- Email: [email protected]
Computerized ultrastructural analysis of the shape of the active synaptic zones in rat spinal cord
EXPERIMENTAL
NEUROLOGY
108,151-155
(1990)
Computerized Ultrastructural Analysis of the Shape of the Active Synaptic Zones in Rat Spinal Cord W. TERRELL STAMPS, Marine
Biomedical Institute, and Biophysics,
RICHARD Department University
E. COGGESHALL,
AND CLAIRE
E. HULSEBOSCH’
of Anatomy and Neurosciences and Department of Texas Medical Branch, Galveston, Texas 77550
Active synaptic zones are cytoplasmic specializations that indicate where synaptic transmission occurs. We have used computerized three-dimensional reconstructions from serial ultrathin sections to define certain features of the geometry of these zones in mammalian spinal cord. Our main finding is that the active zones in the dorsal portion of the spinal cord can be placed in one of two categories with respect to curvature: (1) uncurved or slightly curved and (2) very curved. The very curved category is associated with simple axodendritic type synapses in which the axonal terminal arises from pri0 1990 Academic Press, Inc. mary afferent fibers.
INTRODUCTION The dorsal horn of the mammalian spinal cord is an important area for somatosensory integration. Consequently the synaptic architecture of this region is a subject of importance. Synapses in the dorsal horn, as elsewhere, consist of presynaptic terminals, membrane and cytoplasmic specializations which are referred to as active synaptic zones, (11) and postsynaptic structures which are usually a dendrite or a soma (11, 15). The shape of the active zone and, in particular, the amount of curvature are aspects of dorsal horn synaptic architecture that have not been extensively examined. Since the cytoplasmic specializations that form the active zone are the sites where synaptic transmission presumably occurs, the size, shape, and extent of curvature of the active zone should be determined and related to fiber type if possible. To examine this issue, we have devised an index of curvature for the active zones in laminae I-IV of the rat dorsal horn and then used serial reconstructions and computer modeling to facilitate the gathering of the data. Our goal is to describe quantitatively the shapes and sizes, and particularly the exact extent of curvature of the active zones in laminae I-IV of normal animals, to show that a particular class within this ’ To whom correspondence should be addressed at Marine Biomedical Institute, 200 University, Galveston, TX 77550.
of Physiology
grouping is affected following dorsal rhizotomy. This is a step toward correlating the geometry of spinal synapses with meaningful functional data. METHODS Four adult male Sprague-Dawley rats were anesthetized with sodium pentobarbital (Nembutal, 35 mg/kg ip). When anesthesia was deep, a laminectomy was done, and dorsal roots L3-Ca2 were cut unilaterally midway between the ganglion and the spinal cord. Seventy-two hours later, the animals were reanesthetized as above and perfused intraventricularly with 0.9% NaCl containing 200 IU heparin and 0.02% sodium nitrite. When the vascular system was free of blood, the perfusion mixture was changed to 3% glutaraldehyde, 3% formaldehyde, and 0.1% picric acid in 0.1 M, pH 7.4, cacodylate buffer. Following fixation, the S2 spinal segment was removed and divided transversely into three approximately equal slices with a razor blade. The slices were postfixed in 1% osmium tetroxide and 1.5% potassium ferricyanide. After being embedded in plastic, 14-24 serial thin sections of laminae I-IV of each dorsal horn were cut, placed on Formvar-coated grids, and stained with 0.1% lead citrate. The dorsal horn in the first section was photographed and a montage of approximately 1200~ (actual magnifications from calibrated grids) was made and the boundaries of laminae I-IV were drawn as described previously (12). Points on this micrograph were chosen randomly and used as the centers of high power pictures (approximately 40,000X, actual magnifications from calibrated grids) which were taken through the entire set of serial sections. Section thickness was determined by measuring the thickness of a fold in the section and dividing it by 2. To reconstruct the active zones, mitochondria were selected as fiducial marks to align the planes of the reconstruction. Contours of the active zones and the presynaptic and postsynaptic elements were traced from serial sections onto a digitizing tablet and entered into a 80386 personal computer using three-dimensional reconstruction software, version 1.2, from the Laboratory of High Voltage Electron Microscopy of Boulder Colo151 All
Copyright 0 1990 rights of reproduction
0014-4886/90
$3.00
by Academic Press, Inc. in any
form
reserved.
152
STAMPS,
COGGESHALL,
AND
HULSEBOSCH
SYNAPTIC
SHAPE
ANALYSIS
IN
RAT
SPINAL
153
CORD
B 1.6
60
0.8 0.6
1.61-1.71-1.81-1.91-2.011.70 1.60 1.90 2.00
2.10
D
1.8
65
5.416.60
1.6
60 30 5w
25
i+ 2
20 0.8
t
15 10 5 0 l.OO1.10
l.ll1.20
1.211.30
1.311.40
1.411 so
1.511.60
INDEX OF CURVATURE
1.61-1.71-1.81-1.91-2.011.70 1.80 1.90
2.00
2.10
5.416.50
INDEX OF CURVATURE
FIG. 2. Histograms of indices of curvatures for (A) slightly curved active zones and (B) very curved active zones (abscissa) percentage total synapses (ordinate) in the dorsal horn of normal rats. Histograms of similar data in rats after dorsal rhizotomy in C and D. Note that the very curved synapses (D) are essentially eliminated.
rado (8,9). An enhanced graphics monitor displayed the reconstructions. The three-dimensional reconstruction of serial sections compensates for biases due to sampling in a single plane with respect to the random orientation of synapses within the dorsal horn and thus is in accord with stereological principles. A total of 270 active zones were examined in normal rats (n = 4) and 210 in rhizotomized rats (n = 4). Reconstructed active zones were divided into populations using an index of curvature (Ic)
plotted against are displayed
which was determined by dividing the longest boundary trace length of the active zone cross section by the distance between the end points of the zone. This method yielded the smallest standard deviations for our calculations. An Ic of 1 would indicate a flat active zone and a half circle would have an Ic of 1.57. Active zones were considered slightly curved if they had a value of greater than 1 but less than 1.6 and very curved if they had a value of greater than 1.6. In the dorsal horn, distinctions
FIG. 1. (A) An electron micrograph of several vesicle-filled presynaptic terminals. A curved active synaptic zone is present near the center of the picture, between the two arrows. Calibration bar = 0.5 pm. (B) A three-dimensional reconstruction of a presynaptic terminal region (light gray), a postsynaptic dendrite with a small dendritic spine extending toward the top of the picture (medium gray), and the very curved active synaptic zone which is associated with these two structures (dark lines).
154
STAMPS.
COGGESHALL,
are made between simple axodendritic synapses, where one presynaptic element contacts one postsynaptic element, and glomerular terminals, where one presynaptic element contacts two or more postsynaptic elements. For this reason, we separated the active zone measurements into simple terminals and glomerular terminals to determine whether a zonal type was associated preferentially with either of these synaptic terminal types. RESULTS Active zones appear as straight or curved profiles in electron micrographs (Fig. 1A). Although the degree of curvature varies, the zones always curve such that the surface of the presynaptic terminal is concave. Threedimensional reconstructions reveal subpopulations of the disc-shaped active zones (Fig. 1B). In particular the large majority of active zones are relatively flat or slightly curved (1~‘s of 1.00 to 1.60). There are also a significant number that are very curved with It’s of 1.61 to 6.50 (Figs. 2A and 2B). Very curved active zones make up 4% of the total disc population; 60% of the very curved active zones occur in laminae Iii and III. In normal rats, 28% of the total active synaptic zones were on glomeruli, but only two of these active zones were highly curved. In the highly curved active zone population, 30% are found on dendritic spines. Following dorsal rhizotomy, the major changes are a significant reduction in the numbers of highly curved synaptic zones (P < 0.05, Fisher’s exact test) and a significant reduction in the number of glomeruli (P < 0.05, Fisher’s exact test). Very curved synaptic zones make up only 1% of the total active zone population on the rhizotomized side and no very curved synapses are found in laminae Iii and III (Figs. 2C and 2D). Of the total synaptic zone population in the rhizotomized animals, only 9% are associated with glomeruli and none of these are highly curved. DISCUSSION The typical “chemical” synapse consists of a presynaptic terminal, a postsynaptic element, and one or more active synaptic zones which are the membrane and cytoplasmic specializations that indicate where actual synaptic transmission occurs, Most studies on the morphology of the synapse focus primarily on the presynaptic terminal (1, 7). Those studies that consider the active zones deal with important questions as to whether any of the active zones are perforated or with advanced techniques that reveal previously unknown facets of the organization of the cytoplasmic thickenings that make up the active zones (6, 13, 14, 19). Our data indicate that active zones in the dorsal horn are disc-shaped and not perforated. The extent of curvature of the zones has received less attention, presumably because such data de-
AND
HULSEBOSCH
pend on reconstructions and also because no obvious correlations of active zone curvature with other structural or functional variables were apparent. The present study deals with the extent of curvature of the active zones in rat dorsal horn and correlates the pattern in the normal with that seen after dorsal rhizotomy. An index of curvature was devised to facilitate the comparisons. Computer-aided reconstructions of the active zones made the actual shapes of the active zones less laborious to attain. In normal dorsal horns, the synaptic zones fall into two categories on the basis of curvature: (1) the large majority are flat or slightly curved and (2) a smaller number are very curved. The second population, the very curved zones, is basically lost following dorsal rhizotomy. If the reasonable assumption is made that dorsal rhizotomy removes the primary afferent fibers ipsilaterally, the conclusion is that the very curved active zones are associated primarily with primary afferent synapses. Although the glomeruli that characterize laminae IIII of the mammalian dorsal horn are a major focus of previous studies on himary afferent innervation of the dorsal horn (3,16,17,18,20), the very curved synapses are not specifically associated with these structures. Instead they are associated primarily with dendritic spines and simple axodendritic synapses. An obvious question that follows from this finding is whether the very curved active zones are involved in transmission of a type of information different than that of the more numerous flat or slightly curved active zones. Since the very curved active zones seldom occur on central elements of glomeruli, which are often labeled for substances that characterize small dorsal root ganglion cells (2,4,5, lo), it might be suggested that they are associated with presynaptic terminals that arise from large cells that give rise to rapidly conducting myelinated primary afferent fibers. If this is the case, an experiment of interest will be to determine the numbers and proportions of very curved active zones in rats that received capsaicin at birth since these animals preferentially lose fine primary afferents. It is also of interest that the very curved terminals are associated with small dendritic spines or dendritic “bumps,” a finding which presumably has implications for the integration of information passed to the postsynaptic cells. ACKNOWLEDGMENTS This research was supported by NIH Grants NS 11255, NS 10161, RR 03979, and NS 01217; the Bristol Myers Co.; and the Florence and Marie Hall Endowment for Excellence in Medical Education.
REFERENCES 1.
AKERT, K., K. PFENNINGER, C. SANDRI, AND H. MOOR. 1972. Freeze etching and cytochemistry of vesicles and membrane complexes in synapses of the central nervous system. In Struc-
SYNAPTIC
SHAPE
ANALYSIS
ture and Function of Synapses (G. D. Pappas and D. P. Purpura, Eds.), pp. 67-86. Raven Press, New York. 2. CARLTON, S. M., D. L. MCNEILL, K. CHUNG, AND R. E. COGGESHALL. 1987. A light and electron microscopic level analysis of calcitonin gene-related peptide (CGRP) in the spinal cord of the primate: An immunohistochemical study. Neurosci. Lett. 82: 145-150. 3. COIMBRA, A., A. EIBEIRO-DA-SILVA, AND D. PIGNATELLI. 1984. Effects of dorsal rhizotomy on the several types of primary afferent terminals in laminae I-III of the rat spinal cord. Anat. Embryol.
170:
279-287.
4. COIMBRA, A., B. P. SODRE-BORGES, AND M. M. MAGALHAES. 1974. The substantia gelatinosa Rolandi of the rat. Fine structure, cytochemistry (acid phosphatase) and changes after dorsal root section. J. Neurocytol. 3: 199-217. 5. CSILLIK, B., AND E. KNYIHAR-CSILLIK. 1986. The Protean Gate, Structure
and
Plasticity
of the Primary
Nociceptive
Analyzer.
Akademiai Kiado, Budapest. 6. GEINISMAN, Y., F. MORRELL, AND L. DE TOLEDO-MORRELL. 1987. Axospinous synapses with segmented postsynaptic densities: A morphologically distinct synaptic subtype contributing to the number of profiles of ‘perforated’ synapses visualized in random sections. Brain Res. 423: 179-188. HENSON, J. E., T. S. REESE, M. J. DENNIS, Y. JAN, L. JAN, AND L. EVANS. 1979. Synaptic vesicle exocytosis captured by quick freezing and correlated with quanta1 transmitter release. J. Cell Biol. 81:275-300. HUIJSMANS, D. P., W. H. LAMERS, J. A. Los, AND J. STRACKEE. 1986. Toward computerized morphometric facilities: An aided review of 58 software packages for computer-aided three-dimensional reconstruction, quantification, and picture generation from parallel serial sections. Anat. Rec. 216: 449-470. KINNAMON, J. C., T. A. SHERMAN, AND S. D. ROPER. 1988. Ultrastructure of mouse vallate taste buds. III. Patterns of synaptic connectivity. J. Comp. Neural. 270: l-10.
IN RAT SPINAL
CORD
155
10. KNYIHAR-CSILLIK, AND B. CSILLIK. 1985. FRAP: Histochemistry of the Primary Nociceptive Neuron. Fisches Verlag, Stuttgart/ New York. 11. MAYHEW, T. M. 1979. Stereological approach to the study of synapse morphometry with particular regard to estimating number in a volume and on a surface. J. Neurocytol. 8: 121-138. 12. MCNEILL, D. L., K. CHUNG, C. E. HULSEBOSCH, R. P. BoLENDER, AND R. E. COGGESHALL. 1988. Numbers of synapses in laminae I-IV of the rat dorsal horn. J. Comp. Neural. 279: 453460. 13. NIETO-SAMPEDRO, M., S. F. HOFF, AND C. W. COTMAN. 1982. Perforated postsynaptic densities: Probable intermediates in the synapse turnover. Proc. Natl. Acad. Sci. USA 79: 5718-5722. 14. PETERS, A., AND I. R. KAISERMAN-ABRAMOF. 1969. The small pyramidal neuron of the rat cerebral cortex. 2. Zellforsch. 100: 487-506. 15. RALSTON, H. J., III. 1979. The fine structure of laminae I, II, and III of the macaque spinal cord. J. Comp. Neural. 184: 619-642. 16. RALSTON, H. J., III, AND D. D. RALSTON. 1979. The distribution of dorsal root axons in laminae I, II, and II of the macaque spinal cord: A quantitative electron microscope study. J. Comp. Neural. 184:643-684. 17. RALSTON, H. J., III, AND D. D. RALSTON. 1979. The distribution of dorsal root axons to laminae IV, V, and VI of the macaque spinal cord: A quantitative electron microscope study. J. Comp. Neural. 2 12:435-448. 18. RIBEIRO-DA-SILVA, A., AND A. COIMBRA. 1982. Two types of synaptic glomeruli and their distribution in laminae I-III of rat spinal cord. J. Comp. Neural. 209: 176-186. 19. WESTRUM, L. E., AND E. G. GRAY. 1986. New observations on the substance of the active zone of brain synapses and motor endplates. Proc. R. Sot. Lond. (Biol.) 229: 29-38. 20. WILLIS, W. D., AND R. E. COGGESHALL. 1978. Sensory Mechanisms of the Spinal Cord. Plenum, New York/London.
NEUROLOGY
108,151-155
(1990)
Computerized Ultrastructural Analysis of the Shape of the Active Synaptic Zones in Rat Spinal Cord W. TERRELL STAMPS, Marine
Biomedical Institute, and Biophysics,
RICHARD Department University
E. COGGESHALL,
AND CLAIRE
E. HULSEBOSCH’
of Anatomy and Neurosciences and Department of Texas Medical Branch, Galveston, Texas 77550
Active synaptic zones are cytoplasmic specializations that indicate where synaptic transmission occurs. We have used computerized three-dimensional reconstructions from serial ultrathin sections to define certain features of the geometry of these zones in mammalian spinal cord. Our main finding is that the active zones in the dorsal portion of the spinal cord can be placed in one of two categories with respect to curvature: (1) uncurved or slightly curved and (2) very curved. The very curved category is associated with simple axodendritic type synapses in which the axonal terminal arises from pri0 1990 Academic Press, Inc. mary afferent fibers.
INTRODUCTION The dorsal horn of the mammalian spinal cord is an important area for somatosensory integration. Consequently the synaptic architecture of this region is a subject of importance. Synapses in the dorsal horn, as elsewhere, consist of presynaptic terminals, membrane and cytoplasmic specializations which are referred to as active synaptic zones, (11) and postsynaptic structures which are usually a dendrite or a soma (11, 15). The shape of the active zone and, in particular, the amount of curvature are aspects of dorsal horn synaptic architecture that have not been extensively examined. Since the cytoplasmic specializations that form the active zone are the sites where synaptic transmission presumably occurs, the size, shape, and extent of curvature of the active zone should be determined and related to fiber type if possible. To examine this issue, we have devised an index of curvature for the active zones in laminae I-IV of the rat dorsal horn and then used serial reconstructions and computer modeling to facilitate the gathering of the data. Our goal is to describe quantitatively the shapes and sizes, and particularly the exact extent of curvature of the active zones in laminae I-IV of normal animals, to show that a particular class within this ’ To whom correspondence should be addressed at Marine Biomedical Institute, 200 University, Galveston, TX 77550.
of Physiology
grouping is affected following dorsal rhizotomy. This is a step toward correlating the geometry of spinal synapses with meaningful functional data. METHODS Four adult male Sprague-Dawley rats were anesthetized with sodium pentobarbital (Nembutal, 35 mg/kg ip). When anesthesia was deep, a laminectomy was done, and dorsal roots L3-Ca2 were cut unilaterally midway between the ganglion and the spinal cord. Seventy-two hours later, the animals were reanesthetized as above and perfused intraventricularly with 0.9% NaCl containing 200 IU heparin and 0.02% sodium nitrite. When the vascular system was free of blood, the perfusion mixture was changed to 3% glutaraldehyde, 3% formaldehyde, and 0.1% picric acid in 0.1 M, pH 7.4, cacodylate buffer. Following fixation, the S2 spinal segment was removed and divided transversely into three approximately equal slices with a razor blade. The slices were postfixed in 1% osmium tetroxide and 1.5% potassium ferricyanide. After being embedded in plastic, 14-24 serial thin sections of laminae I-IV of each dorsal horn were cut, placed on Formvar-coated grids, and stained with 0.1% lead citrate. The dorsal horn in the first section was photographed and a montage of approximately 1200~ (actual magnifications from calibrated grids) was made and the boundaries of laminae I-IV were drawn as described previously (12). Points on this micrograph were chosen randomly and used as the centers of high power pictures (approximately 40,000X, actual magnifications from calibrated grids) which were taken through the entire set of serial sections. Section thickness was determined by measuring the thickness of a fold in the section and dividing it by 2. To reconstruct the active zones, mitochondria were selected as fiducial marks to align the planes of the reconstruction. Contours of the active zones and the presynaptic and postsynaptic elements were traced from serial sections onto a digitizing tablet and entered into a 80386 personal computer using three-dimensional reconstruction software, version 1.2, from the Laboratory of High Voltage Electron Microscopy of Boulder Colo151 All
Copyright 0 1990 rights of reproduction
0014-4886/90
$3.00
by Academic Press, Inc. in any
form
reserved.
152
STAMPS,
COGGESHALL,
AND
HULSEBOSCH
SYNAPTIC
SHAPE
ANALYSIS
IN
RAT
SPINAL
153
CORD
B 1.6
60
0.8 0.6
1.61-1.71-1.81-1.91-2.011.70 1.60 1.90 2.00
2.10
D
1.8
65
5.416.60
1.6
60 30 5w
25
i+ 2
20 0.8
t
15 10 5 0 l.OO1.10
l.ll1.20
1.211.30
1.311.40
1.411 so
1.511.60
INDEX OF CURVATURE
1.61-1.71-1.81-1.91-2.011.70 1.80 1.90
2.00
2.10
5.416.50
INDEX OF CURVATURE
FIG. 2. Histograms of indices of curvatures for (A) slightly curved active zones and (B) very curved active zones (abscissa) percentage total synapses (ordinate) in the dorsal horn of normal rats. Histograms of similar data in rats after dorsal rhizotomy in C and D. Note that the very curved synapses (D) are essentially eliminated.
rado (8,9). An enhanced graphics monitor displayed the reconstructions. The three-dimensional reconstruction of serial sections compensates for biases due to sampling in a single plane with respect to the random orientation of synapses within the dorsal horn and thus is in accord with stereological principles. A total of 270 active zones were examined in normal rats (n = 4) and 210 in rhizotomized rats (n = 4). Reconstructed active zones were divided into populations using an index of curvature (Ic)
plotted against are displayed
which was determined by dividing the longest boundary trace length of the active zone cross section by the distance between the end points of the zone. This method yielded the smallest standard deviations for our calculations. An Ic of 1 would indicate a flat active zone and a half circle would have an Ic of 1.57. Active zones were considered slightly curved if they had a value of greater than 1 but less than 1.6 and very curved if they had a value of greater than 1.6. In the dorsal horn, distinctions
FIG. 1. (A) An electron micrograph of several vesicle-filled presynaptic terminals. A curved active synaptic zone is present near the center of the picture, between the two arrows. Calibration bar = 0.5 pm. (B) A three-dimensional reconstruction of a presynaptic terminal region (light gray), a postsynaptic dendrite with a small dendritic spine extending toward the top of the picture (medium gray), and the very curved active synaptic zone which is associated with these two structures (dark lines).
154
STAMPS.
COGGESHALL,
are made between simple axodendritic synapses, where one presynaptic element contacts one postsynaptic element, and glomerular terminals, where one presynaptic element contacts two or more postsynaptic elements. For this reason, we separated the active zone measurements into simple terminals and glomerular terminals to determine whether a zonal type was associated preferentially with either of these synaptic terminal types. RESULTS Active zones appear as straight or curved profiles in electron micrographs (Fig. 1A). Although the degree of curvature varies, the zones always curve such that the surface of the presynaptic terminal is concave. Threedimensional reconstructions reveal subpopulations of the disc-shaped active zones (Fig. 1B). In particular the large majority of active zones are relatively flat or slightly curved (1~‘s of 1.00 to 1.60). There are also a significant number that are very curved with It’s of 1.61 to 6.50 (Figs. 2A and 2B). Very curved active zones make up 4% of the total disc population; 60% of the very curved active zones occur in laminae Iii and III. In normal rats, 28% of the total active synaptic zones were on glomeruli, but only two of these active zones were highly curved. In the highly curved active zone population, 30% are found on dendritic spines. Following dorsal rhizotomy, the major changes are a significant reduction in the numbers of highly curved synaptic zones (P < 0.05, Fisher’s exact test) and a significant reduction in the number of glomeruli (P < 0.05, Fisher’s exact test). Very curved synaptic zones make up only 1% of the total active zone population on the rhizotomized side and no very curved synapses are found in laminae Iii and III (Figs. 2C and 2D). Of the total synaptic zone population in the rhizotomized animals, only 9% are associated with glomeruli and none of these are highly curved. DISCUSSION The typical “chemical” synapse consists of a presynaptic terminal, a postsynaptic element, and one or more active synaptic zones which are the membrane and cytoplasmic specializations that indicate where actual synaptic transmission occurs, Most studies on the morphology of the synapse focus primarily on the presynaptic terminal (1, 7). Those studies that consider the active zones deal with important questions as to whether any of the active zones are perforated or with advanced techniques that reveal previously unknown facets of the organization of the cytoplasmic thickenings that make up the active zones (6, 13, 14, 19). Our data indicate that active zones in the dorsal horn are disc-shaped and not perforated. The extent of curvature of the zones has received less attention, presumably because such data de-
AND
HULSEBOSCH
pend on reconstructions and also because no obvious correlations of active zone curvature with other structural or functional variables were apparent. The present study deals with the extent of curvature of the active zones in rat dorsal horn and correlates the pattern in the normal with that seen after dorsal rhizotomy. An index of curvature was devised to facilitate the comparisons. Computer-aided reconstructions of the active zones made the actual shapes of the active zones less laborious to attain. In normal dorsal horns, the synaptic zones fall into two categories on the basis of curvature: (1) the large majority are flat or slightly curved and (2) a smaller number are very curved. The second population, the very curved zones, is basically lost following dorsal rhizotomy. If the reasonable assumption is made that dorsal rhizotomy removes the primary afferent fibers ipsilaterally, the conclusion is that the very curved active zones are associated primarily with primary afferent synapses. Although the glomeruli that characterize laminae IIII of the mammalian dorsal horn are a major focus of previous studies on himary afferent innervation of the dorsal horn (3,16,17,18,20), the very curved synapses are not specifically associated with these structures. Instead they are associated primarily with dendritic spines and simple axodendritic synapses. An obvious question that follows from this finding is whether the very curved active zones are involved in transmission of a type of information different than that of the more numerous flat or slightly curved active zones. Since the very curved active zones seldom occur on central elements of glomeruli, which are often labeled for substances that characterize small dorsal root ganglion cells (2,4,5, lo), it might be suggested that they are associated with presynaptic terminals that arise from large cells that give rise to rapidly conducting myelinated primary afferent fibers. If this is the case, an experiment of interest will be to determine the numbers and proportions of very curved active zones in rats that received capsaicin at birth since these animals preferentially lose fine primary afferents. It is also of interest that the very curved terminals are associated with small dendritic spines or dendritic “bumps,” a finding which presumably has implications for the integration of information passed to the postsynaptic cells. ACKNOWLEDGMENTS This research was supported by NIH Grants NS 11255, NS 10161, RR 03979, and NS 01217; the Bristol Myers Co.; and the Florence and Marie Hall Endowment for Excellence in Medical Education.
REFERENCES 1.
AKERT, K., K. PFENNINGER, C. SANDRI, AND H. MOOR. 1972. Freeze etching and cytochemistry of vesicles and membrane complexes in synapses of the central nervous system. In Struc-
SYNAPTIC
SHAPE
ANALYSIS
ture and Function of Synapses (G. D. Pappas and D. P. Purpura, Eds.), pp. 67-86. Raven Press, New York. 2. CARLTON, S. M., D. L. MCNEILL, K. CHUNG, AND R. E. COGGESHALL. 1987. A light and electron microscopic level analysis of calcitonin gene-related peptide (CGRP) in the spinal cord of the primate: An immunohistochemical study. Neurosci. Lett. 82: 145-150. 3. COIMBRA, A., A. EIBEIRO-DA-SILVA, AND D. PIGNATELLI. 1984. Effects of dorsal rhizotomy on the several types of primary afferent terminals in laminae I-III of the rat spinal cord. Anat. Embryol.
170:
279-287.
4. COIMBRA, A., B. P. SODRE-BORGES, AND M. M. MAGALHAES. 1974. The substantia gelatinosa Rolandi of the rat. Fine structure, cytochemistry (acid phosphatase) and changes after dorsal root section. J. Neurocytol. 3: 199-217. 5. CSILLIK, B., AND E. KNYIHAR-CSILLIK. 1986. The Protean Gate, Structure
and
Plasticity
of the Primary
Nociceptive
Analyzer.
Akademiai Kiado, Budapest. 6. GEINISMAN, Y., F. MORRELL, AND L. DE TOLEDO-MORRELL. 1987. Axospinous synapses with segmented postsynaptic densities: A morphologically distinct synaptic subtype contributing to the number of profiles of ‘perforated’ synapses visualized in random sections. Brain Res. 423: 179-188. HENSON, J. E., T. S. REESE, M. J. DENNIS, Y. JAN, L. JAN, AND L. EVANS. 1979. Synaptic vesicle exocytosis captured by quick freezing and correlated with quanta1 transmitter release. J. Cell Biol. 81:275-300. HUIJSMANS, D. P., W. H. LAMERS, J. A. Los, AND J. STRACKEE. 1986. Toward computerized morphometric facilities: An aided review of 58 software packages for computer-aided three-dimensional reconstruction, quantification, and picture generation from parallel serial sections. Anat. Rec. 216: 449-470. KINNAMON, J. C., T. A. SHERMAN, AND S. D. ROPER. 1988. Ultrastructure of mouse vallate taste buds. III. Patterns of synaptic connectivity. J. Comp. Neural. 270: l-10.
IN RAT SPINAL
CORD
155
10. KNYIHAR-CSILLIK, AND B. CSILLIK. 1985. FRAP: Histochemistry of the Primary Nociceptive Neuron. Fisches Verlag, Stuttgart/ New York. 11. MAYHEW, T. M. 1979. Stereological approach to the study of synapse morphometry with particular regard to estimating number in a volume and on a surface. J. Neurocytol. 8: 121-138. 12. MCNEILL, D. L., K. CHUNG, C. E. HULSEBOSCH, R. P. BoLENDER, AND R. E. COGGESHALL. 1988. Numbers of synapses in laminae I-IV of the rat dorsal horn. J. Comp. Neural. 279: 453460. 13. NIETO-SAMPEDRO, M., S. F. HOFF, AND C. W. COTMAN. 1982. Perforated postsynaptic densities: Probable intermediates in the synapse turnover. Proc. Natl. Acad. Sci. USA 79: 5718-5722. 14. PETERS, A., AND I. R. KAISERMAN-ABRAMOF. 1969. The small pyramidal neuron of the rat cerebral cortex. 2. Zellforsch. 100: 487-506. 15. RALSTON, H. J., III. 1979. The fine structure of laminae I, II, and III of the macaque spinal cord. J. Comp. Neural. 184: 619-642. 16. RALSTON, H. J., III, AND D. D. RALSTON. 1979. The distribution of dorsal root axons in laminae I, II, and II of the macaque spinal cord: A quantitative electron microscope study. J. Comp. Neural. 184:643-684. 17. RALSTON, H. J., III, AND D. D. RALSTON. 1979. The distribution of dorsal root axons to laminae IV, V, and VI of the macaque spinal cord: A quantitative electron microscope study. J. Comp. Neural. 2 12:435-448. 18. RIBEIRO-DA-SILVA, A., AND A. COIMBRA. 1982. Two types of synaptic glomeruli and their distribution in laminae I-III of rat spinal cord. J. Comp. Neural. 209: 176-186. 19. WESTRUM, L. E., AND E. G. GRAY. 1986. New observations on the substance of the active zone of brain synapses and motor endplates. Proc. R. Sot. Lond. (Biol.) 229: 29-38. 20. WILLIS, W. D., AND R. E. COGGESHALL. 1978. Sensory Mechanisms of the Spinal Cord. Plenum, New York/London.