- Email: [email protected]
The ultrastructural identification of reticulo-hypoglossal axon terminals anterogradely labeled with horseradish peroxidase
Brain Research, 337 (1985) 127-132 Elsevier
127
BRE 20877
Short Communications
The ultrastructural identification of reticulo-hypoglossal axon terminals anterogradely labeled with horseradish peroxidase ROSEMARY C. BORKE and MARTIN E. NAU Department of Anatomy, Uniformed Services University, Bethesda, MD 20814-4799 (U.S.A.) (Accepted January 28th, 1985) Key words: nucleus reticularis parvocellularis- - horseradish peroxidase (HRP) - - wheat germ agglutinin-HRP (WGA-HRP) - anterograde labeling - - axon terminals in hypoglossal nucleus
Injections of horseradish peroxidase (HRP) or wheat-germ agglutinin-horseradish peroxidase (WGA-HRP) into the nucleus reticularis parvocellularis (RPc) produced anterograde labeling of axon terminals within the hypoglossal nucleus. Based on morphological parameters of vesicle population, membrane specializations, and postsynaptic articulations, two types of axon terminals derived from neurons in RPc end on hypoglossal neurons. More than half of the terminals contained spherical vesicles (S-type), established asymmetrical membrane specializations and contacted proximal and medium-sized dendrites. The remaining labeled terminals had flattened vesicles (F-type), symmetrical membrane densities and apposed medium and small dendrites. The morphological differences expressed in the two types of terminals may reflect physiological and/or pharmacological differences in the action of RPc neurons on motoneurons in the hypoglossal nucleus.
The a r r a n g e m e n t of lingual muscles and reflex mechanisms of the tongue contribute to the highly organized and controlled m o v e m e n t s of this muscular organ. Neural pathways mediating c o o r d i n a t e d tongue m o v e m e n t s have not been c o m p l e t e l y identified, but m o t o n e u r o n s in the hypoglossal nucleus provide the only source of m o t o r innervation to the muscle mass. Physiological evidence suggests multisynaptic pathways control the activity of hypoglossal m o t o n e u r o n s in m a m m a l s o t h e r than primates; cortical regions 20 and cranial nerve ganglia9,11,31, 32 are implicated as sources of first o r d e r neurons and second o r d e r neurons reside chiefly in the medulla12,24, 27. Recently, the anatomical substrates providing direct hypoglossal input have been d e l i n e a t e d in the medulla by using r e t r o g r a d e tracers2, 28. Neurons in the dorsolateral p o r t i o n of the reticular formation, namely the nucleus reticularis parvocellularis, constitute the primary source of the hypoglossal afferents. Physio-
logical data12,20, 24 indicate that this portion of the reticular formation acts as a site for convergence of peripheral and cortical fibers and these p r e m o t o r neurons in turn transmit integrated responses to hypoglossal motoneurons. Since RPc assumes a p r i m a r y role in the coordination of tongue m o v e m e n t s , the ways in which RPc and the hypoglossal nucleus relate structurally to each other is a basis of interest. HRps,16,17 and W G A - H R p 1 8 have been used recently as anterograde m a r k e r s to characterize ultrastructural patterns of presynaptic connectivity; size of axon terminals, vesicle population, m e m b r a n e specializations and postsynaptic articulations. The object of the current investigation was to analyze the distribution and synaptic organization of these reticulo-hypoglossal connections and to interpret their significance. Experiments were carried out in 11 adult, male rats of the O s b o r n - M e n d e l strain, anesthetized with
Correspondence: R. C. Borke, Department of Anatomy, Uniformed Services University, 4301 Jones Bridge Road, Bethesda, Maryland 20814-4799, U.S.A. 0006-8993/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
128
Fig. 1. A: diffuse, agranular type of HRP label in axon terminal containing spherical vesicles and making synaptic contact at the site of the asymmetrical membrane specialization (arrow) with a medium-sized dendrite (d) in the neuropil of the XIIth nucleus. HRP injection into RPc, 7 days survival, CO-GOD, unstained, × 34,500. B: diffuse, agranular type of HRP label in axon terminal with flattened vesicles. This terminal establishes synaptic contact of the symmetrical type (arrow) with a small dendrite (d) in the neuropil of the
129 an intraperitoneal injection of sodium pentobarbital (30 mg/kg). For each specimen, the obex served as the stereotaxic reference point. A glass micropipette (20-30 ktm tip diameter) was introduced in a vertical axis into RPc (lateral to the midlength of the XIIth nucleus) using predetermined stereotaxic coordinates (1.1-1.3 mm lateral, 0.7-0.8 mm anterior, 0.5-0.7 mm depth) relative to the reference point. A single injection (15-60 nl) of 30% H R P or 1.8% W G A - H R P in 0.1 M Tris buffer was then made through the micropipette into RPc and the micropipette was left in place for 20 min after the injection to minimize the diffusion of the enzyme along the pipette track. Animals injected with the tracer were allowed to survive 16-24 h, 2, 3, 5 and 7 days and were reanesthetized and transcardially perfused with Ringer's solution then dilute and concentrated aldehydes 22 followed by a solution of 0.1 M phosphate buffer. Immediately after perfusion, and each brain removal, the medulla was isolated and mounted in agar. Coronal sections, 50/~m in thickness, were cut on a vibratome. Sample sections were stained with F i n k - H e i m e r and Nauta silver methods 4. The remaining sections were reacted histochemically for H R P according to the cobalt-glucose oxidase ( C O - G O D ) 10 or tetramethylbenzidine (TMB) 3,23 methods. Some of the sections reacted for H R P were mounted on gelatinized slides and counterstained with neutral red for light microscopic evaluation of the injection site and localization of areas of anterograde labeling. Remaining reacted sections were postfixed in osmium tetroxide, dehydrated and symmetrical halves of the medulla were embedded in Epon according to previously detailed procedures 3. Epoxy resin blocks were trimmed to include only the hypoglossal nucleus. Ultrathin sections of silver interference were cut and left unstained for ultrastructural evaluation. Light microscopic preparations of H R P reacted tissue revealed numerous labeled cell bodies and
processes of neurons in RPc. Nerve fibers filled with reaction product could be traced medially from the injection site into the XIIth nucleus. Occasionally, a few anterogradely labeled fibers traversed the medullary raphe below the ventricle and were distributed to the contralateral nucleus, particularly if the injection extended into the medial aspect of RPc (1.1 mm measurement for lateral stereotaxic coordinate). There was no evidence of direct or indirect damage at the' injection site by the traverse of the pipette or the pressure of the hydraulic injection. Silver staining of vibratome sections adjacent to those reacted for H R P failed to demonstrate anterograde degeneration of terminals in the hypoglossal nucleus or of axons in RPc. Ultrastructural examination of the hypoglossal nucleus disclosed that the reaction product was found almost exclusively in myelinated axons and axon terminals. Both TMB and C O - G O D methods labeled axon terminals containing spherical and flattened vesicles. Quantitative evaluation of 366 C O - G O D reacted terminals revealed: (1) 56% of labeled terminals harbored spherical vesicles (S-type) (Fig. 1A, C, D, E); and (2) 44% of labeled terminals contained flattened vesicles (F-type) (Fig. 1B, F). Asymmetrical membrane specializations were identified on 78% of the S-type terminals (Fig. 1A, C) and were rarely detected on F-type terminals, which characteristically displayed symmetrical membrane densities (Fig. 1B). The majority of S-type terminals (70%) ended on proximal (Fig. 1E) and medium-sized dendrites (Fig. 1A, C, D). Medium and small dendrites (Fig. 1B, F) served as the primary postsynaptic structures for the F-type terminals (72%). Most of the remaining S-type terminals contacted small dendrites (16%) whereas F-type terminals apposed proximal dendrites (20%). The postsynaptic structure could not be ascertained for a small percentage (4%) of the total terminals. The finding of axon terminals making multiple syn-
XIIth nucleus. HRP injection in RPc, 7 days survival, CO-GOD, unstained, × 44,800. C: S-type terminal with diffuse, agranular HRP label, apposes a medium-sized dendrite (d) in the neuropil of the XIIth nucleus. WGA-HRP injection in RPc, 24 h survival, COGOD, unstained, x 30,400. D: S-type terminal labeled with a crystal-like aggregate (arrow) when TMB was used as the chromogen. Terminal is in synaptic contact with medium-sized dendrite (d). WGA-HRP injection into RPc, 24 h survival, TMB, unstained, × 34,200. E: membrane-bound, granular type of HRP label (arrows) in S-type terminal. The postsynaptic structure is a proximal dendrite (d) containing ribosomes (r). Other unlabeled terminals of the S-type (s) and the F-type (f) also contact the dendrite. WGA-HRP injection into the RPc, 24 h survival, CO-GOD, unstained, x 27,300. F: membrane-bound, granular type of HRP label (arrow) in Ftype terminal contacting a small dendrite (d). WGA-HRP injection into RPc, 42 h survival, CO-GOD, unstained, x 37,000.
130 aptic contacts on proximal and medium-sized dendrites typifies the synaptic relationship of RPc neurons ending on other cranial nerve motor nuclei18 and is consistent with data derived from retrograde tracer studies identifying RPc as the source of primary afferents of the hypoglossal nucleus2, 28. RPC is thought to act as the site for convergence and integration of hypoglossal reflexes from spatially separated cortical and peripheral sources. Physiological evidence suggests that neurons located in the dorsolateral portion of the reticular formation adjacent to the hypoglossal nucleus may exert either excitatory or inhibitory effects on postsynaptic neurons 12. Although a close relationship between vesicle morphology and functional synaptic type is not universal, axon terminals containing spherical and flattened vesicles have often been associated respectively with excitatory and inhibitory effects on postsynaptic neurons 29. Whether this relationship applies to the current findings is not known. However, the characterization of two morphological types of axon terminals may reflect physiological and/or pharmacological differences in the actions of RPc neurons on hypoglossal motoneurons. In TMB-treated sections (Fig. 1D), the reaction product formed crystal-like aggregates that often penetrated membranes and disrupted the morphology of the terminals. On the other hand, the ultrastructural integrity of axons and their terminals was preserved with the CO-GOD (Fig. 1A, B, C, E, F) reaction thus confirming other reports TM that this method is preferable for fine structural and semiquantitative evaluation of synaptic relationships. After CO-GOD treatment with diaminobenzidine as the chromogen, the distribution of the reaction product in the axon terminals was characterized by two different patterns. In the majority of terminals, an electron-dense reaction product was applied to the cytomembranes of organelles within the terminals (Fig. 1A-C). This type of labeling has been interpreted to result primarily from intracytoplasmic diffusion of the enzyme into somata and axons damaged by the injection procedure~,5,6 and has been termed diffuse, agranular labeling. Electron-dense deposits of HRP were detected to a lesser extent within membrane-bound vesicles in some terminals (Fig. 1E, F). This kind of enzyme localization represents the incorporation and transport of the tracer by intact neurons in an anterograde direction16.19 and has been
designated membrane-bound, granular label. The types of terminals identified from the anterograde tracer study were compared to those characterized by use of electron microscopic degeneration techniques. Electrolytic lesions (Fig. 2A) comparable in size (<1 mm in diameter) to the HRP injection sites (Fig. 2B) were placed stereotaxically in RPc of 6 rats. Rats were fixed by transcardiac perfusion as indicated previously. Adjacent vibratome sections of the medulla were processed in a routine manner for plastic embedding or stained with silver stains for anterograde axonal and terminal degeneration. Marked degenerative changes were detected by light and electron microscopy in RPc and the hypoglossal nucleus. Degenerative changes were recognized mainly in the same types of terminals as those labeled after injection of HRP into RPc: large terminals with spherical (Fig. 2C, D) and flattened (Fig. 2D) vesicles made synaptic contact chiefly with large and medium and medium and small dendrites, respectively. These changes were apparent in the terminals 1-7 days after lesioning and by 9 days, inclusions containing degenerative debris could only be identified in glial cells. The possibility that some of the labeling of terminals in the hypoglossal nucleus after enzyme injection into RPc resulted from uptake by axons of passage was considered. Although neurons of several cell groups in the brainstem project axons that course through RPc 15, only one of these sources, the spinal V nucleus, pars interpolaris and oralis, supplies afferent input to the hypoglossal nucleus2. Previous evidence indicates that uptake of HRp13, 30 or WGAHRp21,25,26 by intact axonal shafts does not occur in the CNS. Retrograde cell label, far more common than anterograde label7, was never detected in spinal V neuronal somata suggesting that injured spinal V axons did not account for any of the labeled population of axon terminals after RPc injections. Furthermore, anterograde labeling after injection of HRP into pars interpolaris and oralis portions of the spinal V nucleus in two rats occurred in axon terminals that exhibited morphological features of synaptic organization that could be distinguished from the terminals labeled after the RPc injections. These findings will be reported in a separate communication. The current findings therefore suggest that the labeled axon terminals originate from neurons in RPc and that the
131
Fig. 2. A: bright-field micrograph of the lesion site (< 1 mm in diameter) in RPc lateral to the hypoglossal nucleus (XII) at a level of the medulla rostral to the obex. 4 days survival, bar = 1 mm. B: dark-field micrograph of an injection site of comparable size and location to the lesion site in A. Neurons of RPc are labeled with enzyme and small particles of labeled substance are situated in the hypoglossal nucleus (XII). No label is seen in the XIIth nerve (n). WGA-HRP, 2 days survival, bar = 1 mm. C: axon terminal of S-type (*) undergoing anterograde degeneration in the neuropil of the XIIth nucleus. Electrolytic lesion in RPc, 4 days survival, x 25,g00. D: axon terminals of F-type (f) and S-type (s) demonstrating degenerative changes in the neuropil of the XIIth nucleus. Electrolytic lesion in RPc, 4 days survival, x 22,300. H R P technique can be used to study the ultrastructural synaptic organization of reticulo-hypoglossal connections. The author is grateful to Dr. Malcolm Carpenter for his valuable suggestions and comments. The skilled typing of H e a t h e r W o n g is gratefully acknowledged. This work was supported by the D e p a r t m e n t of Defense, U n i f o r m e d Services University of the
Health Sciences, D e p a r t m e n t of Defense G r a n t CO 7019. The opinions or assertions contained herein are the private ones of the author and are not to be construed as official or reflecting the views of the D o D or the U S U H S . The experiments reported herein were conducted according to the principles set forth in the 'Guide for Care and Use of Laboratory Animals', Institute of Laboratory A n i m a l Resources, National Research Council, D H E W Pub. No. (NIH) 78-23.
132 1 Beattie, M. S., Bresnahan, J. C. and King, J. S., Ultrastructural identification of dorsal root primary afferent terminals after retrograde filling with horseradish peroxidase, Brain Research, 153 (1978) 127-134. 2 Borke, R. C., Nau, M. E. and Ringler, R. L. Jr., Brain stem afferents of hypoglossal neurons in the rat, Brain Research, 269 (1983) 47-55. 3 Carson, K. A. and Mesulam, M. M., Electron microscopic demonstration of neural connections using horseradish peroxidase: a comparison of the tetramethylbenzidine procedure with other histochemical methods, ]. Histochem. Cytochem., 30 (1982) 425-435. 4 Ebbesson, S. O. E., The selective silver impregnation of degenerating axons and their synaptic endings in non-mammalian species, In W. J. H. Nauta and S. O. E. Ebbesson (Eds.), Contemporary Research Methods in Neuroanatomy, Springer-Verlag, New York, 1970, pp. 132-161. 5 Gobel, S. and Falls, W. M., Anatomical observations of horseradish peroxidase filled terminal primary axonal arborization in layer II of the substantia gelatinosa of Rolando, Brain Research, 175 (1979) 335-340. 6 Gwyn, D. G., Wilkinson, P. H. and Leslie, R. A., The ultrastructural identification of vagal terminals in the solitary nucleus of the cat after anterograde labeling with horseradish peroxidase, Neurosci. Lett., 28 (1982) 139-143. 7 Hedreen, J. C. and McGrath, S., Observations on labeling of neuronal cell bodies, axons and terminals after injection of horseradish peroxidase into rat brain, J. comp. Neurol., 176 (1977) 225-246. 8 Holstege, J. G. and Dekker, J. J., Electron microscopic identification of mammillary body terminals in the rat's AV thalamic nucleus by means of anterograde transport of HRP. A quantitative comparison with the EM degeneration and EM autoradiographic techniques, Neurosci. Lett., 11 (1979) 129-135. 9 Hunter, I. W. and Porter, R., Glossopharyngeal influences on hypoglossal motoneurons in the cat, Brain Research, 74 (1974) 161-166. 10 Itoh, K., Konishi, A., Nomura, S., Mizuno, N., Na,kamura, Y. and Sugimoto, T., Application of coupled oxidation reaction to electron microscopic demonstration horseradish peroxidase: cobalt-glucose oxidase method, Brain Research, 175 (1979) 341-346. 11 Kaiga, Y., Linguo-hypoglossal reflex elicited by mechanical stimulation in rabbits, J. Osaka Odont. Soc., 43 (1980) 449-460. 12 Lowe, A. A., Excitatory and inhibitory inputs to hypoglossal motoneurons and adjacent reticular formation neurons in cats, Exp. Neurol., 62 (1978) 30-47. 13 Lynch, G., Gall, C., Mensah, P. and Cotman, C., Horseradish peroxidase histochemistry: a new method for tracing efferent projections in the central nervous system, Brain Research, 65 (1974) 373-380. 14 Mawe, G. M., Bresnahan, J. C. and Beattie, M. S., Ultrastructure of HRP-labeled neurons: a comparison of two sensitive techniques, Brain Res. Bull., 10 (1983) 551-558. 15 Mehler, W. R., Observations on the connectivity of the parvicellular-reticular formation with respect to a vomiting center, Brain Behav. Evol., 23 (1983) 63-80. 16 Mizuno, N., Konishi, A., Itoh, K., Iwahori, N. and Nakamura, Y., Identification of axon terminals of cerebello-olivary fibers in the cat: an electron microscopic study using
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
anterograde horseradish peroxidase method, Neurosci. Lett., 20 (1980) 11-14. Mizuno, N., Nomura, S., Itoh, K., Nakamura, Y. and Konishi, A., Commissural interneurons for masticating motoneurons: a light and electron microscope study using horseradish peroxidase tracer technique, Exp. Neurol., 59 (1978) 254-262. Mizuno, N., Yasui, Y., Nomura, S., Itoh, K., Konishi, A., Takada, M. and Kudo, M., A light and electron microscopic study of premotor neurons for the trigeminal motor nucleus, J. comp. Neurol., 215 (1983) 290-298. Oldfield, B. J., Hou-Yu, A. and Silverman, A. J., Technique for simultaneous ultrastructural demonstration of anterogradely transported horseradish peroxidase and an immunocytochemically identified neuropeptide, J. Histochem. Cytochem., 31 (1983) 1145-1150. Porter, R., Cortical actions on hypoglossal motoneurons in cats: a proposed role for a common internuncial cell, J. Physiol. (Lond.), 193 (1967) 295-308. Pugh, W. and Kalia, M., Differential uptake of peroxidase (HRP) and peroxidase-lectin (HRP-WGA) conjugate injected in the nodose ganglion of the cat, J. Histochem. Cytochem., 30 (1982) 887-894. Reese, T. S. and Karnovsky, M. J., Fine structural localization of a blood brain barrier to exogenous peroxidase, J. Cell Biol., 34 (1967) 207-217. Schonitzer, K. and Hollander, H., Anterograde tracing of horseradish peroxidase (HRP) with the electron microscope using the tetramethyl-benzidene reaction, J. Neurosci. Meth., 4 (1981) 373-383. Sessle, B. J., Excitatory and inhibitory inputs to single neurons in the solitary tract nucleus and adjacent reticular formation, Brain Research, 53 (1973) 319-331. Staines, W. A., Kimura, H., Fibiger, H. C. and McGeer, E. G., Peroxidase-labeled lectin as a neuroatomical tracer: evaluation in a CNS pathway, Brain Research, 197 (1980) 485-490. Steindler, D. A., Differences in the labeling of axons of passage by wheat germ agglutinin after uptake by cut peripheral nerve versus injections within the central nervous system, Brain Research, 250 (1982) 159-167. Sumino, R. and Nakamura, Y., Synaptic potential of hypoglossal motoneurons and a common inhibitory interneuron in the trigemino-hypoglossal reflex, Brain Research, 73 (1974) 439-454. Travers, J. B. and Norgren, R., Afferent projections to the oral motor nuclei in the rat, J. cornp. Neurol., 220 (1983) 280-298. Uchizono, K., Characteristics of excitatory and inhibitory synapses in the central nervous system of the cat, Nature (Lond.), 207 (1965) 642-643. Wakefield, C. and Shonnard, N., Observations of HRP labeling following injection through a chronically implanted cannula - - a method to avoid diffusion of HRP into injured fibers, Brain Research, 168 (1979) 221-226. Yamamoto, T., Fujiwara, T., Matsuo, R. and Kawamura, Y., Hypoglossal motor nerve activity elicited by taste and thermal stimuli applied to the tongue in rats, Brain Research, 238 (1982) 89-104. Yokota, T., Nishikawa, Y. and Ohno, S., A hypoglossal reflex elicited by mechanical stimulation of the mandibular mucosa in the cat, Jap. J. Physiol., 28 (1978) 659-666.
127
BRE 20877
Short Communications
The ultrastructural identification of reticulo-hypoglossal axon terminals anterogradely labeled with horseradish peroxidase ROSEMARY C. BORKE and MARTIN E. NAU Department of Anatomy, Uniformed Services University, Bethesda, MD 20814-4799 (U.S.A.) (Accepted January 28th, 1985) Key words: nucleus reticularis parvocellularis- - horseradish peroxidase (HRP) - - wheat germ agglutinin-HRP (WGA-HRP) - anterograde labeling - - axon terminals in hypoglossal nucleus
Injections of horseradish peroxidase (HRP) or wheat-germ agglutinin-horseradish peroxidase (WGA-HRP) into the nucleus reticularis parvocellularis (RPc) produced anterograde labeling of axon terminals within the hypoglossal nucleus. Based on morphological parameters of vesicle population, membrane specializations, and postsynaptic articulations, two types of axon terminals derived from neurons in RPc end on hypoglossal neurons. More than half of the terminals contained spherical vesicles (S-type), established asymmetrical membrane specializations and contacted proximal and medium-sized dendrites. The remaining labeled terminals had flattened vesicles (F-type), symmetrical membrane densities and apposed medium and small dendrites. The morphological differences expressed in the two types of terminals may reflect physiological and/or pharmacological differences in the action of RPc neurons on motoneurons in the hypoglossal nucleus.
The a r r a n g e m e n t of lingual muscles and reflex mechanisms of the tongue contribute to the highly organized and controlled m o v e m e n t s of this muscular organ. Neural pathways mediating c o o r d i n a t e d tongue m o v e m e n t s have not been c o m p l e t e l y identified, but m o t o n e u r o n s in the hypoglossal nucleus provide the only source of m o t o r innervation to the muscle mass. Physiological evidence suggests multisynaptic pathways control the activity of hypoglossal m o t o n e u r o n s in m a m m a l s o t h e r than primates; cortical regions 20 and cranial nerve ganglia9,11,31, 32 are implicated as sources of first o r d e r neurons and second o r d e r neurons reside chiefly in the medulla12,24, 27. Recently, the anatomical substrates providing direct hypoglossal input have been d e l i n e a t e d in the medulla by using r e t r o g r a d e tracers2, 28. Neurons in the dorsolateral p o r t i o n of the reticular formation, namely the nucleus reticularis parvocellularis, constitute the primary source of the hypoglossal afferents. Physio-
logical data12,20, 24 indicate that this portion of the reticular formation acts as a site for convergence of peripheral and cortical fibers and these p r e m o t o r neurons in turn transmit integrated responses to hypoglossal motoneurons. Since RPc assumes a p r i m a r y role in the coordination of tongue m o v e m e n t s , the ways in which RPc and the hypoglossal nucleus relate structurally to each other is a basis of interest. HRps,16,17 and W G A - H R p 1 8 have been used recently as anterograde m a r k e r s to characterize ultrastructural patterns of presynaptic connectivity; size of axon terminals, vesicle population, m e m b r a n e specializations and postsynaptic articulations. The object of the current investigation was to analyze the distribution and synaptic organization of these reticulo-hypoglossal connections and to interpret their significance. Experiments were carried out in 11 adult, male rats of the O s b o r n - M e n d e l strain, anesthetized with
Correspondence: R. C. Borke, Department of Anatomy, Uniformed Services University, 4301 Jones Bridge Road, Bethesda, Maryland 20814-4799, U.S.A. 0006-8993/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
128
Fig. 1. A: diffuse, agranular type of HRP label in axon terminal containing spherical vesicles and making synaptic contact at the site of the asymmetrical membrane specialization (arrow) with a medium-sized dendrite (d) in the neuropil of the XIIth nucleus. HRP injection into RPc, 7 days survival, CO-GOD, unstained, × 34,500. B: diffuse, agranular type of HRP label in axon terminal with flattened vesicles. This terminal establishes synaptic contact of the symmetrical type (arrow) with a small dendrite (d) in the neuropil of the
129 an intraperitoneal injection of sodium pentobarbital (30 mg/kg). For each specimen, the obex served as the stereotaxic reference point. A glass micropipette (20-30 ktm tip diameter) was introduced in a vertical axis into RPc (lateral to the midlength of the XIIth nucleus) using predetermined stereotaxic coordinates (1.1-1.3 mm lateral, 0.7-0.8 mm anterior, 0.5-0.7 mm depth) relative to the reference point. A single injection (15-60 nl) of 30% H R P or 1.8% W G A - H R P in 0.1 M Tris buffer was then made through the micropipette into RPc and the micropipette was left in place for 20 min after the injection to minimize the diffusion of the enzyme along the pipette track. Animals injected with the tracer were allowed to survive 16-24 h, 2, 3, 5 and 7 days and were reanesthetized and transcardially perfused with Ringer's solution then dilute and concentrated aldehydes 22 followed by a solution of 0.1 M phosphate buffer. Immediately after perfusion, and each brain removal, the medulla was isolated and mounted in agar. Coronal sections, 50/~m in thickness, were cut on a vibratome. Sample sections were stained with F i n k - H e i m e r and Nauta silver methods 4. The remaining sections were reacted histochemically for H R P according to the cobalt-glucose oxidase ( C O - G O D ) 10 or tetramethylbenzidine (TMB) 3,23 methods. Some of the sections reacted for H R P were mounted on gelatinized slides and counterstained with neutral red for light microscopic evaluation of the injection site and localization of areas of anterograde labeling. Remaining reacted sections were postfixed in osmium tetroxide, dehydrated and symmetrical halves of the medulla were embedded in Epon according to previously detailed procedures 3. Epoxy resin blocks were trimmed to include only the hypoglossal nucleus. Ultrathin sections of silver interference were cut and left unstained for ultrastructural evaluation. Light microscopic preparations of H R P reacted tissue revealed numerous labeled cell bodies and
processes of neurons in RPc. Nerve fibers filled with reaction product could be traced medially from the injection site into the XIIth nucleus. Occasionally, a few anterogradely labeled fibers traversed the medullary raphe below the ventricle and were distributed to the contralateral nucleus, particularly if the injection extended into the medial aspect of RPc (1.1 mm measurement for lateral stereotaxic coordinate). There was no evidence of direct or indirect damage at the' injection site by the traverse of the pipette or the pressure of the hydraulic injection. Silver staining of vibratome sections adjacent to those reacted for H R P failed to demonstrate anterograde degeneration of terminals in the hypoglossal nucleus or of axons in RPc. Ultrastructural examination of the hypoglossal nucleus disclosed that the reaction product was found almost exclusively in myelinated axons and axon terminals. Both TMB and C O - G O D methods labeled axon terminals containing spherical and flattened vesicles. Quantitative evaluation of 366 C O - G O D reacted terminals revealed: (1) 56% of labeled terminals harbored spherical vesicles (S-type) (Fig. 1A, C, D, E); and (2) 44% of labeled terminals contained flattened vesicles (F-type) (Fig. 1B, F). Asymmetrical membrane specializations were identified on 78% of the S-type terminals (Fig. 1A, C) and were rarely detected on F-type terminals, which characteristically displayed symmetrical membrane densities (Fig. 1B). The majority of S-type terminals (70%) ended on proximal (Fig. 1E) and medium-sized dendrites (Fig. 1A, C, D). Medium and small dendrites (Fig. 1B, F) served as the primary postsynaptic structures for the F-type terminals (72%). Most of the remaining S-type terminals contacted small dendrites (16%) whereas F-type terminals apposed proximal dendrites (20%). The postsynaptic structure could not be ascertained for a small percentage (4%) of the total terminals. The finding of axon terminals making multiple syn-
XIIth nucleus. HRP injection in RPc, 7 days survival, CO-GOD, unstained, × 44,800. C: S-type terminal with diffuse, agranular HRP label, apposes a medium-sized dendrite (d) in the neuropil of the XIIth nucleus. WGA-HRP injection in RPc, 24 h survival, COGOD, unstained, x 30,400. D: S-type terminal labeled with a crystal-like aggregate (arrow) when TMB was used as the chromogen. Terminal is in synaptic contact with medium-sized dendrite (d). WGA-HRP injection into RPc, 24 h survival, TMB, unstained, × 34,200. E: membrane-bound, granular type of HRP label (arrows) in S-type terminal. The postsynaptic structure is a proximal dendrite (d) containing ribosomes (r). Other unlabeled terminals of the S-type (s) and the F-type (f) also contact the dendrite. WGA-HRP injection into the RPc, 24 h survival, CO-GOD, unstained, x 27,300. F: membrane-bound, granular type of HRP label (arrow) in Ftype terminal contacting a small dendrite (d). WGA-HRP injection into RPc, 42 h survival, CO-GOD, unstained, x 37,000.
130 aptic contacts on proximal and medium-sized dendrites typifies the synaptic relationship of RPc neurons ending on other cranial nerve motor nuclei18 and is consistent with data derived from retrograde tracer studies identifying RPc as the source of primary afferents of the hypoglossal nucleus2, 28. RPC is thought to act as the site for convergence and integration of hypoglossal reflexes from spatially separated cortical and peripheral sources. Physiological evidence suggests that neurons located in the dorsolateral portion of the reticular formation adjacent to the hypoglossal nucleus may exert either excitatory or inhibitory effects on postsynaptic neurons 12. Although a close relationship between vesicle morphology and functional synaptic type is not universal, axon terminals containing spherical and flattened vesicles have often been associated respectively with excitatory and inhibitory effects on postsynaptic neurons 29. Whether this relationship applies to the current findings is not known. However, the characterization of two morphological types of axon terminals may reflect physiological and/or pharmacological differences in the actions of RPc neurons on hypoglossal motoneurons. In TMB-treated sections (Fig. 1D), the reaction product formed crystal-like aggregates that often penetrated membranes and disrupted the morphology of the terminals. On the other hand, the ultrastructural integrity of axons and their terminals was preserved with the CO-GOD (Fig. 1A, B, C, E, F) reaction thus confirming other reports TM that this method is preferable for fine structural and semiquantitative evaluation of synaptic relationships. After CO-GOD treatment with diaminobenzidine as the chromogen, the distribution of the reaction product in the axon terminals was characterized by two different patterns. In the majority of terminals, an electron-dense reaction product was applied to the cytomembranes of organelles within the terminals (Fig. 1A-C). This type of labeling has been interpreted to result primarily from intracytoplasmic diffusion of the enzyme into somata and axons damaged by the injection procedure~,5,6 and has been termed diffuse, agranular labeling. Electron-dense deposits of HRP were detected to a lesser extent within membrane-bound vesicles in some terminals (Fig. 1E, F). This kind of enzyme localization represents the incorporation and transport of the tracer by intact neurons in an anterograde direction16.19 and has been
designated membrane-bound, granular label. The types of terminals identified from the anterograde tracer study were compared to those characterized by use of electron microscopic degeneration techniques. Electrolytic lesions (Fig. 2A) comparable in size (<1 mm in diameter) to the HRP injection sites (Fig. 2B) were placed stereotaxically in RPc of 6 rats. Rats were fixed by transcardiac perfusion as indicated previously. Adjacent vibratome sections of the medulla were processed in a routine manner for plastic embedding or stained with silver stains for anterograde axonal and terminal degeneration. Marked degenerative changes were detected by light and electron microscopy in RPc and the hypoglossal nucleus. Degenerative changes were recognized mainly in the same types of terminals as those labeled after injection of HRP into RPc: large terminals with spherical (Fig. 2C, D) and flattened (Fig. 2D) vesicles made synaptic contact chiefly with large and medium and medium and small dendrites, respectively. These changes were apparent in the terminals 1-7 days after lesioning and by 9 days, inclusions containing degenerative debris could only be identified in glial cells. The possibility that some of the labeling of terminals in the hypoglossal nucleus after enzyme injection into RPc resulted from uptake by axons of passage was considered. Although neurons of several cell groups in the brainstem project axons that course through RPc 15, only one of these sources, the spinal V nucleus, pars interpolaris and oralis, supplies afferent input to the hypoglossal nucleus2. Previous evidence indicates that uptake of HRp13, 30 or WGAHRp21,25,26 by intact axonal shafts does not occur in the CNS. Retrograde cell label, far more common than anterograde label7, was never detected in spinal V neuronal somata suggesting that injured spinal V axons did not account for any of the labeled population of axon terminals after RPc injections. Furthermore, anterograde labeling after injection of HRP into pars interpolaris and oralis portions of the spinal V nucleus in two rats occurred in axon terminals that exhibited morphological features of synaptic organization that could be distinguished from the terminals labeled after the RPc injections. These findings will be reported in a separate communication. The current findings therefore suggest that the labeled axon terminals originate from neurons in RPc and that the
131
Fig. 2. A: bright-field micrograph of the lesion site (< 1 mm in diameter) in RPc lateral to the hypoglossal nucleus (XII) at a level of the medulla rostral to the obex. 4 days survival, bar = 1 mm. B: dark-field micrograph of an injection site of comparable size and location to the lesion site in A. Neurons of RPc are labeled with enzyme and small particles of labeled substance are situated in the hypoglossal nucleus (XII). No label is seen in the XIIth nerve (n). WGA-HRP, 2 days survival, bar = 1 mm. C: axon terminal of S-type (*) undergoing anterograde degeneration in the neuropil of the XIIth nucleus. Electrolytic lesion in RPc, 4 days survival, x 25,g00. D: axon terminals of F-type (f) and S-type (s) demonstrating degenerative changes in the neuropil of the XIIth nucleus. Electrolytic lesion in RPc, 4 days survival, x 22,300. H R P technique can be used to study the ultrastructural synaptic organization of reticulo-hypoglossal connections. The author is grateful to Dr. Malcolm Carpenter for his valuable suggestions and comments. The skilled typing of H e a t h e r W o n g is gratefully acknowledged. This work was supported by the D e p a r t m e n t of Defense, U n i f o r m e d Services University of the
Health Sciences, D e p a r t m e n t of Defense G r a n t CO 7019. The opinions or assertions contained herein are the private ones of the author and are not to be construed as official or reflecting the views of the D o D or the U S U H S . The experiments reported herein were conducted according to the principles set forth in the 'Guide for Care and Use of Laboratory Animals', Institute of Laboratory A n i m a l Resources, National Research Council, D H E W Pub. No. (NIH) 78-23.
132 1 Beattie, M. S., Bresnahan, J. C. and King, J. S., Ultrastructural identification of dorsal root primary afferent terminals after retrograde filling with horseradish peroxidase, Brain Research, 153 (1978) 127-134. 2 Borke, R. C., Nau, M. E. and Ringler, R. L. Jr., Brain stem afferents of hypoglossal neurons in the rat, Brain Research, 269 (1983) 47-55. 3 Carson, K. A. and Mesulam, M. M., Electron microscopic demonstration of neural connections using horseradish peroxidase: a comparison of the tetramethylbenzidine procedure with other histochemical methods, ]. Histochem. Cytochem., 30 (1982) 425-435. 4 Ebbesson, S. O. E., The selective silver impregnation of degenerating axons and their synaptic endings in non-mammalian species, In W. J. H. Nauta and S. O. E. Ebbesson (Eds.), Contemporary Research Methods in Neuroanatomy, Springer-Verlag, New York, 1970, pp. 132-161. 5 Gobel, S. and Falls, W. M., Anatomical observations of horseradish peroxidase filled terminal primary axonal arborization in layer II of the substantia gelatinosa of Rolando, Brain Research, 175 (1979) 335-340. 6 Gwyn, D. G., Wilkinson, P. H. and Leslie, R. A., The ultrastructural identification of vagal terminals in the solitary nucleus of the cat after anterograde labeling with horseradish peroxidase, Neurosci. Lett., 28 (1982) 139-143. 7 Hedreen, J. C. and McGrath, S., Observations on labeling of neuronal cell bodies, axons and terminals after injection of horseradish peroxidase into rat brain, J. comp. Neurol., 176 (1977) 225-246. 8 Holstege, J. G. and Dekker, J. J., Electron microscopic identification of mammillary body terminals in the rat's AV thalamic nucleus by means of anterograde transport of HRP. A quantitative comparison with the EM degeneration and EM autoradiographic techniques, Neurosci. Lett., 11 (1979) 129-135. 9 Hunter, I. W. and Porter, R., Glossopharyngeal influences on hypoglossal motoneurons in the cat, Brain Research, 74 (1974) 161-166. 10 Itoh, K., Konishi, A., Nomura, S., Mizuno, N., Na,kamura, Y. and Sugimoto, T., Application of coupled oxidation reaction to electron microscopic demonstration horseradish peroxidase: cobalt-glucose oxidase method, Brain Research, 175 (1979) 341-346. 11 Kaiga, Y., Linguo-hypoglossal reflex elicited by mechanical stimulation in rabbits, J. Osaka Odont. Soc., 43 (1980) 449-460. 12 Lowe, A. A., Excitatory and inhibitory inputs to hypoglossal motoneurons and adjacent reticular formation neurons in cats, Exp. Neurol., 62 (1978) 30-47. 13 Lynch, G., Gall, C., Mensah, P. and Cotman, C., Horseradish peroxidase histochemistry: a new method for tracing efferent projections in the central nervous system, Brain Research, 65 (1974) 373-380. 14 Mawe, G. M., Bresnahan, J. C. and Beattie, M. S., Ultrastructure of HRP-labeled neurons: a comparison of two sensitive techniques, Brain Res. Bull., 10 (1983) 551-558. 15 Mehler, W. R., Observations on the connectivity of the parvicellular-reticular formation with respect to a vomiting center, Brain Behav. Evol., 23 (1983) 63-80. 16 Mizuno, N., Konishi, A., Itoh, K., Iwahori, N. and Nakamura, Y., Identification of axon terminals of cerebello-olivary fibers in the cat: an electron microscopic study using
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
anterograde horseradish peroxidase method, Neurosci. Lett., 20 (1980) 11-14. Mizuno, N., Nomura, S., Itoh, K., Nakamura, Y. and Konishi, A., Commissural interneurons for masticating motoneurons: a light and electron microscope study using horseradish peroxidase tracer technique, Exp. Neurol., 59 (1978) 254-262. Mizuno, N., Yasui, Y., Nomura, S., Itoh, K., Konishi, A., Takada, M. and Kudo, M., A light and electron microscopic study of premotor neurons for the trigeminal motor nucleus, J. comp. Neurol., 215 (1983) 290-298. Oldfield, B. J., Hou-Yu, A. and Silverman, A. J., Technique for simultaneous ultrastructural demonstration of anterogradely transported horseradish peroxidase and an immunocytochemically identified neuropeptide, J. Histochem. Cytochem., 31 (1983) 1145-1150. Porter, R., Cortical actions on hypoglossal motoneurons in cats: a proposed role for a common internuncial cell, J. Physiol. (Lond.), 193 (1967) 295-308. Pugh, W. and Kalia, M., Differential uptake of peroxidase (HRP) and peroxidase-lectin (HRP-WGA) conjugate injected in the nodose ganglion of the cat, J. Histochem. Cytochem., 30 (1982) 887-894. Reese, T. S. and Karnovsky, M. J., Fine structural localization of a blood brain barrier to exogenous peroxidase, J. Cell Biol., 34 (1967) 207-217. Schonitzer, K. and Hollander, H., Anterograde tracing of horseradish peroxidase (HRP) with the electron microscope using the tetramethyl-benzidene reaction, J. Neurosci. Meth., 4 (1981) 373-383. Sessle, B. J., Excitatory and inhibitory inputs to single neurons in the solitary tract nucleus and adjacent reticular formation, Brain Research, 53 (1973) 319-331. Staines, W. A., Kimura, H., Fibiger, H. C. and McGeer, E. G., Peroxidase-labeled lectin as a neuroatomical tracer: evaluation in a CNS pathway, Brain Research, 197 (1980) 485-490. Steindler, D. A., Differences in the labeling of axons of passage by wheat germ agglutinin after uptake by cut peripheral nerve versus injections within the central nervous system, Brain Research, 250 (1982) 159-167. Sumino, R. and Nakamura, Y., Synaptic potential of hypoglossal motoneurons and a common inhibitory interneuron in the trigemino-hypoglossal reflex, Brain Research, 73 (1974) 439-454. Travers, J. B. and Norgren, R., Afferent projections to the oral motor nuclei in the rat, J. cornp. Neurol., 220 (1983) 280-298. Uchizono, K., Characteristics of excitatory and inhibitory synapses in the central nervous system of the cat, Nature (Lond.), 207 (1965) 642-643. Wakefield, C. and Shonnard, N., Observations of HRP labeling following injection through a chronically implanted cannula - - a method to avoid diffusion of HRP into injured fibers, Brain Research, 168 (1979) 221-226. Yamamoto, T., Fujiwara, T., Matsuo, R. and Kawamura, Y., Hypoglossal motor nerve activity elicited by taste and thermal stimuli applied to the tongue in rats, Brain Research, 238 (1982) 89-104. Yokota, T., Nishikawa, Y. and Ohno, S., A hypoglossal reflex elicited by mechanical stimulation of the mandibular mucosa in the cat, Jap. J. Physiol., 28 (1978) 659-666.