Exp. Brain Res. 35,395-406 (1979)
Experimental Brain Research @ Springer-Verlag 1979
Cat Phrenic Nucleus Architecture as Revealed by Horseradish Peroxidase Mapping Ch.L. Webber, Jr., R.D. Wurster, and J.M. Chung ~ Department of Physiology, Loyola University of Chicago, Stritch School of Medicine, 2160 South First Avenue, Maywood, I160153, USA
Summary. Cat phrenic motoneurons, labeled by intradiaphragmatic injection of horseradish peroxidase, formed a tight cluster in the most ventral portion of the ventral horn in lamina IX of the lower cervical cord. Cell counts were symmetrically distributed for 17 to 21 m m along the longitudinal axis of the cord with a unimodal peak at the junction of segments C5 and C6. The phrenic nucleus was bilaterally organized on either side of the cord with anatomical symmetry and in no case was there evidence for the crossing of phrenic axons in the cord. Assessment of cellular geometry and intercellular relationships demonstrated that phrenic cell diameters approximated a normal distribution with a single peak at 26 ptm while longitudinal cell lengths averaged 76 btm. Cells of different size were mixed randomly at all levels of the nucleus. The minimum distance between cells was about 10 btm and the m a x i m u m cell packing density approached 2 cells per 106 ~m 3. The results confirm the location of the cat phrenic nucleus, extend the knowledge of phrenic m o t o n e u r o n a l geometry, and provide an anatomical basis for the understanding of recruitment and synchronization p h e n o m e n a within the phrenic nucleus.
Key words: Cat - Spinal cord - Phrenic motoneurons - Neuron geometry Horseradish peroxidase
There have been several attempts with varying degrees of success to localize the phrenic nucleus in the cat cervical spinal cord. One excellent m a p was constructed by Keswani et al. (1954) who sectioned the left phrenic nerve in adult cats to p r o m o t e chromotolysis in phrenic motoneurons. They localized the phrenic nucleus to a group of cells positioned between the ventromedial and ventrolateral cell columns of the ventral gray horn throughout the fifth and sixth cervical cord segments. Using the retrograde degeneration technique, Schmiedt
1 Present address: Department of Physiology, Yonsei University, School of Medicine, Seoul, Korea
0014-4819/79/0035/0395/$
2.40
396
C.L. Webber, Jr. et al.
( 1 9 6 4 ) a n d Sterling a n d K u y p e r s ( 1 9 6 7 ) h a v e c o n f i r m e d this p h r e n i c nucleus p o s i t i o n in cats. O t h e r s , h o w e v e r , h a v e failed to o b t a i n p o s i t i v e histological results following p h r e n e c t o m y in cats ( A c h e s o n et al., 1942; B a u m g a r t e n et al., 1963) n e c e s s i t a t i n g the use o f o t h e r m a r k i n g t e c h n i q u e s . F o r e x a m p l e , B a u m g a r t e n et al. (1963) d e p o s i t e d i r o n at spinal c o r d sites f r o m which e x t r a c e l l u l a r p h r e n i c a n t i d r o m i c a c t i o n p o t e n t i a l s c o u l d be e v o k e d . W e b b e r and P l e s c h k a (1976) i n t r a c e l l u l a r l y l a b e l e d p h r e n i c m o t o n e u r o n s with P r o c i o n yellow. U n f o r t u n a t e l y , t h e s e two e x p e r i m e n t a l a p p r o a c h e s c o u l d n o t p r a c t i c a l l y d e l i m i t t h e o r g a n i z a t i o n of t h e c o m p l e t e p h r e n i c nucleus in i n d i v i d u a l cats. A l t e r n a t e l y , t h e staining o f the feline spinal c o r d by the classical silver ( K e s w a n i a n d H o l l i n s h e a d , 1955) or G o l g i - C o x (Sterling a n d K u y p e r s , 1967) t e c h n i q u e s h a v e b e e n s h o w n to l a b e l a significant fraction o f t h e p h r e n i c nucleus in any o n e p r e p a r a t i o n ; h o w e v e r , t h e s e t e c h n i q u e s a r e at b e s t i n d i r e c t since the stains do n o t d i f f e r e n t i a t e b e t w e e n p h r e n i c a n d o t h e r spinal cells. I n t e r e s t in the p h r e n i c nucleus goes far b e y o n d its a n a t o m i c a l p o s i t i o n in l a m i n a I X ( R e x e d , 1952) o f t h e cervical spinal cord. U n d e r s t a n d i n g of f u n c t i o n a l n e u r o n a l m e c h a n i s m s such as cell r e c r u i t m e n t a n d s y n c h r o n i z a t i o n within the p h r e n i c nucleus ( I s c o e et al., 1976; W e b b e r a n d Pleschka, 1976) m u s t have an a n a t o m i c a l basis. W h i l e this a p p r o a c h has b e e n i n i t i a t e d for a few p h r e n i c m o t o n e u r o n s ( W e b b e r a n d P l e s c h k a , 1976), no studies h a v e e x a m i n e d q u a n t i t a t i v e l y m a n y i n d i v i d u a l cells o f a single p h r e n i c nucleus. T h e p r e s e n t w o r k was i n i t i a t e d to selectively l a b e l a large fraction of t h e p h r e n i c nucleus utilizing t h e r e t r o g r a d e t r a n s p o r t o f h o r s e r a d i s h p e r o x i d a s e ( H R P ) ( L a V a i l a n d L a V a i l , 1974). L a b e l e d n e u r o n s w e r e m a p p e d to d e f i n e t h e limits o f the p h r e n i c nucleus w h i c h w e r e s u b s e q u e n t l y c o m p a r e d with p r e v i o u s l y c o n s t r u c t e d maps. G e o m e t r i c a l c h a r a c t e r i s t i c s of i n d i v i d u a l p h r e n i c cells a n d cell g r o u p s w e r e t h e n q u a n t i f i e d a n d c o r r e l a t e d with k n o w n f u n c t i o n a l p r o p e r t i e s o f p h r e n i c m o t o n e u r o n s . G e o m e t r i c a l c o m p a r i s o n s w e r e also d r a w n between phrenic and lumbosacral motoneurons.
Methods Fourteen adult cats of either sex (1.5-4.7 kg) were anesthetized with sodium pentobarbital (35-40 mg/kg, i.p.). While maintaining positive pressure ventilation through an endotracheal tube, the thoracic cage was opened. In twelve cats the right phrenic nerve was cut one to two cm above the diaphragm; in two cats both phrenic nerves were left intact. Using a 25 ~1 Hamilton syringe, 0.5 ml of a 20% acqueous HRP solution (Sigman, Type IV) was injected into 60-80 sites of the left hemidiaphragm (or whole diaphragm in the two phrenic intact cats) through either a thoracic cage or abdominal window. After a 48 h survival period the cats were reanesthetized, and perfused through the left ventricle with 300-500 ml of physiological saline followed by 1-1.5 L of fixative solution (2.5 % glutaraldehyde, 0.5 % formaldehyde, 0.1 M phosphate buffer, pH 7.45). A cervical laminectomy was then performed and the cord from mid-C4 to mid-C7 was removed. A needle was inserted longitudinally through the right dorsal horn region of the excised spinal cord for later histological orientation. The dura mater was cut away and the cord stored in perfusate at 4~ C for at least 12 h. The fixed spinal cord was placed in a phosphate buffered solution of 20% sucrose at 4~ C for 12 h. The cord was then mounted in embedding medium and 40 gm frozen sections were cut in the transverse plane on a freezing microtome. Every 25 serial sections were collected and processed in
Cat Phrenic Nucleus Architecture
397
common for 30 rain in Tris buffer (pH 7.6) containing 100 mg diaminobenzidine and 0.75 ml of 3 % hydrogen peroxide per 200 ml. The sections were then mounted on glass slides, each of which held twenty-five 40 gm sections or one mm of spinal cord. Preparations with the best HRP labeling were counterstained with creyslechtviolett.
Results
HRP Labeling of Phrenic Montoneurons All spinal cord slides were carefully examined for cells containing H R P reaction product using light and/or dark field microscopy. O f 14 cats that were injected with H R P (one of which died post-operatively), 8 cats unexplainably failed to show any cell labeling or only weak, sporadic labeling at best. Five cats exhibited excellent phrenic soma labeling with H R P from which the data in this p a p e r are taken. The labeling of non-phrenic somas in the cervical cord was not considered a problem; since the diaphragm is located in the lower thoracic region, excess H R P spreading to intercostal or abdominal muscles should be transported to the thoracic spinal cord. Examples of H R P labeled phrenic motoneurons are given in the photomicrographs of Fig. 1. While the granulated cytoplasm, nucleus and nucleolus are best observed under light microscopy (A), dark field illumination of the same region better delineates cell body borders and the presence of H R P reaction product in the large proximal dendrites (B). The most striking feature of the phrenic nucleus is the tight organization or clustering of neurons at all levels of the nucleus including the rostral (B), middle (C) and caudal (D) regions. Intercellular distances between labeled cells were often below 50 ~tm and even approached 10 ~tm in some instances. In cross section the cell bodies were round to oval. O f the five cats yielding usable data, four presented labeled phrenic m o t o n e u r o n s with unclear nucleoli due to cell shrinkage problems. Labeled neurons from the fifth cat (C-14), however, were consistently well stained and showed neither signs of cellular distortion nor cell shrinkage. Detailed counts of H R P nucleolated cells were thus facilitated in this p r e p a r a t i o n . F o r example, of 1349 labeled phrenic cell sections, only 709 were found to be nucleolated, giving a nucleolated to total ratio of 52.6 %. Since the ratio value was less than unity, individual phrenic cell bodies had to be appearing in more than one 40 ~tm cord section. By calculation, the average longitudinal length of phrenic somas is demonstrated to be 76 ~tm (40 ~tm sections/0.526). Correcting the other cat data to nucleolated counts by using this 5 2 . 6 % factor gave 298, 406, 427, and 656 labeled cells in the left phrenic nucleus.
Anatomical Localization and Distribution of Phrenic Cells in the Cervical Cord Using a camera lucida attachment to the microscope, transverse plane phrenic localization maps were made in 1 m m increments throughout the entire phrenic nucleus. In the four cats with severed right phrenic nerves, labeled neurons could only be found in the left ventral horn of the spinal gray indicating that
398
C.L. Webber, Jr. et al.
Fig. 1A-D. Lightfield (A) and darkfield (B-D) photomicrographsof phrenic motoneurons labeled with horseradish peroxidase (HRP) in the left phrenie nucleus of cat C-14. (A and B) the same transverse section through the rostal nucleus; (C) section through the middle nucleus; (D) section through the caudal nucleus. The cells are viewed from their caudal ends at equivalent magnifications as given in A, with dorsal (D) and lateral (L) directions provided
phrenic axons do not cross over in the cord to join the contralateral ventral root. In Fig, 2 representative maps are reconstructed from one cat to demonstrate phrenic localization and relative cell density at several positions within the phrenic nucleus from upper C5 to lower C6. The phrenic cell cluster was consistently located within the most ventral portion of the anterior horn of the gray matter (lamina IX), about 3 mm below the dorsal cord surface. Starting at the rostral end of the nucleus and progressing caudally, it is evident that the center of the phrenic nucleus remains a constant distance of about 900 ~tm from the ventral median fissure despite the concurrent lateral extension of the gray matter accommodating brachial plexus motoneurons. The phrenic nucleus at C5 starts as a small cluster of cells (A) that grows in diameter until mid C 5 - C 6 (C) below which the diameter of the nucleus again decreases (E). Thus, the phrenic nucleus is spindle shaped with a maximal medial to lateral diameter of 340 ~tm.
Cat Phrenic Nucleus Architecture
399
Fig. 2A-E. Cross sectional maps of the left phrenic nucleus of Cat C-14 at different levels: A upper C5; B middle C5; C junction of lower C5 and upper C6; D middle C6; E lower C6. Each map represents the projection of HRP labeled phrenic motoneurons onto the transverse plane (dots) for 1 mm segments of spinal cord. The photomicrographsof Fig. 1A and B are taken from a section intermediate between A and B above; Fig. 1C and D correspond to C and D above
The longitudinal organization of the phrenic nucleus was studied by plotting the number of nucleolated cells found in each mm of spinal cord throughout the nucleus. Five separate histograms were thus obtained (one from each cat) which were combined by centering rostral and caudal extents (17-21 mm) and averaging the number of phrenic cells (nucleolated counts or estimates) in corresponding histogram bins. The resultant histogram is plotted in Fig. 3. It can be noted that the distribution is unimodal with a high degree of cell overlap at the junction of spinal cord segments C5 and C6. Phrenic maps were also constructed from one succesful cat in which both phrenic nerves were left intact and where the whole diaphragm was injected with HRP. Orientating cord sections to the needle track placed in the right dorsal horn, the two reconstructed nuclei on either side of the cord were found to be comparable in terms of cluster diameter and longitudinal extent. Hence, the assumption that the phrenic nucleus is bilaterally symmetrical is an acceptable notion.
Size and Spacing of Phrenic Motoneurons The major long (L) and minor short (S) diameters of the 709 nucleolated cells of cat C - 1 4 were measured with an eyepiece micrometer at a final magnification of 400 x. Large dendrites were eliminated by application of the imaginary oval
400
C.L. Webber, Jr. et al.
60U3 ._1 ._1 Ill
0
n
40-
2
14.
0 n-
uJ ,.n =E
-'-1
I i
i
20-
::) Z
C~
OI 0
I 5
LONGITUDINAL
I 10 CORD
I 15 DISTANCE,
I 20 mm
Fig. 3. Longitudinal distribution of 2496 H R P labeled phrenic motoneurons in the left cervical spinal cord of 5 cats. The number of phrenic cells (mean + SEM) is plotted as a function of longitudinal distance down the spinal cord (1 m m bins; rostral to the left; caudal to the right). Ventral roots exit at C5 (B) and C6 (D). A - E correspond to Fig. 2 A - E
technique (Burke et al., 1977) and no corrections were introduced for cell shrinkage. As shown in Fig. 4, the long axes are skewed to the right (A) while the short axes are skewed to the left (B). These data indicate that phrenic motoneuron profiles are more oval than round in cross-section. This was quantitatively established by calculating the eccentricity (e) of each phrenic cell body in the transverse plane: e = ~ Only 12% of the cells were round (e = 0) while 88 % were oval with e ranging from 0.359 to 0.949 in a rectangular distribution. Mean cell diameters ( L / 2 + S / 2 ) , ranging in size from 10 to 48 ~m, were more normally distributed about a single mode of 26 ~m (C). To check for distribution differences of cells of different size at various regions in the phrenic nucleus, the spinal cord was parcelled up into seven 3 mm segments along the longitudinal axis. For each segment a histogram distribution of mean cell diameters was plotted as given in Fig. 5. In no case was there any evidence for cells of a specified size to be located in a preferential position within the phrenic nucleus. Thus, phrenic cells of small, medium, and large diameters are arranged at random within the nucleus. Using cat C-14 data, an attempt was made to quantify phrenic cell densities throughout the phrenic nucleus. The cross-sectional areas of phrenic nucleus projections in the transverse plane were measured (cf. Fig. 2) which were individually multiplied by 1 mm of longitudinal cord length to estimate the volume of gray matter in which the phrenic cells were found. The number of labeled, nucleolated cells per volume of cord so calculated gave measures of cell density. For example, phrenic cell densities ranged from 0.99 to 1.25 cells per 106
Cat Phrenic Nucleus Architecture
401
A 150100500 03 ..1 ..J I.U
B 150
Fig. 4A-C. Transverse cell body
dimensions of 709 HRP labeled phrenic motoneurons in the left cervical spinal cord of cat C-14. A distribution of major long (L) axis diameters (2.6 ~m bins); B distribution of minor short (S) axis diameters (2.6 ,am bins); C distribution of mean (L/2+S/2) axis diameters (1.3 ~m bins). Note that: long axis diameters are positively skewed to the right (mode = 26 `am; mean = 30 ~tm); short axis diameters are negatively skewed to the left (mode = 26 ~xm;mean = 22 am); mean axis diameters are more symmetrically distributed (mode = 26 ~tm; mean = 26 gm)
I.t.
o 14.1
IE Z
100 50 0C 755025O-
m
I
9.1
I
I
24.7
I
I
I
40.3
DIAMETER,
I
55.9
pm
~m 3 in u p p e r C5; f r o m 1.36 to 1.87 cells p e r 106 g m 3 at the C 5 - C 6 j u n c t i o n ; a n d f r o m 0.69 to 0.80 cells p e r 106 ~m 3 at l o w e r C6. Thus, p h r e n i c cells t e n d to b e m o r e tightly p a c k e d at the m i d d l e o f t h e n u c l e u s while b o t h r o s t r a l a n d c a u d a l e n d s o f t h e n u c l e u s h a v e r e l a t i v e l y l o w e r cell densities. C o m p a r e d with u p p e r C5, h o w e v e r , l o w e r C6 is m o r e diffusely o r g a n i z e d .
Discussion
Structural Considerations T h e p h r e n i c nucleus m a p s d e f i n e d in this s t u d y b y cell l a b e l i n g with H R P e x a c t l y c o n f i r m the r e t r o g r a d e d e g e n e r a t i o n m a p s o f K e s w a n i et al. (1954), S c h m i e d t (1964), a n d Sterling a n d K u y p e r s (1967). T h e m a p s also a g r e e with the p o s i t i o n o f p h r e n i c m o t o n e u r o n s l o c a l i z e d b y e x t r a c e l l u l a r a c t i v a t i o n ( B a u m g a r t e n et al., 1963) a n d i n t r a c e l l u l a r staining ( W e b b e r a n d P l e s c h k a 1976). A l t h o u g h the nucleus e x t e n d s t h r o u g h o u t two spinal c o r d s e g m e n t s (C5 a n d C6), the
402
C.L. Webber, Jr. et al. 50A
,
1---],,,
e5
B
,
r 1
,oo
0500-
50r ,.I _1 IJJ
C 0-
i
0
50-
D UJ e-,
z
|
0-
50-
E
[
F
I
,,,
0500-
~
90
50G
2
0 I
9.8
I
I
22.8
I
I
35.8
DIAMETER,
I
Fig. 5A-G. Distribution of mean phrenic cell body diameters (L/2+S/2) at different levels of the left phrenic nucleus of cat C-14. The number of cells included in each serial 3 mm segment from upper C5 A to lower C6 G are given at the right. The bin size is set at 6.5 btm. With the exception of A and G (extreme ends of the phrenic nucleus) modal cell diameters are constant at 26 ~xm.Average cell diamters at each level are also comparable: A 30 ~*m;B 27 btm; C 25 gm; D 25 ~xm;E 26 gm; F 27 btm; G 33 Ixm
48.8 pm
constituent m o t o n e u r o n s form a discrete nuclear grouping both in the transverse (Fig. 2) and longitudinal (Fig. 3) planes. T h e deep location and small target size of the phrenic nucleus clearly explain the difficulty in recording extracellularly ( B a n m g a r t e n et al., 1963; P u r p u r a and Chatfield, 1953) and intracellularly (Gill and Kuno, 1963; W e b b e r and Pleschka, 1976) f r o m phrenic m o t o n e u r o n s . H i n s e y et al. (1939) have c o u n t e d 1000 myelinated fibers in the right phrenic nerve and a b o u t 100 myelinated sensory afferents in the left phrenic nerve. A s s u m i n g that left and right nuclear groups are comparable, this would imply that there are approximately 900 phrenic m o t o n e u r o n s in each half nucleus or about 1800 n e u r o n s total. F r o m the nucleolated cell counts in this study, therefore, it is estimated that H R P labeled f r o m 33 % to 79 % of the left phrenic nucleus in the different cat preparations. If myelinated accessory fibers join the phrenic nerve in the peripheral third of its course (Wilson, 1970; but see S a n t ' A m b r o g i o et al., 1963), a possibility not taken into account by Hinsey et al. (1939), this labeling success would be apparently improved. T h e failure to label all phrenic m o t o n e u r o n s in the left phrenic nucleus is related to the m e t h o d s e m p l o y e d in this study. First, not all phrenic efferent
Cat Phrenic Nucleus Architecture
403
axon terminals were exposed to HRP. Considering the structure of the diaphragm, it is a thin, dome-shaped muscle with a large surface area. Given a total H R P injection volume of 0.5 ml, it was difficult to completely cover the entire hemidiaphragm, especially since the circumferential border of the diaphragm was not very accessible. Second, some left phrenic axons picking up H R P were unavoidably severed when the right phrenic nerve was cut close to the diaphragm. This is so because left phrenic efferents have been shown functionally to cross over to the right phrenic nerve in the thoracic cage before crossing back to the left side in the diaphragm itself (Sant'Ambrogio et al., 1963). Third, some smaller phrenic axons may have transported H R P so slowly that the cell bodies, 170 mm distant, were never reached in 48 h. The minimum HRP transport velocity allowed does calculate out to the 3.5 mm/h found by LaVail and LaVail (1974), however, in the chick visual system. Chromatolysis has been used as a successful cell marker for lumbosacral (Romanes, 1951) as well as phrenic motoneurons (Keswani et al., 1954; Schmiedt, 1964; Sterling and Kuypers, 1967). However, as pointed out by Burke et al. (1977), outside of nuclear placement in the cord, little information on neuronal size or spatial organization of lumbosacral motoneurons has been provided by retrograde degeneration studies. The same is true for phrenic motoneurons. The problem of making such measurements may be related to the swelling of cat spinal motoneurons (increases in cross sectional area by 20 %) 7 to 26 days following axotomy (Barr and Hamilton, 1948; but see Schad6 and Harreveld, 1961). Nevertheless, Burke et al. (1977) overcame the problem by studying lumbosacral motoneurons labeled with HRP. Likewise, the same labeling technique has been applied to phrenic motoneurons in the present study, making possible the contribution of the quantification of individual phrenic motoneuronal characteristics for a significant fraction of the left phrenic nucleus. The cell bodies of most phrenic motoneurons are eccentric (oval) in the transverse plane and have modal long, short and average diameters of 26 ~m (Fig. 4). In the longitudinal axis, the neurons are elongated 76 ~tm on the average, giving a length to width ratio of about 3 to 1. Phrenic cells are, therefore, fusiform in shape (Keswani and Hollinshead, 1955; Webber and Pleschka, 1976), in contrast to lumbosacral motoneurons which are more spherical (Aitken and Bridger, 1961). Comparing the dimensions of Procion yellow stained phrenic motoneurons (Webber and Pleschka, 1976) with the present values (assuming shrinkage to be minimal) demonstrates that the intracellular stabs in the former study were selective for the larger cells. Thus, not only is the phrenic target nucleus small, but the individual cell profiles are themselves small making it difficult to establish and maintain stable intracellular recordings. Comparisons can now be made between phrenic and lumbosacral (Burke et al., 1977) motoneurons, both labeled with H R P and neither corrected for shrinkage. The mean diameters of lumbosacral motoneurons (average long and short transverse axes) are distributed bimodally. The larger alpha motoneurons have a mode diameter of 52 ~tm while the smaller gamma motoneurons have a mode diameter of 27 ~m. Phrenic motoneurons, on the other hand, have an
404
C.L. Webber, Jr. et al.
approximately symmetrical distribution of mean diameters around a single mode of 26 ~m (Fig. 4C). These values establish the argument that phrenic motoneurons are smaller in geometrical dimensions than lumbosacral alpha motoneurons (Webber and Pleschka, 1976), explaining the differences in membrane properties between these two cell types (Gill and Kuno, 1963). The lack of a second modal peak in the phrenic distribution implies a deficiency of gamma motoneurons within the phrenic nucleus corresponding to the scarcity of muscle spindles in the diaphragm (Corda et al., 1965). The question of whether the cell bodies of interneurons are positioned directly within the phrenic nucleus still remains an unanswered question. Keswani et al. (1954), have conjectured that the small, round, nonchromalytic cells in the phrenic nucleus were interneurons. However, since the axon reaction technique is not specific for all phrenic motoneurons in a single preparation, this suggestion is not conclusive. For example, Schmiedt (1964), using the degeneration technique in cats, described the phrenic nucleus as extending only 14 mm in the rostro-caudal axis of the cord. Thus, a fair fraction of the nucleus went undetected as compared with the longitudinal nucleus extent of 21 mm in the present study. Indeed, others were not able to consistently mark phrenic motoneurons by axotomy experiments (Acheson et al., 1942; Baumgarten et al., 1963). In the current study, counts of counterstained and nucleolated cells intermingled with HRP labeled phrenic motoneurons showed that there are more than 900 cells in the left phrenic nucleus. This excess of cells over the number of myelinated phrenic efferents (Hinsey et al., 1939) may suggest that interneurons are present. However, Nyberg-Hansen (1965) found that medullary reticulo-spinal fibers terminate for the most part in lamina VII, most likely on interneurons. Only a few terminations were found on large alpha motoneurons in lamina IX. Clearly, further studies need to be carried out to follow up on this interneuron problem. Functional Considerations
During inspiration, phrenic motoneurons show a stable order of recruitment (Iscoe et al., 1976) based upon cell size and excitability differences (Webber and Pleschka, 1976). Therefore, the wide variation in phrenic cell size fi'om 10 ~m to 48 ~tm (Fig. 4) fits well with the smooth continuum of phrenic cell excitabilities and slow-wave membrane potential profiles (Webber and Pleschka, 1976). Since phrenic cells of different size are randomly mixed throughout the nucleus (Fig. 5), the full range of phrenic cell types can be found in any selected portion of the nucleus. This nucleus architecture probably corresponds to the fiber type organization of the cat diaphragm which contains both slow and fast twitch fibers (Davies and Gunn, 1972) which are distributed in a mosaic pattern in the sternal, crural and costal regions (J. A. Faulkner, personal communication). This phrenic organization can be distinguished from that of the medial gastrocnemius motor nucleus which has larger motoneurons congregating in the rostral nucleus which innervate the dorsal portion of the gastrocnemius muscle, and smaller motoneurons in the caudal nucleus which project to the ventral margin of the muscle (Burke et al., 1977).
Cat Phrenic Nucleus Architecture
405
A second functional point of emphasis regarding phrenic cell organization in the cervical spinal cord relates tocellsynchronization duringtheinspiratoryphase of respiration. Although starting times may differ (Iscoe et al., 1976; Webber and Pleschka, 1976), phrenic motoneurons discharge in synchrony as if all components belonged to the same neural net. Indeed, phrenic cells are tightly clustered throughout the entire phrenic nucleus (Fig. 1) with a high degree of cell overlap between C5 and C6 spinal cord segments (Fig. 3). Considering this, several reasons may be put forth to explain this synchrony. (1) Phrenic cells receive synchronous inputs from medullary inspiratory neurons that synchronize phrenic outputs (Cohen et al., 1974). (2) Close approximation of phrenic cell bodies implies extensive overlapping of dendritic fields making possible the simultaneous excitation of different units in the population (Keswani and Hollinshead, 1955; Sterling and Kuypers, 1967; Webber and Pleschka, 1976). Synaptic inputs should thus be distributed over several phrenic units separated in space. (3) Phrenic cell body membranes may be so close as to permit the establishment of ephaptic connections between neighbor neurons as observed for abducens motoneurons (Gogan et al., 1974). This may involve electrical interactions within dendritic bundles that would lead to synchronization (Matthews et al., 1971; Sterling and Kuypers, 1967). (4) A high phrenic cell density favors the accumulation of potassium ions from discharging cells. Elevated extracellular potassium levels could impose a generalized depolarization and synchronization on neurons within the phrenic nucleus as has been described for medullary respiratory neurons in the region of the nucleus ambiguus (Richter et al., 1977). If this potassium mechanism is in operation within the phrenic nucleus, the slow-wave membrane potential of phrenic motoneurons (Webber and Pleschka, 1976) cannot be entirely attributed to the central respiratory drive potential (summed EPSPs) as originally introduced for thoracic respiratory motoneurons (Eccles et al., 1962). Clearly, the structure of the phrenic nucleus implicates several important mechanisms all playing a role in the formation of a final neural output to the diaphragm. Finally, the finding that phrenic axons do not cross in the spinal cord suggests that the crossed-phrenic phenomenon must be mediated by crossed inputs to phrenic motoneurons as opposed to crossed outputs. The response is probably mediated by interneurons in the spinal cord that drive contralateral phrenic motoneurons. Activity in this crossed pathway is directly related to the intensity of the central respiratory discharge (Lewis and Brookhart, 1951). Acknowledgements. This work was supported by National Heart and Lung Institute Research Grant HL 08682. Much appreciationis extended to Mrs. Mira Milosavljevicfor her histologicalexpertise in processingthe many spinal cords for this study. The dedicatedefforts of Mr. Ronald Longinoand Miss Mary McGoldrickare gladlyacknowledgedin the areas of data analysisand figurepreparation. References
Acheson, G. H., Lee, E. S., Morison, R. S.: A deficiencyin the phrenicrespiratorydischargesparallel to retrograde degeneration.J. Neurophysiol.5, 269-273 (1942) Aitken, J. T., Bridger,J. E.: Neuron size and neuron population densityin the lumbosacralregion of the cat's spinal cord. J. Anat. 95, 38-53 (1961) Barr, M.L., Hamilton,J.D.: A quantitativestudy of certain morphologicalchanges in spinal motor neurons during axon reaction. J. Comp. Neurol. 89, 93-121 (1948)
406
C.L. Webber, Jr. et al.
Baumgarten, R. von, Schmiedt, H., Dodich, N.: Microelectrode studies of phrenic motoneurons. Ann. N.Y. Acad. Sci. 1119, 536-544 (1963) Burke, R.E., Strick, P.L., Kanda, K., Kim, C. C., Walmsley, B.: Anatomy of medial gastrocnemius and soleus motor nuclei in cat spinal cord. J. Neurophysiol. 40, 667-680 (1977) Cohen, M.I., Piercey, M.F., Gootman, P.M., Wolotsky, P.: Synaptic connections between medullary inspiratory neurons and phrenic motoneurons as revealed by cross-correlation. Brain Res. 81, 319-324 (1974) Corda, M., Euler, C. yon, Lennerstrand, G.: Proprioceptive innervation of the diaphragm. J. Physiol. (Lond.) 178, 161-177 (1965) DavieS, A.S., Gunn, H.M.: Histochemical fibre type in the mammalian diaphragm. J. Anat. 112, 41-60 (1972) Eccles, R.M., Sears, T.A., Shealy, C.N.: Intra-cellular recording from respiratory motoneurons of the thoracic spinal cord of the cat. Nature 193, 844-846 (1962) Gill, P.K., Kuno, M.: Properties of phrenic motoneurons. J. Physiol. (Lond.) 168, 258-273 (1963) Gogan, P., Gueritaud, J.P., Horcholle-Bossavit, G., Tyd-DuMont, S.: Electrotonic coupling between motoneurons in the abducens nucleus of the cat. Exp. Brain Res. 21, 139-154 (1974) Hinsey, J.C., Hare, K., Phillips, R.A.: Sensory components of the phrenic nerve of the cat. Proc. Soc. Exp. Biol. Med. 41, 411-414 (1939) Iscoe, S., Dankoff, J., Migicovsky, R., Polosa, C.: Recruitment and discharge frequency of phrenic motoneurons during inspiration. Respir. Physiol. 26, 113-128 (1976) Keswani, N.H., Groat, R.A., Hollinshead, W.H.: Localization of the phrenic nucleus in the spinal cord of the cat. J. Anat. Soc. India 3, 82-89 (1954) Keswani, N. H., Hollinshead, W. H.: The phrenic nucleus. III. Organization of the phrenic nucleus in the spinal cord of the cat and man. Proc. Mayo Clin. 30, 566-577 (1955) LaVail, J. H., LaVail, M.M.: The retrograde intraaxonal transport of horseradish peroxidase in the chick visual system: a light and electron microscope study. J. Comp. Neurol. 15"7, 303-358 (1974) Lewis, L.J., Brookhart, J.M.: Significance of the crossed phrenic phenomenon. Am. J. Physiol. 166, 241-254 (1951) Matthews, M.A., Willis, W.D., Williams, V.: Dendrite bundles in lamina IX of cat spinal cord: a possible source for electrical interaction between montoneurons? Anat. Rec. 171, 313-328 (1971) Nyberg-Hansen, R.: Sites and mode of termination of reticulo-spinal fibers in the cat. An experimental study with silver impregnation methods. J. Comp. Neurol. 124, 71-100 (1965) Purpura, D.P., Chatfield, P.O.: Changes in action potentials of single phrenic motor neurons during activity. J. Neurophysiol. 16, 85-92 (1953) Rexed, B.: The cytoarchitectonic organization of the spinal cord in the cat. J. Comp. Neurol. 96, 415-495 (1952) Richter, D.W., Sonnhof, U., Camerer, H.: Changes in extracellular potassium during the spontaneous activity of medullary respiratory neurons. Pflfigers Arch. 376, 139-149 (1978) Romanes, G.J.: The motor cell columns of the lumbo-sacral spinal cord of the cat. J. Comp. Neurol. 94, 313-363 (1951) Sant'Ambrogio, G., Frazier, D.T., Wilson M.F., Agostoni, E.: Motor innervation and pattern of activity of cat diaphragm. J. Appl. Physiol. 18, 43-46 (1963) Schad6, J.P., Harreveld, A. yon: Volume distribution of moto- and interneurons in the peroneus-tibialis neuron pool of the cat. J. Comp. Neurol. 117, 387-398 (1961) Schmiedt, H.: Neurophysiologische und neuroanatomische Untersuchungen an Phrenicusmotoneuronen der Katze. Doctoral dissertation submitted to the faculty of the Georg-August-Universitgt, G6ttingeu 1964 Sterling, P., Kuypers, H. G. J.M.: Anatomical organization of the brachial spinal cord of the cat. II. The motoneuron plexus. Brain Res. 4, 16-32 (1967) Webber, C.L., Jr., Pleschka, K.: Structural and functional characteristics of individual phrenic motoneur0ns. Pflfigers Arch. 364, 113-121 (1976) Wilson, A.S.: Experimental studies on the innervation of the diaphragm in cats. Anat. Rec. 168, 537-548 (1970) Received July 19, 1978