A Cytoarchitectonic Analysis of the Spinal Cord of the R O B E R T B. L E O N A R D * AND DAVID H. COHEN D e p a r t m e n t of Physiology, University of V i r g i n i a , School of Medicine, Charlottesville, Virginia 22901

ABSTRACT

The spinal gray of the pigeon is described cytoarchitectonically to establish a foundation for anatomical and physiological studies of the pigeon spinal cord. The material includes segments from the high cervical cord through the lumbosacral enlargement, and nine cellular layers are described. In addition to this laminar organization, various distinct cell groups such as the dorsal magnocellular column, column of Terni, marginal cells and lobes of Lachi are described. Layers I-IV occupy the head of the dorsal horn, are apparent at all spinal levels examined, and represent the clearest case of laminar organization of the spinal gray of the pigeon. Layer V occupies the full extent of the neck of the dorsal horn at all segmental levels investigated. Also, the dorsal magnocellular column is situated in the central region of this layer from the rostra1 pole of the cervical enlargement through the lumbosacral enlargement, and arguments are aavanced that this cell column is homologous to the column of Clarke. In the intermediate zone a Layer VI is defined, but i t is apparent only at the enlargements. Layers VII-IX constitute the ventral horn, Layer IX being the motoneuronal cell groups. With the exception of the motoneuronal groups, the boundaries of the ventral horn layers are considerably less distinct than those of the dorsal horn, and no attempt is made to distinguish Layers VII and VIII at lumbosacral levels. At the enlargements there is a prominent lateral motoneuronal cell group consisting of large cells. It is generally concluded that the cytoarchitectonic organization of the spinal gray of the pigeon bears a rather close resemblance to that described for various mammalian species, particularly with respect to the dorsal horn.

This and the following report on spinal dorsal root projections in the pigeon (Leonard and Cohen, '75) were undertaken to provide a structural foundation for physiological studies of somato-autonomic coupling in the pigeon spinal cord (Leonard and Cohen, in preparation). In this context, our most immediate concern was to delineate the spino-spinal circuitry through which segmental afferents influence the activity of the sympathetic preganglionic neurons in the column of Terni (Macdonald and Cohen, '70; Cabot et al., in preparation). However, the broader motivation derived from our efforts to develop visually conditioned heart rate change in the pigeon as a vertebrate model system for cellular studies of learning (Cohen, '69, '74a). In this model foot-shock serves as the unconditioned stimulus, and part of our analysis has involved (a) identifying the somatosensory pathways transmitting this uncondiJ. COMP.NEUR.,163: 159-180.

tioned stimulus information and (b) describing their coupling to the motor components of the system (Cohen, '74b). This initial report describes the spinal cytoarchitecture, generally to establish a framework for further spinal investigations and more specifically to provide a The research reported here was supported by NSF grants GB-13816X and GB-35204X and a grant from the Benevolent Foundation of Scottish Rite Freemasonary, Northern Jurisdiction, U . S. A. Portions of this material were presented at the 85th meeting of the American Association of Anatomists, Dallas, Texas, April 1972. Requests for reprints should be sent to: Dr. David H. Cohen, Department of Physiology, School of Medicine, University of Virginia, Charlottesville, Virginia 22901. 2 This report i s based in part on a doctoral dissertation submitted to the University of Virginia by R . B. Leonard who was partially supported by a fellowship from the Benevolent Foundation of Scottish Rite Freemasonary, Northern Jurisdiction, U. S. A. and P. H. S. Training Grant HL-05815 from the National Heart and Lung Institute. Present address: Marine Biomedical Institute, University of Texas Medical Branch at Galveston. 3 Partially supported by P. H. S. Research Career Development Award HL-16579 from the National Heart and Lung Institute.

159

160

ROBERT B. LEONARD AND DAVID H. COHEN

basis for describing the results of the subsequent experiments on spinal terminal fields of dorsal root fibers. Regarding earlier descriptions of the avian spinal cord, the older comparative neuroanatomical literature contains scattered reports for various species. The most extensive for the pigeon is that of Huber ('36), and that has constituted the basis for most discussions of the avian spinal gray (e.g., Ariens-Kappers et al., '36; Nieuwenhuys, '64). In these descriptions only the distinct columns of large cells were discussed, such as the dorsal magnocellular column at the base of the dorsal horn and the motoneuronal cell groups. Although a substania gelatinosa capping the dorsal horn was noted, the greater part of the spinal gray was described as containing scattered cells without apparent grouping. Since the inception of the experiments presented here, two germane publications have appeared. Van den Akker ('70) has analyzed the pigeon spinal cord, and his monograph includes a short section on cytoarchitecture. Although he does not propose a laminar scheme in the sense of Rexeds ('52) description of the cat spinal cord, his delineation of seven zones clearly reflects the influence of the laminar approach. Brinkman and Martin ('73) have also published a cytoarchitectonic analysis of the chick brachial spinal cord, and in contrast to van den Akker ('70) they identify nine discrete laminae, including a subdivision of the dorsal horn into six laminae. Our interpretation of the spinal gray of the pigeon differs substantially from those of both van den Akker ('70) and Brinkman and Martin ('73), and a detailed discussion of these differences follows the presentation of our material. MATERIALS AND METHODS

The present analysis is based on material gathered from many cases over a period of several years, and it includes segments from upper cervical, brachial, thoracic, lumbar and upper sacral spinal cord. In all instances the material was obtained from White Carneaux pigeons (Columba Zivia) weighing 350-650 g and ranging in age from 2-6 months. A complete series of representative segments from two spinal cords has been prepared in 15, 20, 40 and 60 p sections in the transverse plane and 20 and

40 p sections in the sagittal plane. Sections were stained with either cresylechtviolet or a modified Kluver-Barrera ('53) procedure in which Lux01 Fast Blue MBS was used in the concentration suggested by Szabo ('65) with the differentiation and Nissl staining procedure of Kluver-Barrera ('53). Segments representing both enlargements were also prepared in 100 p transverse sections and stained with cresylechtviolet. Moreover, several lumbar segments were prepared with a reduced silver method, and the experimental material for characterizing dorsal root terminations (Leonard and Cohen, '75) provided additional cases with sections stained with cresylechtviolet, Fink-Heimer procedure I (Fink-Heimer, '67), or the Ebbesson-Heimer procedure (Ebbesson, '70). In all cases the birds were sacrificed with an overdose of sodium pentobarbital and perfused through the left ventricle with avian saline followed by 10% formalin. The vertebral columns were removed, and the cords were exposed and allowed further immersion fixation in 10% formalin. Normal material was embedded in paraffin for sectioning at 15 or 20 p and in celloidon for sectioning at 40, 60 or 100 p. Experimental material was embedded in albumin and sectioned frozen. All cell measurements were made from 40 p sections embedded in celloidon. An ocular reticle calibrated with a stage micrometer was used at magnifications of 400 or 1,000 with no correction for shrinkage. RESULTS

Huber's ('36) macroscopic description of the pigeon spinal cord was confirmed in this study and will not be repeated in detail. Briefly, the spinal cord extends almost the entire length of vertebral canal, and according to Huber ('36) there are 14 cervical, 6 thoracic, 4 lumbar, and 15 sacro-coccygeal segments. The brachial and lumbrosacral intumescenses are distinct, and the lumbosacral enlargement is characterized by a unique separation of the dorsal funiculi to form the rhomboid sinus (fig. 1) containing a gelatinous structure composed of densely packed, glycogen-rich cells (e.g., Matulinois, '72). Microscopically, nine cellular regions can be seen extending through the gray in an approximately laminar arrangement.

SPINAL CYTOARCHITECTURE IN THE PIGEON

161

Fig. 1 Dissection of the lumbosacral enlargement and plexus. Dorsal root ganglia 21 (top) through 27 (bottom) are exposed. The sciatic nerve is at the lower right. Magnification x 3.5.

These are referred to as layers to suggest their analogy to Rexed's ('52, '54, '64) laminar description of the cat spinal cord. In the pigeon this laminar pattern is most distinct in the head of the dorsal horn, less so in its neck, and least prominent in the more ventral regions of the gray. The detailed cytoarchitecture of the cervical enlargement will be described first, followed by that of the lumbosacral enlargement and finally by summaries of the differences found at other spinal levels.

Cervical enlargement A representative section from segment 12 of the cervical enlargement is shown in figure 2 A (see Huber "361 for segmental nomenclature), and our proposed cytoarchitectonic boundaries at this level are outlined in figure 2B. Layer I caps the dorsal horn, and both its medial and lateral aspects extend ventrally (fig. 3). Groups of neurons sometimes protrude slightly into

the white matter, such that the Layer I boundaries are indistinct at certain points. These neurons are frequently fusiform in shape (fig. 4A), particularly along the lateral aspects of the layer. In contrast, the cells along its dorsal aspect tend to be larger and triangular in shape (fig. 4B). Small round cells are also numerous in Layer I (fig. 4C), and these generally measure 5-8 g in diameter; the large, dorsally situated cells are 15-18 CL along their longest dimension (fig. 4B), while the long axis of the fusiform neurons may reach 50 fi (fig. 4A). The small neurons have relatively clear nuclei with distinct nucleoli, and their cytoplasm is restricted to a narrow perinuclear region with the Nissl substance concentrated close to the nuclear membrane. The larger cells are richer in cytoplasm with more distributed Nissl bodies that appear as coarse granules in the larger elongated neurons. Layer I1 follows the contour of Layer I

162

ROBERT B. LEONARD A N D DAVID H . C O H E N

SPINAL CYTOARCHITECTURE I N THE PIGEON

163

Fig. 3 Transverse section (40 p ) showing Layers 1-111 of the head of the dorsal horn in the cervical enlargement (segment 12). DF, Dorsal funiculus. Cresylechtviolet stain. Calibration: 50 w .

(figs. 2, 3 ) , extending ventrally along both the medial and lateral aspects of the spinal gray. The medial aspect of the layer is penetrated by fascicles of fibers from the dorsal funiculus (fig. 5). The packing density is not homogeneous across the mediolateral extent of the layer, and several cellular aggregates are apparent (fig. 3 ) . In transverse section the cells appear round with little cytoplasm. Many scattered neurons of approximately 13 p diameters are present, but most cells are smaller and measure 6 7 p . The Nissl substance is sparse and encircles the large clear nucleus. At the medial border with Layer I the cells tend to be slightly larger with more distinct Nissl granules. Layer 111 parallels Layer I1 (figs. 2 , 3 ) and is broader dorsoventrally than either Layers I or I1 at the center of the dorsal horn. Its ventral border, both medially and laterally, is indistinct where it merges with the ventral aspect of Layer 11; centrally this border is more distinct. Packing density is highest centrally where the cells tend to be largest, and both packing density and cell size de-

crease medially and even more so laterally. Most of the centrally situated neurons are somewhat elongated in a dorsolateral orientation with their nuclei slightly displaced toward one pole. These cells are maximally 18-20 p in their longest dimension and 10 p in the other. Medially, but especially laterally, they tend to appear rounder and only slightly larger than the cells of Layer 11, although in other respects they generally resemble Layer I1 neurons. Layer IV is the most distinct area of the head of the dorsal horn by virtue of its larger cells with their prominent aggregation of Nissl substance (fig. 6A). This layer fills the concavity formed by Layers 1-111 (fig. 2 ) , and laterally it does not reach the funiculus but borders Layers I1 and 111. Medially the border is less distinct where Layer IV and the ventral extents of Layers I1 and 111 abut the dorsal funiculus and are penetrated by entering fiber bundles (fig. 5). The ventrolateral border of this layer, though recognizable, is less distinct than the ventromedial, since large cells are also

164

ROBERT B. LEONARD AND DAVID H . COHEN

Fig. 4 Examples of three neuronal types from Layer I . Panel A : Large fusiform cell from the lateral aspect of Layer I. Panel B : Triangular cell from the dorsal aspect of Layer I. Panel C : Cell type commonly found throughout Layer I. All three cells are from 40 k sections stained with cresylechtviolet. Calibration: 10 k.

present laterally in Layer V . Though small neurons are found in Layer IV, larger polygonal or triangular cells predominante (fig. 6B). These larger neurons are 2 5 3 0 X 15 p , while the more elongated cells often have a long axis of 40-45 k . The nuclei are generally centrally located, though having

a slight tendency toward eccentricity, and the cytoplasm contains numerous aggregates of Nissl substance (figs. 6 A , B ) . The less numerous small cells resemble those of Layer I11 but are slightly larger. Thus, four distinct cellular zones can be recognized in the head of the dorsal horn.

SPINAL CYTOARCHITECTURE I N T H E PIGEON

165

Fig. 5 The pattern of fiber entry into the dorsal horn shown in a transverse section (15 p.) from segment 13. Modified Kluver-Barrera stain. Calibration: 100 b .

The first three are clearly laminar and are concave ventrally with their medial and lateral aspects following the contour of the horn. The fourth area is less obviously laminar, with its dorsal portion filling the concavity formed by Layers 1-111 and its ventral portion extending mediolaterally across most of the gray. Cell size increases from the first through fourth layers, save for the long fusiform cells of Layer I. The differentiation of the neck of the dorsal horn is less distinct than that of layers I-IV, since gradations in cell size and geometry occur in both mediolateral and dorsoventral directions. In addition, a prominent column of large cells occupies the central region from the base of the dorsal horn and extending dorsally through twothirds of the neck (figs. 2, 7). The extent of

this dorsal magnocellular column will be discussed in more detail below. Layer V is a relatively broad zone in the neck of the dorsal horn (figs. 2, 7). Its dorsomedial aspect borders the dorsal funiculus, and its ventral portion extends to the midline. Many of the constituent neurons, particularly dorsomedially, are fusiform in shape, the larger ones measuring 18-25 X 10-12 w . Smaller cells are also intermingled varying from slightly fusiform to almost round in conformation with diameters less than 18 p . All of these cells are characterized by large nuclei and relatively scant cytoplasm. The Nissl substance of the smaller cells tends to concentrate close to the nuclei with a finer distribution throughout the remaining cytoplasm; in contrast, the Nissl substance of the larger cells distrib-

166

ROBEKT B. LEONARD AND DAVID H. COHEN

Fig. 6 Panel A : Neuronal field from Layer IV at the cervical enlargement (segment 12). Transverse section (40 p ) stained with cresylechtviolet. Panel B . A cell from Layer IV shown at higher magnification. Calibration: 10 p .

Utes homogeneously throughout the cytoplasm. Laterally, Layer V neurons are larger and more varied in geometry, most measuring 20 p or more along their smallest dimension. Ventrolaterally, many of the cells are 30-35 p , though smaller cells are

also scattered throughout the area. The Nissl substance of these cells, both large and small, is seen as discrete clumps throughout the cytoplasm and presents a more intense picture with cresylechtviolet than that observed in more medial cells.

SPINAL CYTOARCHITECTURE I N THE PIGEON

167

Fig. 7 Transverse section (40 p ) showing the spinal gray of the cervical enlargement (segment 1 2 ) ventral to Layer 1%’. DF, Dorsal funiculus; LF, Lateral funiculus; VF, Ventral funiculus. Cresylechtviolet stain. Calibration: 100 p .

Layer VI extends across the entire spinal gray from the midline, slightly dorsal to the central canal, to the lateral funiculus (figs. 2 , 7). Its cytoarchitecture resembles that of Layer V, and the border between the two is frequently indistinct. The medial cells are

larger than those of Layer V, often 25-30 p , and they tend to be round or polygonal rather than fusiform in conformation. Laterally, the packing density diminishes; however, the cells are larger, 3G-35 p , and small cells like those of lateral Layer V are

168

ROBERT B . I.EONARD A N D DAVID H. CO H EN

Figure 8

S P I N A L CYTOARCHITECTURE I N THE PI(;EON

less numerous. Although the Nissl substance is more abundant in Layer VI as compared to Layer V neurons, a similar mediolateral gradient in coarseness and basophilia is observed. The zona intermedia and ventral horn present the same delineation problem as do the neck and base of the dorsal horn; namely, distinct borders are difficult to identify because of the continuous and gradual shift in cell size and packing density. In certain respects the most apparent difference is in the mediolateral dimension. Ventromedial to Layer VI cells become slightly smaller, and the variability of cell shape increases; this region is defined as Layer VIII (figs. 2, 7 ) . Dorsally in Layer VIII the neurons are 15-18 p , but ventrally larger cells reaching 30 p in diameter appear before the medial motoneurons are encountered. Smaller cell are present most ventrally and are more numerous medially than laterally. Layer VII lies lateral to Layer VIII and extends to the ventral funiculus between the medial and lateral motoneuronal cell columns (figs. 2, 7). Layer VII neurons are larger than those of Layer VIII, many exceeding 30 p , and they are polygonal or slightly elliptical in shape. Smaller, round cells are also present, measuring 15 p , but these are not numerous. Neurons of both Layers VII and VIII have similar distributions of Nissl substance which appears as clumps throughout the cytoplasm of the larger neurons, while concentrating closer to the nucleus in the smaller cells. Motoneurons are evident medially and in the lateral extension of the ventral horn (figs. 2 , 7), and both of these neuronal groups are included in Layer IX. Subdivisions are obvious in the lateral group, while medially only a single group is evident. The cells of the lateral group have a n appearance typically associated with motoneurons; namely, they are large, generally pyramidal or polygonal in transverse section, and measure 40 p or more. The medial motoneurons are smaller, 30 p being the upper limit of their size distribution, and smaller cells of approximately 20 p are present though not numerous. Fig. 8 Panel A shows a representative transverse section (40 p ) from the lumbosacral enlargement (segment 23) of the pigeon. Calibration: 100 p. Panel B shows a high contrast illustration of the same section as in Panel A with the cytoarchitectonic regions outlines. CIC, Clarke’s column; LL. lobe of Lachi. Cresylechtviolet stain.

169

Briefly reviewing the intermediate zone and ventral horn of the brachial cord, precise delimitation of cell groups is difficult with the exception of the motoneuronal columns. Two regions are identified in addition to the motoneurons, the laterally situated Layer VII and the medially situated Layer VIII having smaller cells than VII. The dorsal boundary of Layer VII is not sharp and merges with the lateral aspect of Layer VI, both layers containing similar neurons. Medially, the dorsal boundary of Layer VIII is easier to define, as its neurons are slightly smaller than those of the overlying Layer VI. Neurons of Layer VIII become larger ventrally, approaching the size of the medial motoneurons.

Lumbosncrnl enlargement The cytoarchitectonic organization at the lumbosacral enlargement is similar to that described above for the cervical enlargement (fig. 8). Four regions are identifiable in the head of the dorsal horn, the first three forming laminae that follow the contour of the spinal gray (fig. 9). As in the cervical enlargement, Layer IV fills the concavity formed by Layers 1-111, while its ventral boundary extends across most of the gray. Layers 1-111 extent a considerable distance ventrally along the medial and lateral aspects of the dorsal horn, and the ventral extensions of Layers I1 and I11 are indistinct from the medial and lateral borders of Layer IV. The neurons of each layer resemble those of homologous layers in the cervical enlargement, with the exception of Layer IV where cells are somewhat smaller. Layer V extends across the neck of the dorsal horn (fig. 8), and its ventral boundary extends to the rhomboid sinus separating the gray matter of each side at lumbosacral levels. As in the brachial cord, the laterally situated neurons are larger than those along the medial boundary. The dorsal magnocellular column is also evident throughout the lumbosacral enlargement; however, i t is located somewhat more dorsally than at brachial levels. Moreover, it rarely extends into Layer VI, and it separates the medial and lateral aspects of Layer V. Layer VI is relatively thin, extending across the gray ventral to Layer V (fig. 8). Although its lateral neurons are larger, those located more centrally have coarser, more intensely staining clumps of Nissl

170

ROBERT B . LEONARD AND DAVID H. COHEN

Fig. 9 Transverse section (40 p ) showing Layers I-IV of the head of the dorsal horn in the lumbosacral enlargement (segment 23). DF, Dorsal funiculus; LF, Lateral funiculus. Cresylechtviolet stain. Calibration: 50 p .

substance and dominate the layer in many mainder of the ventral horn contains a heterogeneous neuronal population that is sections. Motoneuronal cell columns are easily more difficult to subdivide than at the ceridentified in the ventral horn and are la- vical enlargement. Therefore, it seems most belled as Layer IX (fig. 8). However, the re- reasonable to denote this area tentatively

SPINAL CYTOARCHITECTURE IN THE PIGEON

as Layer VII-VIII (fig. 8) and not to attempt further subdivision at this time. The majority of cells in this region are fusiform with long axes of 30-33 p , but in addition there is a heterogeneous collection of neurons ranging upward in size from 1 5 p and having various conformations. H i g h cervical Spinal segments rostra1 to the cervical enlargement present a somewhat different cytoarchitectonic picture. The gray is reduced in comparison to the enlargements, and only eight areas can be clearly identified (fig. 10). In addition, the dorsal magnocellular column is no longer present, facilitating subdivision of the neck and base of the dorsal horn. The border between Layers I and I1 is less clear than at the cervical enlargement; however, the border between Layers I1 and I11 is distinct centrally. Layers 1-111 curve ventrally around the head of the dorsal horn as described previously, and Layer IV occupies the concavity (fig. 10). Layer V fills the neck of the horn, as the dorsal magnocellular column is absent. Neurons resembling those of Layer VI at the enlargements are not evident at these levels, and Layer VII is present in the intermediate zone. Only one cell group other than the motoneurons can be identified in the ventral horn, and this extends across the entire gray between the motoneurons and Layer VII, its boundary sloping ventrolaterally. The neurons of this layer resemble those of the medial ventral horn at the enlargements, and thus this region is denoted as Layer VIII (fig. 10).

171

midline and is denoted the column of Terni (fig. 11). A distinct Layer VI is not identifiable at thoracic levels, but lateral to the column of Terni Layer VII occupies the intermediate zone, extending into the ventral horn (fig. 11). Layer VIII then extends ventrally from Layer VII to the motoneuronal cell group, Layer IX, only one such group being present at thoracic levels.

Marginal cells An unusual column of neurons is located at the extreme lateral borders of the spinal cord from cervical through thoracic segments and is referred to as Hoffman's minor nuclei in the older literature (e.g., Huber, '36). In lumbar segments these neurons form distinct protuberances from the lateral funiculi which are designated the accessory lobes of Lachi (Huber, '36) (fig. 8). The column is larger between root entry zones, and no cells are seen on many sections through these zones. In the lumbosacral enlargement there are also numerous neurons scattered throughout the ventrolateral funiculus between the ventral horn and the lobes of Lachi; these are the paragriseal cells (fig. 8). They are not seen in other spinal segments, and it is unclear as to whether they constitute a distinct group or are associated with the lobes of Lachi. The neurons of the lobes of Lachi differ somewhat from the marginal cells of nonlumbar segments (fig. 12), having more cytoplasm with finely clumped, widely distributed Nissl granules. In other segments, the marginal cells are more irregular, and the aggregates of Nissl substance are larger. However, the scattered paragriseal cells of Thoracic the lumbar spinal cord resemble the marThe head of the dorsal horn is more dis- ginal cells of other segments. tinctly rounded in thoracic segments as Column of Terni compared to the enlargements (fig. 11). Layers 1-111 occupy the same positions and The column of Terni can be recognized are of the same relative sizes as at the en- in all thoracic segments and is located imlargements (fig. 11). Layer IV still occupies mediately dorsal and dorsolateral to the the concavity formed by Layers 1-111; how- central canal (fig. 11). Its cells tend to ocever, it is smaller and its lateral aspect ex- cur in small clusters, especially at the rostends more ventrally. Layer V fills most of tral and caudal limits of the column, renthe neck of the dorsal horn and contains dering the precise boundaries difficult to the dorsal magnocellular column (fig. 11). determine. Some cell clusters are seen in Ventrally, the border of Layer V extends the 13-14 intersegment of the cervical ento the midline just dorsal to a distinct cell largement, but the column does not extend column which is present in thoracic and into the thirteenth segment proper. Cauupper lumbar segments. This column oc- dally a similar situation obtains; some cells cupies a semicircular area straddling the can be observed below nerve 2 1 , the first

172

ROBERT €3.

I.EONARD A N D DAVID H. C O H E N

174

ROBERT B. L.EONARD A N D DAVID H. COHEN

Fig. 12 Marginal cells from the lateral borders of the spinal cord Panel A : Cell from Hoffman's minor nucleus in the high cervical spinal cord. Panel B: Cells from the lobe of Lachi i n the lumbosacral enlargement. From transverse section (40 p ) stained with cresylechtviolet. Calibration: 10 p .

segment contributing to the lumbosacral plexus, but the column does not extend into segment 22.

Dorsal magnocellular c o l u m n The dorsal magnocellular column, located in the neck and base of the dorsal horn, extends rostrocaudally from just above the cervical enlargement through most of the lumbosacral enlargement (figs. 2, 8, 11). At the enlargements the cells are large, 40-50 p (fig. 13), and the column extends into the base of the dorsal horn (figs. 7, 8). In thoracic segments the cells are somewhat smaller, 30-35 p , but otherwise similar, and the column tends to be restricted to the neck of the dorsal horn (fig. 11). DISCUSSION

Our findings suggest that on a cytoarchitectonic basis the spinal gray of the pigeon can be differentiated into nine layers. It must be emphasized, however, that while some layers are clear and consistently observable, others are less so, and their delineation is somewhat arbitrary in certain sections. In general, the head of the dorsal horn is the most easily resolved area of the

gray and, with certain exceptions, the more ventral areas are the least clear.

Head of t h e dorsal horn In the present material four laminae are clearly evident in the head of the dorsal horn. While Huber ('36) states that a substantia gelatinosa caps the horn, he provides neither precise delineation of this area nor further subdivision. In contrast, van den Akker ('70) delineates three regions, designated Areas 1-3, and their organization resembles that described here for the head of the dorsal horn. In this regard, our Layer I closely corresponds to van den Akker's ('70) Area 1, described as a thin cell band along the dorsal margin of the horn. However, in our material cells characteristic of Layer I were also seen extending ventrally along the medial aspect of the gray. Within van den Akker's ('70) Area 2 he distinguishes its dorsal aspect as containing smaller, more closely packed cells, but he does not formally separate the dorsal and ventral regions. In the present study, in addition to this dorsoventral cell size difference, we observed that the two regions have a distinct laminar appear-

SPINAL CYTOARCHITECTURE IN THE PIGEON

175

Fig. 13 Neurons from the dorsal magnocellular (Clarke's) column in the pigeon. From a transverse section (40 p ) through segment 12. Cresylechtviolet stain. Calibration: 10 p.

ance in thicker sections (40 p or greater), and this is also apparent upon close inspection of the Kluver-Barrera stained sections shown by van den Akker ('70). Thus, we have separated Area 2 into Layers I1 and 111, both turning ventrally and paralleling Layer I across the head of the dorsal horn. Layer I V of the present study corresponds in the main to van den Akker's ('70) Area 3, most of which lies within the concavity formed by the more dorsal layers. However, we indicate the ventral border of Layer IV as extending further laterally than Area 3, though this lateral border is not always distinguishable from Layer 111. In the chick Brinkman and Martin ('73) also describe four areas in the head of the dorsal horn. However, their organization differs considerably from that described here and by van den Akker ('70). Since the discrepancy seemed sufficiently large to make species difference a n unlikely explanation, the brachial cord of a 3-day old chick was prepared and stained with the modified Kluver-Barrera procedure and then cytoarchitectonically analyzed. Certain gross differences between the pigeon and chick spinal gray are clearly apparent (fig. 14). In the chick there is a n increased

lateral curvature, and the head of the dorsal horn is larger than in the pigeon. In addition, while most dorsal root fibers enter the pigeon gray on the medial aspect of the horn, in the chick they enter more dorsally and appear to divide the head of the dorsal horn into medial and lateral zones. Brinkman and Martin ('73) focus on this apparent division and describe Laminae 1 and 2 a s situated lateral to the fiber entry zone and Lamina 3 medial to it. However, a plausible alternative is illustrated in figure 14 where the first three layers are shown extending across the region of fiber entry. Lamina 4 is maintained a s described by Brinkman and Martin ('73) with only a slight reduction in its dorsal extent. When viewed in this manner, the cytoarchitecture of the head of the chick dorsal horn is similar to that of the pigeon (cf., figs. 2 and 14). To summarize, the head of the pigeon's dorsal horn contains four distinct cell layers that are recognizable in all spinal segments examined in our study. The similarity of this division to that proposed by Rexed ('52, '54) for the cat is apparent, and this correspondence is particularly striking when compared with the cat cervical spinal

176

ROBERT B . LEONARD AND DAVID H . COHEN

Fig. 14 Transverse section (15 p ) showing t h e head of the dorsal horn in t h e cervical e n largement of t h e chick spinal cord. Suggested cytoarchitectonic boundaries are d r a w n . Modified Kliiver-Barrera stain. Calibration: 50 u.

cord (cf., Rexed ('54 - figs. 2 - 4 ) . A comparison with the cervical spinal cord of the monkey is equally compelling (cf., Shriver et al. "681 -fig. 4).

Neck of the dorsal horn Layer V in our analysis occupies the full extent of the neck of the dorsal horn. Its medial and lateral regions differ, the lateral containing a greater number of larger and irregularly shaped cells; however, this mediolateral transition is gradual. Van den Akker ('70). in contrast, describes only a single large area which extends from the head of the dorsal horn to the motoneuronal cell columns. Although he notes a trend toward larger cells ventrally and laterally, his only subdivision is of the dorsal magnocellular column which is designated Area 5 . Huber ('36) also makes no distinction between the neck of the dorsal horn and more ventral areas, although he too notes the dorsal magnocellular column and describes it a s extending from just rostra1 to the cer-

vical enlargement through the lumbosacral enlargement. Both Huber ('36) and van den Akker ('70) comment on the similarity of this column to the mammalian column of Clarke. However, neither considered it the avian homologue of Clarke's column, principally because it extends beyond thoracic and upper lumbar segments. Moreover, van den Akker ('70) considers its dorsal position as further cause for reservation. However, Huber ('36) noted that a greater proportion of the gray is dorsal to the central canal in birds as compared to man. Since this also appears the case in comparing the pigeon and cat, van den Akker's ('70) reservation with respect to the dorsal position of the column seems less serious. Despite the longitudinal extent of the magnocellular column, we suggest its homology to Clarke's column for several reasons. First, notwithstanding Rexed's ('52) assertion that Clarke's column always lies in Lamina VII, its general position relative to the central canal resembles that of

SPINAL CYTOARCHITECTURE I N T H E PIGEON

Clarke's column. Second, it forms a distinct column extending through thoracic and upper lumbar segments. Third, lesions of the dorsolateral funiculus or spinal hemisection produce retrograde degenerative change in the ipsilateral dorsal magnocellular neurons (van den Akker, '70). Fourth, as described in the subsequent paper (Leonard and Cohen, '75), sectioning lumbosacral or cervical dorsal roots yields dense terminal degeneration in the column which extends into the thoracic cord, similar to observations in the cat (Grant and Rexed, '58; Sprague, '58; Sprague and Ha, '64). Finally, Karten ('63, personal communication) has provided evidence in the pigeon for a fiber tract of spinal origin that enters the cerebellum via the restiform body, suggesting the existence of a dorsal spinocerebellar tract in the bird (cf., Oscarsson et al., '63). While these observations do not conclusively establish the homology between the dorsal magnocellular and Clarke's columns, the distinctiveness of the cells, similarity in position, efferent fiber course, and pattern of dorsal root termination support such a view. Therefore, following Rexeds ('52) precedent, this cell column is not designated as a separate layer, in contrast to van den Akker's ('70) proposed nomenclature. Regarding the chick, Brinkman and Martin ('73) describe two areas in the neck of the dorsal horn. Their Lamina 6 clearly includes the dorsal magnocellular column in the cervical enlargement, and they do not recognnize this lamina more rostrally. Furthermore, they describe this lamina as extending across the lateral twothirds of the gray with a Lamina 5 located dorsomedially. However, this description is at variance with the pigeon material, and inspection of the Brinkman and Martin figures (cf., '73 - figs. 3 and 4), as well as analysis of our sections from this region of the chick cord, support the position that the larger cells characteristic of the region are restricted centrally in the neck of the dorsal horn. In summary, the above analysis of the neck of the dorsal horn suggests that it is composed of one rather wide lamina, Layer V, extending rostrocaudally throughout all levels of the cord investigated in our material. In addition, the column of Clarke is identified as occupying the central region

177

of this layer, from the rostra1 pole of the cervical enlargement through the lumbosacral enlargement.

Base of the dorsal horn, intermediate zone and ventral horn These three regions are treated together, as previous investigators have not attempted their subdivision beyond the column of Terni and the motoneuronal cell groups. The area is, perhaps, best described by proceeding from the motoneurons dorsally. In high cervical and thoracic segments a single group of motoneurons, Layer IX, occupies the most ventral part of the gray. Immediately dorsal, a separate Layer VIII can be identified, and Layer VII i s seen extending dorsally to meet Layer V. At thoracic and upper lumbar levels a group of small cells is situated around the midline just dorsal to the central canal. This cell column was originally described by Terni ('23) and has been designated the column of Terni by subsequent investigators (e.g., Huber, '36). The present material differs from Huber's ('36) report only in that the column appears less continuous than suggested in his description. Terni ('23) suggested that these neurons were the sympathetic preganglionic cells, and this has been verified experimentally by retrograde degeneration (Macdonald and Cohen, '70) and retrograde transport of horseradish peroxidase (Cabot et al., in preparation). In the cervical and lumbosacral enlargements the ventral horn extends laterally and contains the large lateral motoneurons. Cells similar to those of thoracic Layer VIII are found concentrated primarily ventromedially, while those resembling Layer VII neurons extend laterally and ventrally, separating the medial and lateral motoneuronal columns. In addition, the intermediate zone contains another neuronal area which is not easily included with Layer V or with Layers VII or VIII; this is defined as Layer VI. At the enlargements the borders between Layers VII and VIII are not distinct, particularly at lumbar levels where no division has been attempted. The dorsal border of Layer VI is fairly clear, as there is an increase in cell size. This pattern resembles that of the cat (Rexed, '52, '54, '64) with the following differences. In the cervical enlargement of the pigeon a medial extension of Layer VII dorsal to Layer VIII is

178

ROBERT B. LEONARD AND DAVID H. COHEN

not clearly observed, and in the lumbar enlargement a distinction between Layers VII and VIII cannot be made. In addition, laterally the border between Layers VI and VII is vague. To summarize these findings, beyond the motoneurons three layers are delineated below the neck of the dorsal horn at cervical and lumbosacral enlargements, while only two are evident elsewhere. With the exception of the motoneuronal cell groups, the boundaries between the various layers in this region are considerably less distinct than those in the dorsal horn. CONCLUSIONS

The cytoarchitectonic analysis of the pigeon spinal cord presented here is similar to that Rexed ('52, '54) describes in the cat. In certain instances the numbering convention established by Rexed was used in preference to that of either van den Akker ('70) or Brinkman and Martin ('73), partly to facilitate comparison with mammalian results. However, there seems little point at present in undertaking more detailed lamina by lamina comparisons with mammalian results until further connectional data are available. Suffice it to say that many of the laminar relations described by Rexed ('52, '54) exist among the layers delineated here, most particularly in the case of the head of the dorsal horn. ACKNOWLEDGMENTS

We are greatly indebted to Mrs. Doris Hannum for her assistance in the processing of the histological material. We would also like to express our gratitude to Mrs. Nancy Richardson for her secretarial help. LITERATURE CITED Brinkman, R., and A. H. Martin 1973 A cytoarchitectonic study of the spinal cord of the domestic fowl Gtcllus gccllzts domesticus. 1. Brachial region. Brain Res., 5 6 : 4 S 6 2 . Cabot, J . B., D. H. Cohen a n d R. B. Leonard 1975 Spinal localization of' sympathetic preganglionic neurons in the pigeon (Colztmhtt tivicc) with retrograde transport of horseradish peroxidase. In preparation. Cohen, D. H. 1969 Development of a vertebrate experimental model for cellular neurophysiologic studies of learning. Cond. Ref., 4 : 61-80. 1974a The neural pathways and informational flow mediating a conditioned autonomic response. I n : Limbic and Autonomic Nervous System Research. L. V. DiCara, ed. Plenum Press, New York. 1974b Analysis of the final common

path for heart rate conditioning. In: Cardiovascular Psychophysiology. P. A. Obrist, A . H . Black, J. Brener and L. V. DiCara, eds. Aldine Publishing Co., Chicago. Ebbesson, S. 0. E. 1970 The selective silver impregnation of degenerating axons and their synaptic endings in non-mammalian species. In: Contemporary Research Methods in Neuroanatomy. W . J. H. Nauta and S. 0. E. Ebbesson, cds. Springer-Verlag, New York. Fink, R. P., a n d L. Heimer 1967 Two methods for selective silver impregnation of degenerating axons and their synaptic endings in the central nervous system. Brain Res., 4 : 369-374. Grant, G . , and B. Rexed 1958 Dorsal spinal root afferents to Clarke's column. Brain, 81 : 567-576. Huber, J. F. 1936 Nerve roots and nuclear groups i n the spinal cord of the pigeon. J. Comp. Neur., 6 5 : 43-91. Kappers, C. U. A., G. C. Huber and E. C. Crosby 1936 The Comparative Anatomy of the Nervous System of Vertebrates, Including Man. MacMillan Co., New York. Karten, H. 1963 Ascending pathways from the spinal cord of the pigeon (Colztmhcc livitr). Proc. 16th. Int. Congr. Zool., Washington, D.C., 2 : 23. Kliiver, H., and E. Barrera 1953 A method for combined staining of cells and fibers in the nervous system. J. Neuropath. Exp. Neurol., 1 2 : 400-404. Leonard, R. B., and D. H. Cohen 1975 Spinal terminal fields of dorsal root fibers in the pigeon (Columbci l i v i t r ) . J. Comp. Neur., 163: 181-192. 1975 Responses of' postganglionic sympathtic neurons in the pigeon to peripheral nerve stimulation. I n preparation. Macdonald, R. L., and D. H. Cohen 1970 Cells of origin of sympathetic pre- and postganglionic cardioacceleratory fibers in the pigeon. J . Comp. Neur., 4 0 : 343-358. Matulionis, D. H . 1972 Analysis of the developing avian glycogen body. I. Ultrastructural morphology. J. Morph., 1 3 7 : 46-82, Nieuwenhuys, R. 1964 Comparative anatomy of the spinal cord. Prog. Brain Res., 11. 1-57. Oscarsson, O., I . Rosen and N. Uddenberg 1963 Organization of ascending tracts in the spinal cord of the duck. Acta Physiol. Scand., 5 9 : 143153. Rexed, B. 1952 The cytoarchitectonic organization of the spinal cord i n the cat. J. Comp. Neur., 9 6 : 41-95, 1954 A cytoarchitectonic atlas of the spin a l cord in the cat. J. Comp. Neur., 100: 297379. 1964 Some aspects of the cytoarchitectonics and synaptology of the spinal cord. Prog. Brain Res., I 1 : 58-92. Shriver, J. E., B. M. Stein and M. B. Carpenter 1968 Central projections of spinal dorsal roots in the monkey. I. Cervical and upper thoracic dorsal roots. Am. J. Anat., 123: 27-73. Sprague, J . M. 1958 The distribution of dorsal root fibers on motor cells in the lumbosacral spin a l cord of the cat, and the site of excitatory and inhibitory terminals i n monosynaptic pathways. Proc. Roy. SOC. (London), Series B., 1 4 9 : 534-556. Sprague, J. M., and H. Ha 1964 The terminal fields of dorsal root fibers in the lumbosacral

SPINAL CY'I'OARCl1ITECTUKE I N THE PIGEON spinal cord of t h e cat, a n d the dendritic organization of the motor nuclei. Prog. Brain Kes., I I : 120-1 5 4 . Szaho, 1. 1965 Combined staining of CNS tissue with lux01 fast blue a n d basic fuchsin. Acta Morphol. Acad. Sci. Hung., 13: 251-253.

179

Terni, T. 1923 Kicerche anatomiche sul sistema nervoso autonoma degli Uccelli. Arch. Ital. Anat. Embriol., 20: 4 3 S 5 1 0 . van den Akker, L. M. 1970 An Anatomical Outline of the Spinal Cord of the Pigeon. van Gorcum and Comp. N.V., Assen.

A cytoarchitectonic analysis of the spinal cord of the pigeon (Columba livia).

The spinal gray of the pigeon is described cytoarchitectonically to establish a foundation for anatomical and physiological studies of the pigeon spin...
3MB Sizes 0 Downloads 0 Views