Brain Research, 144 (1978) 1-10 © Elsevier/North-Holland Biomedical Press
1
Research Reports
I N T R A - A X O N A L F E R R I C I O N - F E R R O C Y A N I D E S T A I N I N G OF N O D E S OF R A N V I E R A N D I N I T I A L S E G M E N T S I N C E N T R A L M Y E L I N A T E D FIBERS
STEPHEN G. WAXMAN* and DONALD C. QUICK** Department of Neurology, Harvard Medical School, Beth Israel Hospital, Boston, Mass. 02215 and Harvard-MIT Program in Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, Mass. 02139 (U.S.A.)
(Accepted July 21st, 1977)
SUMMARY Ferric ion and ferrocyanide were used to stain central nervous tissue from the spinal cords of rats following fixation in cacodylate-buffered aldehydes. At the nodes of Ranvier in myelinated fibers, the stain was localized primarily on the inner surface of the unmyelinated nodal axolemma, as had been reported previously for peripheral nodes. Unmyelinated initial segments of myelinated neurons were similarly stained, but the axon hillock, cell body and dendrites were not stained. Synapses also exhibited no staining. Details of stain localization and comparison of these results with other ultrastructural data suggest that the stain is specific for the node of Ranvier and the axon initial segment, and are consistent with the idea that the axolemma at these sites may be structurally different from the cell membrane in other regions of the neuron, including paranodal and internodal regions of the axon. The electron-dense substance underlying the cytoplasmic surface of the membrane at the nodes and initial segments may represent a substrate that serves to bind together membrane structures in specialized regions of the axolemma.
INTRODUCTION Recent evidence from a variety of sources demonstrates that the axolemma in
* Please address reprint requests to Dr. Waxman at the Harvard Neurological Unit, Beth Israel Hospital, 330 Brookline Ave., Boston, Mass. 02215, U.S.A. ** Present address : DepartmentofNeurology, UniversityofMinnesota MedicalSchool, Minneapolis, Minn. 55455, U.S.A.
myelinated fibers exhibits a high degree of local structural specialization. This linear differentiation is spatially related to the location of the myelinated internodes, with nodal membrane exhibiting properties different from internodal or paranodal membrane~,a,v,l~,~a,lv,ls, 2`z. This phenomenon of local membrane differentialion raises basic questions of how the axon membrane specializations are developed and maintained, and how dysfunction of the axon may be related to various pathological conditions such as demyelination (see e.g. ref. 17 and 21). I n our initial studies of these topics, we explored the feasibility of using ferric ion and ferrocyanide for specific cytochemical staining of certain regions of the axolemma in the peripheral nervous system 13,1'~,23. In this report, we extend our work to include the central nvrvous system. Either ferric ion or ferrocyanide will bind to the cytoplasmic side of the axolemma at nodes of Ranvier in peripheral nerves j'~. We have shown that this cytochemical binding is specific for normal nodes, in that there is no reaction with internodal or paranodal axolemma 13, unmyelinated C-fiber axolemma~:L or the inexcitable nodes of Sternarchus electric organ fibersr'L We expected that central nodes of Ranvier should also be intensely stained by the ferric ion-ferrocyanide technique. Further, we suspected that the unmyelinated initial segments of myelinated fibers should also react strongly with ferric ion and ferrocyanide, because of the morphological and physiological similarity of the axon membrane at the iaitial segments '2,1°,le and nodes of Ranvier ~,:~,~1. In the present report, we documented the staining of central nodes and initial segments with ferric ion and ferrocyanide, and discuss the relation of our results to the concept of local differentiation of the axolemma in myelinated fibers. MATERIALS A N D METHODS
Adult Sprague-Dawley rats were anesthetized with pentobarbital and perfused through the heart with a solution of 4 }iS paraformaldehyde and 5 ~ glutaraldehyde in 0.15 M cacodylate buffer at pH 7.3.5,~. The cervical and thoracic spinal cord was immediately removed, immersed in fresh fixative, and gently teased longitudinally into a number of thin pieces. Teasing of the cord facilitates penetration of the osmium fixative and staining solutions in later stages of preparation. After 2.5 h initial fixation, the tissue was thoroughly washed in cacodylate buffer, postfixed in 1'),, OsO,~ in cacodylate buffer (pH 7.3) at room temperature for 1.5 h, washed in distilled water 3 :~: 5 rain, immersed in 0.01 M FeCla for 60 rain, rinsed 5 min in distilled water, immersed in 1 i~ K4Fe(CN)6"3H20 (pH 2) for 20 rain, rinsed 5 rain in distilled water, dehydrated in a graded series of ethanol solutions, and embedded in Epon-Araldite. 3ktm sections were mounted on glass slides and examined by light microscopy;sections were photographed (with an orange filter, but without optical contrasting), and were resectioned for electron microscopy ~9. Silver ultrathin sections were examined in a Jeol 100B electron microscope operating at 60 kV, without further staining. RESULTS
At central nodes of Ranvier, an electron-dense precipitate is found at the inner
Fig. 1. A node of Ranvier (arrows) in rat spinal cord, stained with ferric ion and ferrocyanide. The stain appears as an electron-dense precipitate in the nodal axoplasm, a, axoplasm; scale - 10 t*m. Fig. 2. An enlargement from Fig. 1, showing that the primary site of stain deposition is on the cytoplasmic surface of the nodal axolemma. Some stain usually diffuses into the axoplasm, but, in this case, the diffusion is minimal, a, axoplasm; m, myelin terminal loops; scale -- 1/*m.
Fig. 3. An initial segment (i) in rat spinal cord, stained with ferric ion and ferrocyanide. The unmyelinated initial segment is above the arrows; the first myelinated internode is below. Note that the stain has diffused rather extensively into the axoplasm. In this case, the densest region o f staining is near the distal end of the initial segment. A nearby synapse (s) is not stained, m, myelin ; scale ! m~
Fig. 4. Electron micrograph o f parts of the initial segment (i) and cell body (b) of a spinal neuron, stained with ferric ion and ferrocyanide. The initial segment is densely stained, but the axon hillock (h) is not appreciably stained. Scale = 1/zm. Inset: light micrograph o f the same cell. Scale = 10#m.
Fig. 5. Electron micrograph o f part of the cell body (b), the axon hillock (h) and part ot the initial segment (i) o f a spinal neuron stained with ferric ion and ferrocyanide. In this case, there is a gradient o f stain, with the densest accumulation at the proximal region of the initial segment. Scale .... I #m. Inset: light micrograph o f the same cell, showing staining of the initial segment. Scale .... 10 t¢m.
(axoplasmic) surface of the nodal axolemma (Figs. 1 and 2). The stain usually diffuses to some extent into the nodal axoplasm, but the densest concentrations are located adjacent to the nodal axolemma, lntracellular membranes are impermeable to the stain, even after osmium fixation; thus, the dense precipitate is confined to the axoplasmic space and is excluded from endoplasmic vesicles and mitochondria (Fig. 2). Myelinated regions of the axolemma, including both the internodes and the paranodal segments where the layers of myelin terminate, are not stained. The extracellular 'gap substance' at the node is also left unstained. These results are similar to those that we have previously described for peripheral nerves 13,15. Nodes that serve as branch points (at which two myelinated secondary branches arise) are stained similarly; there is a dense precipitate subjacent to unmyelinated parts of the axolemma, both in the main fiber and in the myelinated branch. Observations were not made on unmyelinated collaterals arising at nodes, nor were they made at nodes at which synapses arise `)°. In the initial segment of a myelinated fiber, the ferric ion-ferrocyanide stain is localized inside the unmyelinated axolemma. As with nodes of Ranvier, the densest deposits of stain are usually located adjacent to the inner surface of the axolemma, but ~he tendency for stain to diffuse into the axoplasm is generally more pronounced in the case of initial segments. Microtubules appear to have some slight affinity for the stain (Figs. 3 5). At the distal end of the initial segment, staining is observed to end where the myelin begins (Fig. 3). At the proximal end of the initial segment, where the axon originates from the cell body, the stain may extend to the base of the axon, but the perikaryon itself is not stained. Dendrites are not stained, and, if there is a distinct axon hillock, that region is also left unstained (Fig. 4). The intensity of staining in the initial segment may be relatively uniform (Fig. 4), or it may show a gradient towards one end or the other (Fig. 5). Axosomatic and axodendritic (Fig. 3) synapses do not exhibit staining. In central nodes, and in initial segments, the stain may appear blue or black in the light microscope, depending on the chemical nature of the precipitate. The blue stain is Prussian blue (Fe4[Fe(CN)6]3); the black stain results from reaction of ferrocyanide with an osmium compound 15. For our purposes, this duplicity is of little importance, since the two stains appear to be identical in their localization properties. I f the ferric chloride step is eliminated from the staining procedure, only the black stain results. If ferrocyanide is eliminated (treatment with ferric chloride only), there is no apparent staining. These results are consistent with our previous findings on the cytochemistry of peripheral nodes 15. DISCUSSION
The present results demonstrate the binding of ferric ion and ferrocyanide to the cytoplasmic surface of the axolemma at nodes of Ranvier and axon initial segments in the central nervous system. These results confirm and extend the earlier findings of similar staining at peripheral nodes of Ranvier. The factors which determine localization of ferric ion and ferrocyanide at nodes are complex, and are discussed in an earlier
publication 15. In light microscopic studies of cupric ion and ferrocyanide binding at nodes of Ranvier in cat spinal cord fixed in phosphate-buffered glutaraldehyde, Hildebrand 4 reported the localization of precipitates primarily in the nodal gap substance but also, in some cases, within the nodal axon. Our own data, on both unfixed 14 and fixed 15 peripheral axons, indicate that stain localization is highly dependent on preparative conditions. In particular, binding of ferric ion to the nodal gap substance s,9 depends on pretreatment with inorganic phosphate. Following fixation in cacodylate-buffered aldehydes and osmium tetroxide, as was done in the present study, ferric ion and ferrocyanide are bound to the cytoplasmic surface of the axolemma at normal peripheral nodes of Ranvier, but not the the paranodal or internodal axolemma, the axolemma of C-fibers, or the specialized inexcitable nodes of Ranvier in the gymnotid Sternarchus 13,~'~','~:3. Other data concerning the local differentiation of central and peripheral fibers has come from a variety of sources. Freeze-fracture studies of central nodes of Ranvier have shown a high density of external (E) face membrane particles at the nodal axolemma, and it has been suggested that these may be related to sodium channels ts. In the Sternarchus electrocyte axons, freeze-fracture studies showed a higher density of E-face particles at type 1 nodes, some of which are excitable, than at the larger inexcitable type 1l nodes, which exhibited particle density values in the same range as the internodes 7. Electrophysiological data suggests that, in some peripheral myelinated fibers, the internodal axolemma may be relatively inexcitablO 6. It has also been shown through labeling with radioactive saxitoxin that sodium channels in peripheral myelinated fibers are concentrated in the unmyelinated nodal portion of the axolemma 17. The results of the present study provide cytochemical evidence for a differentiation of the nodal and internodal regions of the axon membrane in the central nervous system. The present findings also add weight to the suggestion 10 that the axon initial segment may exhibit structural modifications of the cell surface related to specific membrane properties. In this regard, freeze-fracture studies of the initial segment should be expected to reveal a high density of external face intramembranous particles. The freeze-fracture 7,1s and pharmacologicaP 7 data, together with our previous results13, 2'~ and those presented in this paper, thus all point toward a high degree of regional differentiation, in terms of membrane structure, in the axolemma of myelinated fibers. Since membranes are generally accepted as fluid under normal conditions, this raises the question as to how this differentiation, which may involve the distribution of ionic pumps or channels, is achieved. Rosenbluth 18 suggested that a combination of membrane flow and barriers at nodal membranes could account for the concentration of certain membrane elements at the nodes. Another possibility is that specialized membrane structures, such as ion channels or ion pumps, may be kept in clusters by cross-linking to one another, to other membrane structures, or to nonmembranous elements. In this regard, we are led to suggest that the electron-dense substance that underlies the axolemma at excitable nodes of Ranvier 1,3,1 t,~e and at the initial segment 2,1°,~2 may represent a substrate that serves to bind together membrane structures in specialized regions of the axolemma.
ACKNOWLEDGEMENTS T h i s w o r k was s u p p o r t e d in p a r t by g r a n t s f r o m t h e N a t i o n a l I n s t i t u t e s o f H e a l t h , the N a t i o n a l M u l t i p l e Sclerosis S o c i e t y , a n d the H e a l t h Sciences F u n d . Dr. W a x m a n is the r e c i p i e n t o f a R e s e a r c h C a r e e r Scientist A w a r d f r o m the N a t i o n a l I n s t i t u t e o f N e u r o l o g i c a l a n d C o m m u n i c a t i v e D i s o r d e r s a n d Stroke. W e t h a n k M i s s E. H a r t w i e g f o r excellent technical assistance.
REFERENCES 1 Andres, K. H., Uber die Feinstruktur besonderer Einrichtungen in markhaltigen Nervenfasern des Kleinhirns der Ratte, Z. ZellJbrsch., 65 (1965) 701-712. 2 Conradi, S., Observations on the ultrastructure of the axon hillock and initial axon segment of lumbosacral motoneurons in the cat, Actaphysiol. seand., Suppl., 332 (1969) 65 84. 3 Elfvin, L. G., The ultrastructure of the nodes of Ranvier in cat sympathetic nerve fibers, J. ultrastruct. Res., 5 (1961) 374-387. 4 Hildebrand, C. and Skoglund, S., Histochemical studies of adult and developing feline spinal cord white matter, Actaphysiol. scand., Suppl., 364 (1971) 145-173. 5 Karnovsky, M. J., A formaldehyde-glutaraldehyde fixative of high osmolality for use inelectron microscopy, J. Cell Biol., 27 (1965) 137a-138a. 6 Karnovsky, M. J., The ultrastructural basis of capillary permeability studied with peroxidase as a tracer, J. Cell Biol., 35 (1967) 213-236. 7 Kristol, C., Akert, K., Sandri, C., Wyss, U., Bennett, M. V. L. and Moor, H., The Ranvier nodes in the neurogenic electric organ of the knife fish, Sternarehus: a freeze-etching study on the distribution of membrane-associated particles, Brain Research, 125 (1977) 197-212. 8 Landon, D. N. and Langley, O. K., The local chemical environment of nodes of Ranvier: a study of cation binding, J. Anat. (Lond.), 108 (1971) 419-432. 9 Langley, O. K. and Landon, D. N., A light and electron histochemical approach to the node of Ranvier and myelin of peripheral nerve fibers, J. Histochem. Cytochem., 15 (1968) 722-731. 10 Palay, S. L., Sotelo, C., Peters, A. and Orkand, P. M., The axon hillock and the initial segment, J. Cell Biol., 38 (1968) 193-201. 11 Peters, A., The node of Ranvier in the central nervous system, Quart. J. exp. Physiol., 51 (1966) 229 236. 12 Peters, A., Proskauer, C. C. and Kaiserman-Abramof, I. R., The small pyramidal neuron of the rat cerebral cortex: the axon hillock and initial segment, J. Cell BioL, 39 (1968) 604-619. 13 Quick, D. C. and Waxman, S. G., Specific staining of the axon membrane at nodes of Ranvier with ferric ion and ferrocyanide, J. nem'ol. Sci., 31 (1977) 1-11. 14 Quick, D. C. and Waxman, S. G., Evidence for inorganic phosphate binding at nodes of Ranvier in peripheral nerves, J. neurol. Sci., 33 (1977) 207 212. 15 Quick, D. C. and Waxman, S. G., Ferric ion, ferrocyanide and phosphate as cytochemical reactants at peripheral nodes of Ranvier, J. Neurocytol. (1977) in press. 16 Rasminsky, M. and Sears, T. A., Internodal conduction in undissected demyelinated nerve fibres, J. Physiol. (Lond.), 227 (1972) 323-350. 17 Ritchie, J. M. and Rogart, R. B., The density of sodium channels in mammalian myelinated nerve fibers and the nature of the axonal membrane under the myelin sheath, Proc. nat. Acad. Sci., (Wash.), 74 (1977) 211-215. 18 Rosenbluth, J., Intramembranous particle distribution at the node of Ranvier and adjacent axolemma in myelinated axons of the frog brain, J. Neurocytol., 5 (1976) 731-745. 19 Schabtach, E. and Parkening, T. A., A method for sequential high resolution light and electron microscopy of selected areas of the same material, J. Cell Biol., 61 (1974) 261-264. 20 Waxman, S. G., Ultrastructural differentiation of the axon membrane at synapticand non-synaptic central nodes of Ranvier, Brain Research, 65 (1974) 338-342. 21 Waxman, S. G., Conduction in myeli nated, unmyelinated, and demyelinated fibers, Arch. Neurol. (Chit'.), (1977) in press.
10 22 Waxman, S. G., Pappas, G, D. and Bennett, M. V. L., Morphological correlates of t~unctional differentiation of nodes of Ranvier along single fibers in the neurogenic electric organ of the k~il'e fish, Sternarchus, J. Cell Biol., 53 (1972) 210-224. 23 Waxman, S. G. and Quick, D. C., Cytochemical differentiation of the axon mernbv~tnc i~l A- ~nd C-fibers, J. Neurol. Neurosurg. Psychiat., 40 (1977) 379-386.