J. Ballantyne (Eds.), Ultrastructural Atlas of the Inner Ear, Butterworths, London, 1984, pp. 165-183. 25 Romand, R. and Romand M.R., Qualitative and quantitative observations of spiral ganglion development in the rat, Hearing Res., 18 (1985) 111-120. 26 Romand, R., Hafidi, A. and Despres, G., Immunocytochemical localisation of neurofilaments protein subunits in the spiral ganglion of the adult rat, Brain Res., 462 (1988) 167-173. 27 Rosenbluth, J., The fine structure of acoustic ganglion of the rat, J. Cell Biol., 12 (1962) 33-41. 28 Schlaepfer, W.W. and Freeman, L.A., Neurofilament
29 30 31
32
33 34
35 36
37
38
protein of rat peripheral nerve and spinal cord, J. Cell Biol., 78 (1978) 653-662. Schwartz, A.M., Parakkal, A.M. and Gulley, R.L., Postnatal development of spiral ganglion cells in the rat, Am. J. Anat., 167 (1983) 33-41. Shaw, G. and Weber, K., Differential expression of neurofilament triplet proteins in brain development, Nature (Lond.), 198 (1982) 177-179. Shaw, G., Osborn, M. and Weber, K., Reactivity of a panel of neurofilament antibodies on phosphorylated and dephosphorylated neurofilaments, Eur. J. Cell Biol., 42 (1986) 1-9. Soifer, D., Czosnek, H.H., Mack, K. and Wisniewski, H.M., Properties and dynamics of neurofilament proteins. In D.G. Wiess (Ed.), Axoplasmic Transport, Springer, Berlin, 1982. Spoendlin, H., Innervation densities of the cochlea, Acta Otolaryngol., 73 (1972) 235-248. Sternberger, L.A. and Sternberger, N.H., Monoclonal antibodies distinguish phosphorylated and non phosphorylated forms of neurofilaments in situ, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 6126-6130. Tapscott, S.J., Bennett, G.S., Toyama, Y., Kleinbart, E and Holtzer, H., Intermediate filament proteins in developing chick spinal chord., Dev. Biol., 86 (1981) 40-54. Vitadello, M., Triban, C., Fabris, M., Dona, M., Gorio, A. and Schiaffino, S., A developmentally regulated isoform of 150,000 molecular weight neurofilament protein specifically expressed in autonomic and small sensory neurons, Neuroscience, 23 (1987) 931-941. Weber, K. and Geisler, N., Intermediate filaments from wool alpha keratin to neurofilaments: A structural overview. In A.J. Levine et al. (Eds.), Cancer Cells. 1. The Transformed Phenotype, Cold Spring Harbor Labs, Cold Spring Harbor, NY, 1984, pp. 153-159. Yehoash, R., Marshak, G., Barash, A. and Geiger, B., Modulation of intermediate-filament expression in developing cochlear epithelium, Differentiation, 35 (1987) 151162.
143
BRD 50935
First appearance of Type II neurons during ontogenesis in the spiral ganglion of the rat. An immunocytochemical study A. Hafidi and R. Romand Laboratoire de Neurobiologie, Universitd Blaise Pascal, Ensemble scientifique des Cdzeauz, Aubidre (France) (Accepted 7 February 1989) Key words: Neurofilament subunit; Spiral ganglion; Ontogenesis; Type II neuron; Immunocytochemistry; Rat
Ontogenesis of spiral ganglion in the rat was studied using antibodies to three subunits of neurofilaments (NFs): NF 68 KDa, NF 160 KDa and NF 200 KDa. The expression of immunoreactivity was examined with 3 immunocytochemical methods: indirect immunofluorescence, peroxidase-antiperoxidase and avidin-biotin complex. Aim of the study was to detect the time of differentiation of the spiral ganglion type II neurons. At 16 and 18 days of gestation, most neuron cell bodies express immunoreactivity to only two NF subunits: NF 68 and NF 160, but at birth they react with the antibodies to all 3 subunits albeit weakly. Nevertheless, a small population (about 7%) of nerve cells that strongly reacts against all 3 NF subunits emerges in the basal turn, already at 20 days of gestation. Two to 3 days after birth, the strongly stained cells are dispersed throughout the entire ganglion. The intensity of their reaction to the NF antibodies is similar to that seen in the adult animal. The strong immunoreactivity of this selective neuronal population suggest, that they correspond to the type II spiral ganglion neurons. Our results imply that the differentiation between the type I and the type II of spiral neurons in the rat occurs perinatally.
INTRODUCTION Spiral ganglion cells are afferent neurons o f the organ of Corti; their peripheral and central processes connect auditory receptors with the cochlear nucleus. Spiral ganglion contains at least two clearly distinct populations of neurons. These two types of cells are well characterized by their structural, ultrastructura124'27'3a and by biochemical features 3 and by their connections with receptors 14. Type I neurons (T I) represent 90-95% of ganglion cells; they have a large perikaryon rich in rough endoplasmic reticulum and ribosomes. Type II neurons (T II) represent 5 - 1 0 % of ganglion cells; they have a small perikaryon and show a characteristic accumulation of neurofilaments (NFs) 3,24,26,33. In the rat, T II neurons can be differentiated ultrastructurally only around the end of the first postnatal week 25'29. Since the accumulation of NFs is the main ultrastructural criterion distinguishing the
two types of neurons, their emergence is not very easy to detect in the electron microscope. In the present study, therefore we have used the antibodies to the 3 subunits of NFs. NFs are the neuron-specific intermediate filaments15,37; in mammals they are composed of 3 subunits with apparent molecular weights of 200 kDa (NF 200), 160 k D a (NF 160) and 68 k D a (NF 68) 11'18'28. The 3 subunits are immunochemically distinct 16,31,34. It has been shown previously that adult T II neurons strongly react with the 3 NF subunits s'26. The present study was undertaken to evaluate the presence of NF subunits in the developing ganglion cells and to characterize the first appearance of T II neurons. MATERIALS AND METHODS We used S p r a g u e - D a w l e y rat fetuses of 16, 18 and 20 days of gestation, newborn and 2-6, 8 and 10
Correspondence: R. Romand, Laboratoire de Neurobiologie, Universit6 Blaise Pascal, Ensemble scientifique des C6zeaux, 63177 Aubi6re C6dex, France. 0165-3806/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
144 day old pups and adult rats. At least 6-10 cochleae per age were studied. After decapitation, the cochleae were removed and immediately put into 4 °C Carnoy fixative (absolute alcohol/chloroform/glacial acetic acid, 6:3:1). Dissection was performed in the same fixative. Cochleae were kept in the fixative for 2-3 h, and were then rinsed repeatedly in phosphate saline buffer (PBS) for several hours. Cochleae from 5 days on were decalcified in 10% EDTA, in pH 7.4 and rinsed in PBS. Subsequently they were put into a gradient of sucrose and frozen for cryostat sections of 6-8 btm. Cryostat sections were used for indirect immunofluorescence. For peroxidase-antiperoxidase (PAP) and avidin-biotin complex (ABC) peroxidase methods we used Carnoy fixed sections embedded in plastic medium. We used monoclonal antibodies raised against mouse lgG (Boehringer-Mannheim) against the 3 NF subunits. For indirect immunofluorescence, the second antibody was goat IgG conjugated with fluorescein isothiocyanate (FITC) (Boehringer). For the PAP method, the PAP complex was a mouse product (Jacksori Lab.). The bridging antibody was a sheep immunoglobulin product which recognizes mouse IgG. All the reactions for immunofluorescence and PAP were carried out at room temperature, in a humid chamber. For indirect immunofluorescence, sections were incubated with fetal calf serum (20%) for 30 min, to block unspecific binding. Sections were drained without washing and incubated with the primary antibody diluted 1/5 (4 /xg/ml), for 60-90 min. Sections were rinsed and covered with the second antibody coupled to FITC for 60-90 min, then rinsed and mounted for microscopic observations. For the PAP method, the first antibody was applied as for the immunofluorescence study. The second antibody was used in a 1/120 dilution (10 pg/ml), for 1 h; after rinsing, the PAP complex was
used in a 1/120 dilution, for 90 min, rinsed in Tris-HCl buffer and covered with a solution of 0.05 diaminobenzidine (DAB)/0,015 H20 2 in the same buffer. The reaction was stopped by removing DAB solution and flooding the section with distilled water. Sections were dehydrated and mounted for microscopic observations. ABC peroxidase (Vector Lab.) sections were first incubated with debiotinylated blocking serum for 20 min, drained without washing, and incubated with the first antibody as for the PAP method. Sections were washed in PBS for 15 min and incubated with a diluted biotinylated antibody solution for 30 min. After washing, sections were incubated with the Vectastain ABC reagent. The horseradish peroxidase reaction was then developed as for the PAP method. The number of reacting versus non-reacting neurons was evaluated from sections spaced 10/~m apart for at least two cochleae for each stage. All measurements were restricted to spiral ganglia from the basal turn. RESULTS
At all stages, from the 16th gestational day to the 10th postnatal day, peripheral and central processes of spiral neurons, as well as the fibers in the intraganglionic bundle react with the three antibodies to NF subunits, but the reaction is stronger with the NF 160 subunit in younger stages (Figs. 1A,B). The soma of prenatal spiral ganglion neurons, during 16, 18 and 20 days of gestation, present an immunoreactivity with only two antibodies, to the NF 68 and the NF 160 subunits. The reaction with the antibody NF 160 was the easiest to obtain (Fig. 1A,B). Around birth the spiral ganglion cell bodies react weakly against all 3 antibodies. During the early development, the microscopical
Fig. 1. Immunohistochemical reaction of spiral ganglion in neonatal stages. A,B: 16 day old rat fetus treated with monoclonal antibody against NF 160 subunit and visualized with ABC method. A: a reaction is visible throughout all the auditory nerve (large arrows) and spiral ganglion (small arrows). A higher magnification of the basal turn (B) shows a strong reaction either in the spiral ganglion neurons and in preganglionic fibers going towards the receptors (arrow) and postganglionic fibers (large arrow). A, x90; B, x360. C,D: 20 day old rat fetus treated with a monoclonal antibody against the NF 160 subunit. C: PAP method. D: indirect immunofluorescence. Many neurons still present a reaction, although some show stronger labelling (arrowheads) that is restricted to the perikaryon, x990. E,F: newborn rat pups treated with monoclonal antibody against NF I60 subunit (E) and NF 68 subunit (F) and revealed by ABC technique. Although several neurons still present a slight reaction, one clearly shows a stronger reaction product (arrow) in the perikaryon. E, ×450; F, x990.
145
146 detection of immunoreactivity in neuronal soma of spiral ganglion cells was difficult due to the small volume of their cytoplasm. Between 16 and 20 days of gestation, a narrow ring of immunoreactivity could be detected using immunofluorescence. At the
20th day of gestation, in the basal turn, a few neurons began to show a distinctly stronger immunoreactivity (Fig. 1C,D). These cells are easier to identify using PAP and ABC methods than with immunofluorescence. In the newborn animal, this
147 population of cells presents 3.7-9.5% (with a mean of 6.7%) and it is still restricted to the basal turn only (Fig. 1E,F). But already in the 2-3 day old cochlea, some spiral neurons display strong immunoreactivity within the entire ganglion. The reactivity for each NF subunit was restricted to the cytoplasm (Fig. 2A-F). The percentage of reacting neurons, measured in the basal turn at 6 days, equals 2.9-7.2 (with a mean of 5.2). This value remained similar for all ages studied including the adult (Fig. 2G-I). In the organ of Corti, labelling with the antibodies against all 3 subunits of NFs gives a strong reaction below both types of sensory cells, but particularly below the outer hair cells (OHCs). This distribution prevails up to about the 8th postnatal day; after which time the labelling intensity below the OHCs decreases. In the 10 day old animal, as in the adult, the immunoreactivity against all 3 subunits is still vivid below the inner hair cells (IHCs) but very weak below the OHCs. DISCUSSION
The cytoplasmic accumulation of NFs in the neuronal subpopulation is characteristic to many ganglia. This feature has been revealed by immunocytochemical methods in the trigeminal ganglion 2°, the dorsal root ganglion 17'2x and in sympathetic and parasympathetic neurons 36. Antibodies against NFs have been successfully
used in immunohistochemical methods to differentiate neurons in the spiral ganglion of the adult rat 3"1°'26. We used the same methods in the developing animals since it is difficult to distinguish the earliest appearance of T II neurons in electron microscopic observations 25"29. It has been established from the electron microscopic observations in the rat that the T II neurons were first observed around 6-8 days after birth in the rat. Using 3 different immunohistochemical methods, we found a subpopulation of spiral ganglion neurons presenting an accumulation of NFs in their perikarya in the perinatal stage. Their percentage corresponds closely to that of T II neurons in the adult. It has been shown previously26 that in the adult rat the percentage of neurons reacting against NF antibodies corresponds closely to the percentage of T II neurons found by more conventional methods 13. According to Anniko et al. ~, in the newborn mouse the immunoreactivity to a NF antibody is present only in the neuronal processes. Using the same technique (indirect immunofluorescence) but different antibodies we were able to detect in the rat reactive neurons already at the 16th day of gestation. We consider, however, the difference between our data as inconclusive because of the difference in experimental procedures. In the human fetus Yehoash et al. 38 found the immunoreactivity against NF in the spiral ganglion very early during development. In addition Anniko et al. 2 found in the human fetus of 14 weeks of gestation a discrete subpopulation of
Fig. 2. Immunohistochemical characterization of T II spiral ganglion neurons in rat pups and adult rats. A: 3 day old rat pup treated with a monoclonal antibody against the NF 160 subunit and revealed with the PAP method. Only a few neurons present a strong reaction in the ganglion (arrows) as the fibers from the intraganglionic spiral bundle (open arrowheads). ×450. B: 3 day old rat pup incubated with a monoclonal antibody against the NF 200 subunits and treated with the ABC method. At this magnification the reaction product is clearly visible in the neuron (large arrow). This neuron is surrounded by numerous non-reacting cells (small arrows), x 1350. C: 3 day old rat pup treated with a monoclonal antibody against the NF 160 subunit and revealed with the PAP technique. As for the previous figure, the reaction products are well restricted to the perikaryon. The nucleus presents an excentric position. × 1350. D-F: 6 day old rat pups treated with monoclonal antibodies against various subunits; (D, NF 200; E, NF 160; F, NF 68) and revealed with the indirect immunofluorescence method. High magnification of the few neurons that present a strong immunofluorescence to the 3 antibodies. For each antibody the reaction is restricted to the perikaryon, while the black round area in the center of T II cells corresponds to the nucleus. Non-responding neurons in the background represent T I neurons. × 1080. G: 10 day old rat pup incubated with monoclonal antibody against NF 200 subunit and revealed by indirect immunofluorescence method. A few neurons in the entire spiral ganglion present strong reactivity, while most preganglionic (small arrowheads) and postganglionic fibers (large arrows) show strong immunofluorescence. At this stage there is no difference in size between the two types of neurons, x360. H,I: adult rats treated with monoclonal antibody against NF 200 subunit and revealed with indirect immunofluorescence (H) and (I). H: in this section only 5 neurons present a strong reaction (arrowheads). Compare the size of the reacting neurons with those in the background which are larger. Small bright dots correspond to fibers cut transversally. Many come from the intraganglionic spiral bundle (large arrow), x360. I: higher magnification of a darkly labelled neuron (large arrow) that corresponds to T II neuron with a neighboring neuron presenting a slight reaction in its perikaryon (small arrow) that corresponds to T I neurons, x1080.
148 neurons that reacted strongly to a N F antibody. N F proteins are expressed in the neurons at the onset of their differentiation 4,7,s,23. It is therefore not surprising that most of the spiral ganglion cells reacted to N F antibodies. During early d e v e l o p m e n t all neurons synthesize a large quantity of N F necessary for the growth of their processes. In our observations, neuronal processes reacted with the 3 N F subunits, supporting the view that all N F subunits are present before birth 4"35. But perinatal neuronal soma as well as processes reacted stronger against the N F 160 subunit than against the two o t h e r subunits. The composition of NFs changes during d e v e l o p m e n t , with the highest molecular weight being either absent or u n p h o s p h o r y l a t e d in fetal or early postnatal nervous s y s t e m 4-6'22'3°. It has been shown that our antibody against the N F 200 subunit reacts only with the p h o s p h o r y l a t e d form of N F protein 9'31 and fails to stain p e r i k a r y a of pyramidal cells, m o t o n e u r o n s and Purkinje cells 31. In this respect, spiral ganglion neurons are exceptional, although recently, reacting cell bodies were observed from fetal neurons of the spinal cord s . The immunoreactivity of most neuronal soma subsides a r o u n d birth which may correspond to a
stabilization of neuronal growth 12. Conversely, the T II neurons still show evidence of strong i m m u n o r e activity due to either a strong synthesis or an accumulation of NFs in their cell bodies. The early detection of T II neurons by immunohistochemical methods is due to the fact that they can detect N F subunits before those form dense packs of filaments visualized ultrastructurally. The latency of N F subunits assembly into NFs seems to be very short 32. The late detection of T II neurons in the electron microscope may be due to the delay in aggregation of NFs into thick bundles. The strong immunohistochemical reactions observed below I H C s and O H C s must be related to peripheral processes of spiral ganglion neurons or efferent fibers. The modification of reactivity below O H C s during d e v e l o p m e n t may be related to variations of innervation in the organ of Corti 19. ACKNOWLEDGEMENTS This study was s u p p o r t e d by I N S E R M , R e s e a r c h G r a n t 86434. The authors would like to thank Dr. H. Sobkowicz for her helpful c o m m e n t s on the manuscript.
REFERENCES 1 Anniko, M., Thornell, L.E., Gustavsson, H. and Virtanen, I, Intermediate filaments in newborn inner ear of the mouse, O.R.L., 48 (1986) 98-106. 2 Anniko, M., Thornell, L.E. and Virtanen, I., Cytoskeletal organization of the human inner ear. II Characterization of intermediate filaments in the cochlea, Acta Otolaryngol., 437 (1987) 29-54. 3 Berglund, A.M. and Ryugo, D.K., A monoclonal antibody label type II neurons of the spiral ganglion, Brain Res., 383 (1986) 327-332. 4 Cochard, P. and Paulin, D., Initial expression of neurofilament and vimentine in the central and peripheral nervous system of the mouse embryo in vivo J. Neurosci., 8 (1984) 2080-2094. 5 Dahl, D., Crosby, C.J., Gardner, E.E. and Bignami, A., Delayed phosphorylation of the largest neurofilament subunit in rat optic nerve development, J. Neurosci. Res., 15 (1986) 513-519. 6 Dahl, D., Maturation of the large neurofilament protein NF 150 in rat postnatal development, J. Neurosci. Res., 17 (1987) 367-374. 7 Dahl, D., Gardner, E.E. and Crosby, C.J., Axonal maturation in development - - I. Characterization of monoclonal antibodies reacting with axon-specific neurofilament epitopes, Int. J. Dev. Neurosci., 5 (1987) 17-27. 8 Dahl, D., Early and late appearance of neurofilament phosphorylated epitopes in rat nervous system develop-
9
10 11
12 13 14 15 16
ment: in vivo and in vitro study with monoclonal antibodies, J. Neurosci., 20 (1988) 431-441. Debus, E., Weber, K. and Osborn, M., Monoclonal antibodies specific for glial fibriUary acidic (GFA) protein and for each of the neurofilament triplet polypeptides, Differentiation, 25 (1983) 193-202. Hafidi, A., Romand, R. and Ontoniente, B., Immunohistochemical characterization of T II cells in the spiral ganglion of the rat, Soc. Neurosci. Abstr., 13 (1987) 149. Hoffman, P.N. and Lasek, R.J., The slow component of axonal transport: Identification of major structural polypeptides of the axon of their generality among mammalian neurons, J. Cell Biol., 66 (1975) 351-360. Hoffman, EN., Griffin, J.W. and Price, D.L., Slowing of neurofilament transport and the radial growth of developing nerve fibers, J. Neurosci., 5 (1985) 2920-2929. Keithley, E.M. and Feidman, M.L., Spiral ganglion cell counts in an age graded series of rat cochleas, J. Comp. Neurol., 188 (1979) 429-442. Kiang, N.Y.S., Rho, J.M., Northrup, C.C., Liberman, M.C. and Ryugo, D.K., Hair cell innervation by spiral ganglion cells in adult cats, Science. 217 (1982) 175-177. Lazarides, E., Intermediate filaments: A chemically heterogenous, developmentally regulated class of proteins, Annu. Rev. Biochem., 51 (1982) 219-250. Lee, V.M.-Y., Wu, H.L. and Schlaepfer, W.W., Monoclonal antibodies recognize individual neurofilament triplet proteins, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 60896092.
149 17 Lee, V.M.-Y., Carden, M.J., and Schlaepfer, W.W., Monoclonal antibodies distinghuish several differentially phosphorylated states of the two largest rat neurofilament subunits (NF-H and NF-M) and demonstrate their existence in the normal nervous system of adult rats, J. Neurosci., 7 (1987) 3474-3488. 18 Liem, R.H.K., Yen, S.H., Solomon, G.E. and Shelanski, M.L., Intermediate filament in nervous tissues, J. Cell. Biol., 79 (1978) 637-645. 19 Lenoir, M., Shnerson, A. and Pujol, R., Cochlear receptor development in the rat with emphasis on synaptogenesis, Anat. Embryol., 160 (1980) 253-262. 20 Maeda, T., Sensory innervation of the periodontal ligament in the incisor and molar of the monkey Macaca fuscata. An immunohistochemical study for neurofilament protein, glial specific S-100 protein, Arch. Histol. Jpn., 50 (1987) 437-454. 21 Oblinger, M.M., Characterization of posttranslational processing of the mammalian high molecular weight neurofilament protein in vivo, J. Neurosci., 7 (1987) 2510-2521. 22 Patcher, J.S. and Liem, R.H.K., The differential appearance of neurofilament triplet polypeptides in the developing rat optic nerve, Dev. Biol., 103 (1984) 200-210. 23 Raju, T., Bignami, A. and Dahl, D., In vivo and in vitro differentiation of neurons and astrocytes in the rat embryo, Dev. Biol., 85 (1981) 344-357. 24 Romand, R. and Romand, M.R., The spiral ganglion. In I. Friedmann and J. Ballantyne (Eds.), Ultrastructural Atlas of the Inner Ear, Butterworths, London, 1984, pp. 165-183. 25 Romand, R. and Romand M.R., Qualitative and quantitative observations of spiral ganglion development in the rat, Hearing Res., 18 (1985) 111-120. 26 Romand, R., Hafidi, A. and Despres, G., Immunocytochemical localisation of neurofilaments protein subunits in the spiral ganglion of the adult rat, Brain Res., 462 (1988) 167-173. 27 Rosenbluth, J., The fine structure of acoustic ganglion of the rat, J. Cell Biol., 12 (1962) 33-41. 28 Schlaepfer, W.W. and Freeman, L.A., Neurofilament
29 30 31
32
33 34
35 36
37
38
protein of rat peripheral nerve and spinal cord, J. Cell Biol., 78 (1978) 653-662. Schwartz, A.M., Parakkal, A.M. and Gulley, R.L., Postnatal development of spiral ganglion cells in the rat, Am. J. Anat., 167 (1983) 33-41. Shaw, G. and Weber, K., Differential expression of neurofilament triplet proteins in brain development, Nature (Lond.), 198 (1982) 177-179. Shaw, G., Osborn, M. and Weber, K., Reactivity of a panel of neurofilament antibodies on phosphorylated and dephosphorylated neurofilaments, Eur. J. Cell Biol., 42 (1986) 1-9. Soifer, D., Czosnek, H.H., Mack, K. and Wisniewski, H.M., Properties and dynamics of neurofilament proteins. In D.G. Wiess (Ed.), Axoplasmic Transport, Springer, Berlin, 1982. Spoendlin, H., Innervation densities of the cochlea, Acta Otolaryngol., 73 (1972) 235-248. Sternberger, L.A. and Sternberger, N.H., Monoclonal antibodies distinguish phosphorylated and non phosphorylated forms of neurofilaments in situ, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 6126-6130. Tapscott, S.J., Bennett, G.S., Toyama, Y., Kleinbart, E and Holtzer, H., Intermediate filament proteins in developing chick spinal chord., Dev. Biol., 86 (1981) 40-54. Vitadello, M., Triban, C., Fabris, M., Dona, M., Gorio, A. and Schiaffino, S., A developmentally regulated isoform of 150,000 molecular weight neurofilament protein specifically expressed in autonomic and small sensory neurons, Neuroscience, 23 (1987) 931-941. Weber, K. and Geisler, N., Intermediate filaments from wool alpha keratin to neurofilaments: A structural overview. In A.J. Levine et al. (Eds.), Cancer Cells. 1. The Transformed Phenotype, Cold Spring Harbor Labs, Cold Spring Harbor, NY, 1984, pp. 153-159. Yehoash, R., Marshak, G., Barash, A. and Geiger, B., Modulation of intermediate-filament expression in developing cochlear epithelium, Differentiation, 35 (1987) 151162.