- Email: [email protected]
Development of the human pontine nuclei: a morphometric study
13
Developmental Brain Research, 65 (1992) 13-20 ~) Elsevier Science Publishers B.V. All rights reserved. 0165-3806192/$05.00 BRESD 51385
Development of the human pontine nuclei: a morphometric study Hidetsugu
N o z a k i 1'2, N o b o r u
Goto 3 and Takahiro
Nara 2
ZDepartment of Pediatrics, Jikei University School of Medicine, Tokyo (Japan), ZDivision of Neurology, Saitama Children's Medical Center, Saitama (Japan) and 3Department of Anatomy, Showa University School of Medicine, Tokyo (Japan) (Accepted 29 August 1991)
Key words: Pontine nuclei; Morphometry; Human brain; Brain development; Large neuron
The morphometric development of the pontine nuclei in the human fetus from 16 to 40 gestational weeks, in a 2-month-old infant and in a 63-year-old adult were examined employing a serial celloidin section method and computer assisted electronic planimeter. The results of our study can be summarized as follows: (1) the development of the human pontine nuclei accelerated in volume after 32 gestational weeks and continued after birth, (2) neuron numbers remained relatively constant after 27 gestationai weeks. It was difficult to clearly distinguish neurons from gila before 27 gestationai weeks. The total estimated neuronal numbers were not indicative of the gestational stages in infants 27 gestationai weeks and older, (3) individual neurons appeared to continue to develop after 32 gestational weeks in accordance with size, distribution and circularity ratio, (4) many islet-shaped groups of large neurons appeared and were scattered throughout the oontine nuclei after 32 gestational weeks. INTRODUCTION The pontine nuclei are one of the important relay nuclei which have an apparent significance in the conduction of impulses from almost all parts of the cerebral cortices to the cerebellum (neocerebellum). The normal cytogenesis ,and histogenesis of this nucleus was initially studied by Essick 4-e in man, pig and rabbit, and subsequently by others in the chick t2 and mouse 2e. More recently, Rakic ~'24 reported the neurophilic mode of migration of pontine neurons in developing rhesus monkey. Altman and Bayer 3 reported that the pontine neurons arise later than those of the inferior olive originating from the neuroepithelium of the rhombic lip of the fourth Ventricle. T hey do not appear to differentiate until they reach their final destination adjacent to the longitudinally oriented fibers of the cerebral peduncle. Mihailoff et a1.19 demonstrated in the rat that an intense wave of synaptic proliferation was observed during the postnatal period. We think the morphometric and developing studies on the human pontine nuclei still have not been fully analyzed, especially focussing on the perinatal period. In this study, the development of the pontine nuclei in the human fetus, infant and adult was examined employing a serial section method combined with an electronic planimeter to determine morphometric cytoarchitectonic parameters.
MATERIALS AND METHODS Nine human brains were used for this study. Seven were from fetuses of 16, 21, 23, 27, 32, 35 and 40 weeks of gestation (WG). The other two were from a 2-month-old infant (2M) and a 63-yearold adult (63Y). Five of the fetal brains were obtained by hysterectomy performed because of myoma uteri (16WG) or from spontaneous abortions resulting from cervical incompetence (16, 21, 23 and 27WG). The live born fetuses obtained at 32, 35 and 40WG died respectively of sepsis, erythroblastosis fetalis and massive aspiration. There was no evidence of macroscopic abnormalities externally and internally, including the central nervous system. The crown-rump length and body weight were within normal ranges. After fixation in formalin (3.7% formaldehyde solution), the whole brains from the fetuses at 16, 21, 23 and 27WG and the brain stems with the cerebelliT(of the remaining 5 brains) were immersed in a secondary fixative, which was composed of 5% potassium dichromate and 5% potassium chromate (1:4, v:v). The secondary fixative was changed several Umes over a 3-week period (for the first two weeks the fixative was maintai~ed at room temperature, for the last week at 37"C) en bloc for the fetuses and after removal of the cerebrum for the others. The brains were mounted in celloidin after washing in running water and dehydrated with alcohol. The celioidin blocks were cut into 30./~m-(fetai and neonate brains) or 40-pm-thick (infant and adult brains) serial transverse sections. All sections were numbered with India ink. Every tenth section was stained by the KI0ver-Barrera (K-B) method. Approximately 10% shrinkage of the sections occurred during the entire procedure. The cytoarchitectonic development of the pontine nuclei was studied using both qualitative and quantitative variables. For the volumetry of structures, ten to twelve sections containing pontine nuclei were chosen at equal intervals. For each brain, enlarged negative prints were made from the selected sections utilizing the same magnification (16-22 times). In each print, areas of pontine regions were measured with an optical electronic planimeter (Digiplan, Kontron Co.). The sectional area (SA) of the pontine nuclei was determined as
Correspondence: H. Nozaki, Department of Pediatrics, Jikei University School of Medicine, 25-8 Nishi-shinbashi 3, Minato-ku, Tokyo 105, Japan. Fax: (81) 03-3435-8665.
14
4.0 ¸ 2000 1000: 3.0"
E .c
r~
x
::' 500. >o
k, 2.0, E
I6WG' 21WG 23WG 27WG 32WG 35WG 40WG 2M" HY'"
Fig. 1. Total volume of pontine nuclei (right and left) in mm 3. The volume is calculated from the section thickness, number of sections containing the nucleus and the average area per section. WG, gestational age in weeks; M, post-natal age in months; Y, years.
2/WG* 32WG 35WG 40WG 2M** 63Y***' Fig. 2. Neuronal number of the pontine nuclei. Before 27 weeks of gestation (WG), it is difficult to distinguish neurons from glia. Total neuronal number of each brain after 32WG is calculated from the total volume of the nuclei, section thickness and mean values of sampling areas in qach brain. The abbreviations are the same as those in Fig. 1.
follows: SA is the outlined area of the pontine nuclei minus the sum of areas consisting of descending cortical fibers and transverse pontine fibers. Volumes were calculated from the section thickness, number of sections containing the nuclei and the average of SA. Because of their large number, there were difficulties in ensuring that all neurons of the pontine nuclei were being evaluated. For microscopic measurements, 20 to 24 areas were systematically selected from each brain (one region on each side from each sectional level)9. This sampling method is a systemic sampling with a random start and is considered to be a quasi-random sampling method. Each sampling area covered 1.65 × 10-3 mm squared. Numbers, perimeters and areas of the pontine neurons in each sampling area were measured with the help of a combination of microscope, drawing tube (or camera lucida) and electronic planimeter. In each brain, total neuron numbers were calculated with the total volume of the nuclei, the see don thickness and the mean neuronal number per sampling area. A computer (PC-9801 VX2, NEC) was employed for storing data on-line, calculation and statistical analysis.
The criterion governing the selection of neurons was that the nucleus had to contain a distinct nucleolus (diameter: 1/~m) t. The probability of split nucleoli is considered to be extremely low. The accuracy of the instrumental system was 0.1 m m ± 1/2 digit for coordinates (an accidental error was less than 1% in some areas) according to the specifications of the electronic planimeter. Qualitative variables were examined including the amount and characteristics of Nissl bodies in the perikaryon and the nature of proximal neuronal processes, For an analysis of mature neurons, the circu. larity ratios of neurons, 4~AIL 2 (A = area in mm2; L = perimeter in mm), were calculated ~7, An immature neuron may present a circularity ratio value close to 1.0. A decrease in circularity ratio
PERCENT
50
,GO0-
AG|27WQ ~t'm692t~ thl 1724~
50
5O
AG|3 ~ .~t,lgA14id~ 4hi1419]
AS 35~ "~'r,,t04:k23~
50
50 A~ ZU | -"lk",t33~42~ t II~'s204~61me (ntWi
tn,343i gt',|tOi62~'~ (naqhl
~n,1341l tr.t52~41~
{h=11041
50 A~ ~v "1r,290t82~* M,3942126~ t1231t
400'
300 c
< ~00.
! i
l I [] []
L
F. an
i 100"
E
3
Fig. 3. Frequency histogram and mean values per/~m2 of neuronal areas (white columns: systematically selected medium-sized neurons; black columns: large neurons, see Figs. 6-8 for the distributed locations). The abbreviations are the same as those in Fig. 1.
15
Fig. 4. Cytoarchitectonic development of the pontine nuclei. A: 27WG; B: 32WG; C-a: large neurons, C-b: medium-sized neurons in 35WG; D-a: large neurons, D-b: medium-sized neurons in 40WG; E-a: large neurons, E-b: medium-sized neurons in 2M; F-a: large neurons, F-b: medium-sized neurons in 63Y. The scale bar for A-l, B-l, C, D, E and F - 100 ~tm; that for A-2, B-2 and insets 10 ~m. The abbreviations are the same as those in Fig. 1.
16
Circularity Ratio = 4 7r A / L = A " area, L" perimeter
1.0
0.5
i I I t tll
4~ mL'"ZU .D
L ~: . "
o medium-sized neurons e " large neurons •
27WG 32WG 35WG 40WG 2M
~~
63Y
Fig. 5. Mean circularity ratios of pontine neurons; open circles: systematically selected medium-sized neurons; closed circles: large neurons). The abbreviations are the same as those in Fig. 1. A, neuronal area in mmZ; L, neuronal perimeter in mm.
Fig. 7. Illustration of the distributed locations of large neurons at 40WO. The abbreviations are the same as those in Fig. 6.
may occur with the maturation of neuronal processes. During this study, it was apparent that the neurons of the pontinc nuclei did not develop uniformly, at least after 32WG. Some groups of large neurons appeared after 3SWG. Therefore, we separately examined the distribution, neuronal area and circularity ratio of these large neurons. A comparison of the mean value between medium-sized and large neurons was analyzed using the Student's
I
t.test.
RESULTS Serial sections of all fetus brains did not reveal any neuropathology.
Nuclear column length
....
.....
Fig. 6. Illustration of the distributed locations of large neurons at 3$WG. Islet-shaped black areas show the site of groups of large neurons araong the medium-sized neurons. Fbpot, fibrae pontis transversae (trsasverse pontine fibers); Fit, fasciculi longitudinafis (descending cortical fibers); Lm, lemniscus medialis (medial iem. n/scus); Pcm, pedunculus cerebellaris medius (middle cerebellar peduncle); L, left side; R, right side.
The orocaudal lengths of pontine nuclei in the 9 specimens were measured separately on each side. Their average lengths were 5.7 (right: 5.1; left: 6.3) m m at 16WG, 10.5 (10.8; 10.2) mm in 21WG, 9.0 (9.0; 9.0) mm at 23WG, 10.4 (9.9; 10.8) mm at 27WG, 9.3 (9.3; 9.3) m m at 32WG, 12.2 (11.4; 12.9) m m at 35WG, 13.0 (13.0; 13.0) mm at 40 W G , 10.7 (10.2; 11.2) m m at 2M and 23.8 (24.0; 23.6) m m at 63Y. The orocaudal lengths may be classified into three: (1) 16WG, (2) from 21WG to 2M and (3) 63Y. Based on these findings, it was concluded that the development of the nuclear orocaudal length accelerated after 16WG, 32WG and 2M.
17
Transverse sectional area of the nuclei
L
The m e a n value of transverse sectional areas of the pontine nuclei was calculated (see Materials and Methods) separately on each side. The average areas were 2.4 (right 2.4; left 2.4) mm 2 at 16WG, 6.1 (6.2; 6.1) mm 2 at 21WG, 6.0 (6.1; 5.9) mm 2 at 23WG, 7.1 (6.9; 7.2) mm 2 at 27WG, 5.9 (5.6; 6.2) mm 2 at 32WG, 9.8 (9.7; 10.0) mm 2 at 35WG, 16.1 (15.6; 16.6) mm 2 at 40WG, 17.0 (16.1; 17.8) mm 2 at 2M and 44.0 (43.5; 44.6) mm 2 at 63Y, respectively. Here, the developmental groups may be grouped into four: (1) 16WG, (2) from 21 to 32WG (3) from 35 W G to 2M and (4) 63Y. The results indicated that the development of the average sectional areas was accelerated in three steps: after 16WG, 32WG and 63Y.
R
Nuclear volume The value of pontine nuclear volume on the right and left sides of each specimen showed a close approximation with each other: any differences between the sides were found to be insignificant by the paired t-test. Each volume of the pontine nuclei is shown in Fig. 1. The volume increased stepwise with gestational and postnatal age except at 63Y when the volume was five times as great as a term newborn. Four groups are established, similar to those used for sectional area. The volume of the pontine nuclei accelerated stepwise after 16WG, 32WG and 63Y. Fig. 8. Illustration of the distributed locations of large neurons at 63Y. The abbreviations are the same as those in Fig. 6.
average 85WG :4.8"1"2.5% 40WG : 6.5:1: 3.5% 2M :4.8"1"2.8% 68Y : 1.9"1"1.6%
20%
(n-IS) (nffil2) (nffi18) (nffi20)
(O
2 c o) i
:
I
•
|
•
•
o%
I
I
; O
2M
Neuronal areas
Development of the cytoarchitecture I v
40WG
10 7 .
450 ~tm2.
m,
85 WG
Before 27WG, there were many neurons which were difficult to distinguish from glia. All the neurons contained a distinct nucleolus and were able to be clearly distinguished from gila at and after 27WG. The estimated total numbers of neurons are shown in Fig. 2. They ranged from 2.6 x 107 (fetus) to 3.6 x 107 (adult), except for one infant (32WG) whose count was 1.6 x
The distribution and mean values of neuronal areas of the pontine nuclei after 27WG are shown in Fig. 3 (white columns). From 27 to 32WG, the neuronal areas were narrowly distributed below 125/tm 2. After 32WG, the neuronal area became more widely distributed according to gestational and postnatal age. At 63Y, the distribution of neuronal areas was the widest, ranging from 150 to
c 10%
Neuronal numbers
63Y
Fig. 9. Scatter diagram between percentile numbers of large neurons in the pontine nuclei and age. The abbreviations are the same as those in Fig. 1. n ffi number.
Cytoarchitectonic development of the pontine nuclei was examined (Fig. 4) with special reference to the amount and character of Nissl bodies, the morphology of neuronal proximal processes and the cell arrangement. The Nissl bodies were hardly visible between 16 and
18
32WG. After 32WG, Nissi bodies gradually became visible and increased significantly after birth. The circularity ratios of neurons were similar to each other before 40WG (Fig. 5; open circles). The ratio was markedly decreased in 2M and 63Y.
Development of large neurons The groups of large cells were definitely observed at and after 35WG. The locations of large neurons are illustrated in Figs. 6-8 (35WG, 40WG, and 63Y). These neurons fon~'~ed several small islets among the mediumsized neurons. The percentage of large neurons among all neurons was calculated to be as follo~vs: 4.3 __ 2.5% (n - 18) at 35WG, 6.5 -+ 3.5% (n - 12) at 40WG, 4.3 - 2.8% (n - 18) at 2M and 1.9 -+ 1.6% (n - 20) at 63Y (Fig. 9). In general neurons were distributed mainly in the ventrolateral portion of the pontine nuclei around the longitudinal descending tracts. At 35WG, 40WG and 2M, the large neuron islets were w~dei~, distributed as compared to the adult where they were few and present as smaller islets. The distributions and mean values of large neuron areas are given in Fig. 3 (black columns). These neurons were larger for each brain (P < 0.005), and their size distdbt~tion was wider for the large neurons than the more common mediumsized neurons. The circularity ratios of these neurons also showed statistically smaller values than those of the other groups except for the 2M (P < 0.005; Fig. 5, closed circles). DISCUSSION The qualitative and quantitative development of the pontine nuclei was analyzed utilizing an electronic planimeter combined with a computer. The accuracy of measurement was affected by several sources of error. It is generally agreed that paraffin embedding causes a considerable shrinkage of sections. To avoid such shrinkage, we adopted the method of celloidin embedding after secondary fixation 7. The brains obtained after birth appeared to have developed normally. The short duration of illness and the cause of death did not influence normal brain maturation. This conclusion is supported by our mulli-variable data including pontine nuclear volumes, neuronal areas, circularity ratios of neurons and characteristics of Nissl bodies. In the brainstem, the pontine nuclei are very unique in view of having a single source or origin, a unique pathway and selective destinations. The normal cytogenesis and histogenesis of pontine nuclei were initially studied by Essick4-6 in man, and subsequently by others in the chick t2 and mouse 26. In these studies, pontine neurons
are generated in a specialized sector of the subventricular zone called the rhombic lip which is one of transitory developmental organs. The migrating neurons of the pontine nuclei, inferior olivary nucleus and arcuate nucleus constitute a distinct sheet of tissue present transiently from 8 to 20WG. Essick~ named it the corpus pontobulbare. More recently, the system of the migration of pontine neurons in developing rhesus monkey was studied combining with both [3H]thymidine autoradiography and electron microscopy by Rakic23"e4. In those studies, the migration of neurons from rhombic lip to the gray nucleus of the pons in the brain stem did not take a gliophilic mode, but a neurophilic mode which was an exceptional pattern in the central nervous system. Altman and Bayer 3 reported that pontine neurons arose later than those of the inferior olive originating from the ne~roepithelium of the rhombic lip of the fourth ventricle. The convoluted form of the inferior olivary nucleus is established during the period from 15 to 22WG. Cells destined for the pontine nuclei arise and differentiate somewhat later than the cells of olivary nuclei ~3. They do not appear to differentiate until they reach their final destination adjacent to the longitudinally oriented fibers of the cerebral peduncle. At the beginning of the 6th lunar month, the genesis of neurons in the rhombic lip and their migration outward gradually cease. The corpus pontobulbare disappears in this stage. Further development in the formation of the pons and medulla oblongata derives from the growth of intrinsic and extrinsic fiber systems, growth of cell bodies and dendrites, gliogenesis, and myelination t3. Mihailoff et al. t9 demonstrated in the rat that an intense wave of synaptic proliferation was observed during the postnatal period extending from day 6 to 13 after birth. The results of our study can be summarized as follows: (1) the pontine nuclear volume developed in 4 steps: 16WG, 21-32WG, 35WG-2M and 63Y, (2) the pontine neuronal number did not indicate the developmental stage, (3) the pontine neuronal sizes developed in and after 27WG in three steps: 27-32WG, 35WG-2M and 63Y, (4) large neurons appeared after 32WG and developed abruptly from the view point of areal distribution and circularity ratio. Although the total estimated number of pontine neurons was around 3.0 x 107 except for the infant of 32WG, the volumes of the nuclei increased and the neuronal areas were widely distributed in three steps: 27-32WG, 35WG-2M and 63Y. At 63Y, the nuclear volume remarkably increased and the neuronal areas were widely distributed. This suggests that postnatal development may be also indispensable for the pontine nuclei. Qualitatively, the Nissl bodies increased in amount and with the development of cell processes, the
19 configuration of neuronal somata gradually changed in shape from round to oval, triangular or multipolar. This was confirmed by quantitation showing a gradual decrease of neuronal circularity ratios with age. These results suggest that the development of pontine nuclei may accelerate after 32WG and more rapidly at 2M. Ontogenetically, the corticopontocerebeilar system develops early in the human brain but is far behind the other parts of the brainstem phylogenetically except for the pyramidal system. This system was investigated in the rat and found to develop primarily during the postnatal period 2. At the end of the 2rid lunar month, the ventricular zone of the 4th ventricle is already exhausted, and virtually all the neurons of the cranial nerve nuclei have been generated t3. Comparing our results with other data from hypoglossal nucleus 2t, motor trigeminal nucleus t° and mesencephalic trigeminal nucleus n, it appears that pontine nuclei develop during later stages of gestation than that of motor or sensory cranial nuclei. Although our data suggest that postnatal changes may occur, the number of subjects in this study was insufficient to examine adequately the growth phases of postnatal maturation. Thus, it will be necessary to conduct further developmental studies on postnatal brains using similar procedures. The existence of large neurons forming small groups within the pontine gray matter has not been reported. Large neurons were more clearly observed in perinatai brains compared to the adult brain. The changes in size of large neuron groups suggests that the perinatal period is also important in the development of human pontine nuclei. However, these data must be expanded after careful observation at wider collections of brain specimens including more postnatal infants. REFERENCES 1 Abercrombie, M., Estimation of nuclear population from microtome sections, Anat. Rec., 94 (1946) 239-247. 2 Adams, C.E., Mihailoff, G.A. and Woodward, D.J., A transient component of the developing corticospinal tract arises in visual cortex, Neurosci. Lett., 36 (1983) 243-248. 3 Altman, J. and Bayer, S.A., Prenatal development of the cerebellar system in the rat. II. Cytogenesis and histogenesis of the inferior live, pontine gray, and the precerebellar reticular nuclei, J. Comp. Neurol., 1979 (1978) 49-76. 4 Essick, C.R., The corpus ponto-bulbare -- a hitherto undescribed nuclear mass in the human hindbrain, Am. J. Anat., 7 (1907) 119-135. 5 Essick, C.R., On the embryology of the corpus ponto-bulbare and its relationship to the development of the pons, Anat. Rec., 3 (1909) 254-257. 6 Essick, C.R., The development of the nuclei pontis and the nucleus arcuatus in man, Am. J. Anat., 13 (1912) 25-54. 7 Goto, N., Discriminative staining methods for the nervous system: luxol fast blue-periodic acid Schiff-hematoxylin triple stain and subsidiary staining methods, Stain Technol., 62 (1987) 305-315. 8 Graybiel, A.M., Ragsdale, Jr., C.W., Yoneoka, E.S., et al., An immunohistochemical study of enkephalins and other neuropeptides in the striatum of the cat with evideace that the opiate
The difference in neuronal size may be dependent upon the lengths of dendrites. This assumption is supported by the results of Jones 14't5 who demonstrated that the neuronal size in the sensory-motor cortex was related to the terminal distribution of fibers and that large cells had long apical dendritic projections. The presence of large neurons in other nuclei, such as the striatum 25 or putamen 2°, has been also reported. Kemp and Powell ~6 classified neurons in the cat striatum into six groups: one group of large cells, four groups of medium-sized cells, and one group of small cells. Graybiel s reported that large neurons in the rat striatum are considered to be intrinsic neurons containing acetylcholine as a neurotransmitter. The determination of neurochemical differences among the pontine nuclei should be evaluated employing other methods, such as immunohistological staining. Studies on human brain development are by necessity limited. However, we have found that limited material can be maximally utilized using the serial section approach employing the method of secondary fixation with chromic acid, celloidin embedding and computer-assisted electronic planimeter combined with microscopic examination.
Acknowledgments. We are grateful to Prof. K. Maekawa (De. partment of Pediatrics, Jikei University School of Medicine) for his encouragement and to Prof. L.E. Becker (Prof. and Head, Division of Neuropathology, University of Toronto, Canada) for his useful comments and suggestions toward the draft of this manuscript. We are also grateful to the followingcolleagues whose assistance helped bring this investigation to completion: Ms. R, Takagi, neurohistological technology; Dr. A. Okada, computer programming; and Dr. A. Komatsu who provided us with some of the ~ormal brains obtained on pathological postmortem examinations.
peptides are arranged to form mosaic patterns in register with the compartments visible by acetylcholinesterase staining, Neu. roscience, 6 (1981) 377-397. 9 Gundersen, H.J.G. and Jensen, E.B., The efficiency of systematic sampling in stereology and its prediction, J. Microsc., 147 (1987) 229-263. 10 Hamano, S., Goto, N. and Nara, T., Development of the human motor trigeminal nucleus: a morphometric study, Pediatr. Neurosci., 14 (1988) 230-235. 11 Hamano, S., Goto, N., Nara, T., Yamaguchi, K. and Maekawa, K, Development of the human mesencephalic trigeminal nuclzus: a Morphometric Study, Dev. Med. Child Neurol., 32 (1990) 621-628. 12 Harkmark, W., Cell migrations from the rhombic lip to the inferior olive, the nucleus raphe and the pons. A morphological and experimental investigation on chick embryos, J. Comp. Neurol., 100 (1954) 115-209. 13 Haymaker, W. and Adams, R.D. (Eds.), Histology and Histo. pathology of the Nervous System, C.C. Thomas, Springfield. 14 Jones, E.G., Coulter, J.D., Burton, H. and Porter, R., Cells of origin and terminal distribution of corticostriatal fibers arising in the sensory-motor cortex of monkeys, J. Comp. Neurol., 173 (1977) 53-80. 15 Jones, E.G. and Wise, S.P., Size, laminar and columnar distribution of efferent cells in the sensory-motor cortex of monkeys,
20 J. Comp. Neurol., 175 (1977)391-438. 16 Kemp, J,M. and Poweli, T.P.S., The structure of the caudate nucleus of the cat: light and electron microscopy, Phil. Trans. R. Soc. Set. B., 262 (1971) 383-401. 17 Kojima, T., Circularity ratio, Okaiimas Folia Anat. lpn., 48 (1971) 153-162. 18 Langman, J. and Haden, C.C., Formalin and migration of neuroblasts in the spinal cord of chick embryo, J. Comp. Neurol., 138 (1970). 19 Mihailoff, G.A., Adams, C.E. and Woodward, D.J., An autoradiographic study of the postnatal development of sensorimotor and visual components of the coorticopontine system, J. Comp. Neurol., 222 (1984) 116-127. 20 Nakae, Y., Goto, N. and Nara, T., Development of the human putamen: a morphometric study, Acta Anat., 137 (1990) 272277. 21 Nara, T., Goto, N. and Yamaguchi, K., Development of the
22
23 24 25 26
Human Hypoglossal Nucleus, Dev. Neurosci., 11 (1989) 212220. Nornes, H.O. and Das, G.D., Temporal pattern of neurogenesis in spinal cord of rat. I. An autoradiographic study -- Time and sites of origin and migration and settling patterns of neuroblasts, Brain Res., 73 (1974) 121-138. Rakic, P., Contact regulation of neuronal migration. In G.M.E. Edelman and J.P. Tiery, (Eds.), The Cell in Contact, Wiley, New York, pp. 67-91. Rakic, P., Principles of neural cell migration, Experientia, 46 (1990) 882-891. Schr6der, K.E, Hopf, A. and Thorner, G., Morphometrischstatische Strukturanalysen des Striatum, Pallidum, und Nucleus Subthalamicus bein Menschen,/. Hirnforsch, 16 (!975) 333-350. Taber and Pierce, E., Histogenesis of the nuclei griseum pontis, corporis pontobulbaris and reticularis tegmenti pontis (Bechterew) in the mouse, I. Comp. Neurol,, 126 (1966) 219-240.
Developmental Brain Research, 65 (1992) 13-20 ~) Elsevier Science Publishers B.V. All rights reserved. 0165-3806192/$05.00 BRESD 51385
Development of the human pontine nuclei: a morphometric study Hidetsugu
N o z a k i 1'2, N o b o r u
Goto 3 and Takahiro
Nara 2
ZDepartment of Pediatrics, Jikei University School of Medicine, Tokyo (Japan), ZDivision of Neurology, Saitama Children's Medical Center, Saitama (Japan) and 3Department of Anatomy, Showa University School of Medicine, Tokyo (Japan) (Accepted 29 August 1991)
Key words: Pontine nuclei; Morphometry; Human brain; Brain development; Large neuron
The morphometric development of the pontine nuclei in the human fetus from 16 to 40 gestational weeks, in a 2-month-old infant and in a 63-year-old adult were examined employing a serial celloidin section method and computer assisted electronic planimeter. The results of our study can be summarized as follows: (1) the development of the human pontine nuclei accelerated in volume after 32 gestational weeks and continued after birth, (2) neuron numbers remained relatively constant after 27 gestationai weeks. It was difficult to clearly distinguish neurons from gila before 27 gestationai weeks. The total estimated neuronal numbers were not indicative of the gestational stages in infants 27 gestationai weeks and older, (3) individual neurons appeared to continue to develop after 32 gestational weeks in accordance with size, distribution and circularity ratio, (4) many islet-shaped groups of large neurons appeared and were scattered throughout the oontine nuclei after 32 gestational weeks. INTRODUCTION The pontine nuclei are one of the important relay nuclei which have an apparent significance in the conduction of impulses from almost all parts of the cerebral cortices to the cerebellum (neocerebellum). The normal cytogenesis ,and histogenesis of this nucleus was initially studied by Essick 4-e in man, pig and rabbit, and subsequently by others in the chick t2 and mouse 2e. More recently, Rakic ~'24 reported the neurophilic mode of migration of pontine neurons in developing rhesus monkey. Altman and Bayer 3 reported that the pontine neurons arise later than those of the inferior olive originating from the neuroepithelium of the rhombic lip of the fourth Ventricle. T hey do not appear to differentiate until they reach their final destination adjacent to the longitudinally oriented fibers of the cerebral peduncle. Mihailoff et a1.19 demonstrated in the rat that an intense wave of synaptic proliferation was observed during the postnatal period. We think the morphometric and developing studies on the human pontine nuclei still have not been fully analyzed, especially focussing on the perinatal period. In this study, the development of the pontine nuclei in the human fetus, infant and adult was examined employing a serial section method combined with an electronic planimeter to determine morphometric cytoarchitectonic parameters.
MATERIALS AND METHODS Nine human brains were used for this study. Seven were from fetuses of 16, 21, 23, 27, 32, 35 and 40 weeks of gestation (WG). The other two were from a 2-month-old infant (2M) and a 63-yearold adult (63Y). Five of the fetal brains were obtained by hysterectomy performed because of myoma uteri (16WG) or from spontaneous abortions resulting from cervical incompetence (16, 21, 23 and 27WG). The live born fetuses obtained at 32, 35 and 40WG died respectively of sepsis, erythroblastosis fetalis and massive aspiration. There was no evidence of macroscopic abnormalities externally and internally, including the central nervous system. The crown-rump length and body weight were within normal ranges. After fixation in formalin (3.7% formaldehyde solution), the whole brains from the fetuses at 16, 21, 23 and 27WG and the brain stems with the cerebelliT(of the remaining 5 brains) were immersed in a secondary fixative, which was composed of 5% potassium dichromate and 5% potassium chromate (1:4, v:v). The secondary fixative was changed several Umes over a 3-week period (for the first two weeks the fixative was maintai~ed at room temperature, for the last week at 37"C) en bloc for the fetuses and after removal of the cerebrum for the others. The brains were mounted in celloidin after washing in running water and dehydrated with alcohol. The celioidin blocks were cut into 30./~m-(fetai and neonate brains) or 40-pm-thick (infant and adult brains) serial transverse sections. All sections were numbered with India ink. Every tenth section was stained by the KI0ver-Barrera (K-B) method. Approximately 10% shrinkage of the sections occurred during the entire procedure. The cytoarchitectonic development of the pontine nuclei was studied using both qualitative and quantitative variables. For the volumetry of structures, ten to twelve sections containing pontine nuclei were chosen at equal intervals. For each brain, enlarged negative prints were made from the selected sections utilizing the same magnification (16-22 times). In each print, areas of pontine regions were measured with an optical electronic planimeter (Digiplan, Kontron Co.). The sectional area (SA) of the pontine nuclei was determined as
Correspondence: H. Nozaki, Department of Pediatrics, Jikei University School of Medicine, 25-8 Nishi-shinbashi 3, Minato-ku, Tokyo 105, Japan. Fax: (81) 03-3435-8665.
14
4.0 ¸ 2000 1000: 3.0"
E .c
r~
x
::' 500. >o
k, 2.0, E
I6WG' 21WG 23WG 27WG 32WG 35WG 40WG 2M" HY'"
Fig. 1. Total volume of pontine nuclei (right and left) in mm 3. The volume is calculated from the section thickness, number of sections containing the nucleus and the average area per section. WG, gestational age in weeks; M, post-natal age in months; Y, years.
2/WG* 32WG 35WG 40WG 2M** 63Y***' Fig. 2. Neuronal number of the pontine nuclei. Before 27 weeks of gestation (WG), it is difficult to distinguish neurons from glia. Total neuronal number of each brain after 32WG is calculated from the total volume of the nuclei, section thickness and mean values of sampling areas in qach brain. The abbreviations are the same as those in Fig. 1.
follows: SA is the outlined area of the pontine nuclei minus the sum of areas consisting of descending cortical fibers and transverse pontine fibers. Volumes were calculated from the section thickness, number of sections containing the nuclei and the average of SA. Because of their large number, there were difficulties in ensuring that all neurons of the pontine nuclei were being evaluated. For microscopic measurements, 20 to 24 areas were systematically selected from each brain (one region on each side from each sectional level)9. This sampling method is a systemic sampling with a random start and is considered to be a quasi-random sampling method. Each sampling area covered 1.65 × 10-3 mm squared. Numbers, perimeters and areas of the pontine neurons in each sampling area were measured with the help of a combination of microscope, drawing tube (or camera lucida) and electronic planimeter. In each brain, total neuron numbers were calculated with the total volume of the nuclei, the see don thickness and the mean neuronal number per sampling area. A computer (PC-9801 VX2, NEC) was employed for storing data on-line, calculation and statistical analysis.
The criterion governing the selection of neurons was that the nucleus had to contain a distinct nucleolus (diameter: 1/~m) t. The probability of split nucleoli is considered to be extremely low. The accuracy of the instrumental system was 0.1 m m ± 1/2 digit for coordinates (an accidental error was less than 1% in some areas) according to the specifications of the electronic planimeter. Qualitative variables were examined including the amount and characteristics of Nissl bodies in the perikaryon and the nature of proximal neuronal processes, For an analysis of mature neurons, the circu. larity ratios of neurons, 4~AIL 2 (A = area in mm2; L = perimeter in mm), were calculated ~7, An immature neuron may present a circularity ratio value close to 1.0. A decrease in circularity ratio
PERCENT
50
,GO0-
AG|27WQ ~t'm692t~ thl 1724~
50
5O
AG|3 ~ .~t,lgA14id~ 4hi1419]
AS 35~ "~'r,,t04:k23~
50
50 A~ ZU | -"lk",t33~42~ t II~'s204~61me (ntWi
tn,343i gt',|tOi62~'~ (naqhl
~n,1341l tr.t52~41~
{h=11041
50 A~ ~v "1r,290t82~* M,3942126~ t1231t
400'
300 c
< ~00.
! i
l I [] []
L
F. an
i 100"
E
3
Fig. 3. Frequency histogram and mean values per/~m2 of neuronal areas (white columns: systematically selected medium-sized neurons; black columns: large neurons, see Figs. 6-8 for the distributed locations). The abbreviations are the same as those in Fig. 1.
15
Fig. 4. Cytoarchitectonic development of the pontine nuclei. A: 27WG; B: 32WG; C-a: large neurons, C-b: medium-sized neurons in 35WG; D-a: large neurons, D-b: medium-sized neurons in 40WG; E-a: large neurons, E-b: medium-sized neurons in 2M; F-a: large neurons, F-b: medium-sized neurons in 63Y. The scale bar for A-l, B-l, C, D, E and F - 100 ~tm; that for A-2, B-2 and insets 10 ~m. The abbreviations are the same as those in Fig. 1.
16
Circularity Ratio = 4 7r A / L = A " area, L" perimeter
1.0
0.5
i I I t tll
4~ mL'"ZU .D
L ~: . "
o medium-sized neurons e " large neurons •
27WG 32WG 35WG 40WG 2M
~~
63Y
Fig. 5. Mean circularity ratios of pontine neurons; open circles: systematically selected medium-sized neurons; closed circles: large neurons). The abbreviations are the same as those in Fig. 1. A, neuronal area in mmZ; L, neuronal perimeter in mm.
Fig. 7. Illustration of the distributed locations of large neurons at 40WO. The abbreviations are the same as those in Fig. 6.
may occur with the maturation of neuronal processes. During this study, it was apparent that the neurons of the pontinc nuclei did not develop uniformly, at least after 32WG. Some groups of large neurons appeared after 3SWG. Therefore, we separately examined the distribution, neuronal area and circularity ratio of these large neurons. A comparison of the mean value between medium-sized and large neurons was analyzed using the Student's
I
t.test.
RESULTS Serial sections of all fetus brains did not reveal any neuropathology.
Nuclear column length
....
.....
Fig. 6. Illustration of the distributed locations of large neurons at 3$WG. Islet-shaped black areas show the site of groups of large neurons araong the medium-sized neurons. Fbpot, fibrae pontis transversae (trsasverse pontine fibers); Fit, fasciculi longitudinafis (descending cortical fibers); Lm, lemniscus medialis (medial iem. n/scus); Pcm, pedunculus cerebellaris medius (middle cerebellar peduncle); L, left side; R, right side.
The orocaudal lengths of pontine nuclei in the 9 specimens were measured separately on each side. Their average lengths were 5.7 (right: 5.1; left: 6.3) m m at 16WG, 10.5 (10.8; 10.2) mm in 21WG, 9.0 (9.0; 9.0) mm at 23WG, 10.4 (9.9; 10.8) mm at 27WG, 9.3 (9.3; 9.3) m m at 32WG, 12.2 (11.4; 12.9) m m at 35WG, 13.0 (13.0; 13.0) mm at 40 W G , 10.7 (10.2; 11.2) m m at 2M and 23.8 (24.0; 23.6) m m at 63Y. The orocaudal lengths may be classified into three: (1) 16WG, (2) from 21WG to 2M and (3) 63Y. Based on these findings, it was concluded that the development of the nuclear orocaudal length accelerated after 16WG, 32WG and 2M.
17
Transverse sectional area of the nuclei
L
The m e a n value of transverse sectional areas of the pontine nuclei was calculated (see Materials and Methods) separately on each side. The average areas were 2.4 (right 2.4; left 2.4) mm 2 at 16WG, 6.1 (6.2; 6.1) mm 2 at 21WG, 6.0 (6.1; 5.9) mm 2 at 23WG, 7.1 (6.9; 7.2) mm 2 at 27WG, 5.9 (5.6; 6.2) mm 2 at 32WG, 9.8 (9.7; 10.0) mm 2 at 35WG, 16.1 (15.6; 16.6) mm 2 at 40WG, 17.0 (16.1; 17.8) mm 2 at 2M and 44.0 (43.5; 44.6) mm 2 at 63Y, respectively. Here, the developmental groups may be grouped into four: (1) 16WG, (2) from 21 to 32WG (3) from 35 W G to 2M and (4) 63Y. The results indicated that the development of the average sectional areas was accelerated in three steps: after 16WG, 32WG and 63Y.
R
Nuclear volume The value of pontine nuclear volume on the right and left sides of each specimen showed a close approximation with each other: any differences between the sides were found to be insignificant by the paired t-test. Each volume of the pontine nuclei is shown in Fig. 1. The volume increased stepwise with gestational and postnatal age except at 63Y when the volume was five times as great as a term newborn. Four groups are established, similar to those used for sectional area. The volume of the pontine nuclei accelerated stepwise after 16WG, 32WG and 63Y. Fig. 8. Illustration of the distributed locations of large neurons at 63Y. The abbreviations are the same as those in Fig. 6.
average 85WG :4.8"1"2.5% 40WG : 6.5:1: 3.5% 2M :4.8"1"2.8% 68Y : 1.9"1"1.6%
20%
(n-IS) (nffil2) (nffi18) (nffi20)
(O
2 c o) i
:
I
•
|
•
•
o%
I
I
; O
2M
Neuronal areas
Development of the cytoarchitecture I v
40WG
10 7 .
450 ~tm2.
m,
85 WG
Before 27WG, there were many neurons which were difficult to distinguish from glia. All the neurons contained a distinct nucleolus and were able to be clearly distinguished from gila at and after 27WG. The estimated total numbers of neurons are shown in Fig. 2. They ranged from 2.6 x 107 (fetus) to 3.6 x 107 (adult), except for one infant (32WG) whose count was 1.6 x
The distribution and mean values of neuronal areas of the pontine nuclei after 27WG are shown in Fig. 3 (white columns). From 27 to 32WG, the neuronal areas were narrowly distributed below 125/tm 2. After 32WG, the neuronal area became more widely distributed according to gestational and postnatal age. At 63Y, the distribution of neuronal areas was the widest, ranging from 150 to
c 10%
Neuronal numbers
63Y
Fig. 9. Scatter diagram between percentile numbers of large neurons in the pontine nuclei and age. The abbreviations are the same as those in Fig. 1. n ffi number.
Cytoarchitectonic development of the pontine nuclei was examined (Fig. 4) with special reference to the amount and character of Nissl bodies, the morphology of neuronal proximal processes and the cell arrangement. The Nissl bodies were hardly visible between 16 and
18
32WG. After 32WG, Nissi bodies gradually became visible and increased significantly after birth. The circularity ratios of neurons were similar to each other before 40WG (Fig. 5; open circles). The ratio was markedly decreased in 2M and 63Y.
Development of large neurons The groups of large cells were definitely observed at and after 35WG. The locations of large neurons are illustrated in Figs. 6-8 (35WG, 40WG, and 63Y). These neurons fon~'~ed several small islets among the mediumsized neurons. The percentage of large neurons among all neurons was calculated to be as follo~vs: 4.3 __ 2.5% (n - 18) at 35WG, 6.5 -+ 3.5% (n - 12) at 40WG, 4.3 - 2.8% (n - 18) at 2M and 1.9 -+ 1.6% (n - 20) at 63Y (Fig. 9). In general neurons were distributed mainly in the ventrolateral portion of the pontine nuclei around the longitudinal descending tracts. At 35WG, 40WG and 2M, the large neuron islets were w~dei~, distributed as compared to the adult where they were few and present as smaller islets. The distributions and mean values of large neuron areas are given in Fig. 3 (black columns). These neurons were larger for each brain (P < 0.005), and their size distdbt~tion was wider for the large neurons than the more common mediumsized neurons. The circularity ratios of these neurons also showed statistically smaller values than those of the other groups except for the 2M (P < 0.005; Fig. 5, closed circles). DISCUSSION The qualitative and quantitative development of the pontine nuclei was analyzed utilizing an electronic planimeter combined with a computer. The accuracy of measurement was affected by several sources of error. It is generally agreed that paraffin embedding causes a considerable shrinkage of sections. To avoid such shrinkage, we adopted the method of celloidin embedding after secondary fixation 7. The brains obtained after birth appeared to have developed normally. The short duration of illness and the cause of death did not influence normal brain maturation. This conclusion is supported by our mulli-variable data including pontine nuclear volumes, neuronal areas, circularity ratios of neurons and characteristics of Nissl bodies. In the brainstem, the pontine nuclei are very unique in view of having a single source or origin, a unique pathway and selective destinations. The normal cytogenesis and histogenesis of pontine nuclei were initially studied by Essick4-6 in man, and subsequently by others in the chick t2 and mouse 26. In these studies, pontine neurons
are generated in a specialized sector of the subventricular zone called the rhombic lip which is one of transitory developmental organs. The migrating neurons of the pontine nuclei, inferior olivary nucleus and arcuate nucleus constitute a distinct sheet of tissue present transiently from 8 to 20WG. Essick~ named it the corpus pontobulbare. More recently, the system of the migration of pontine neurons in developing rhesus monkey was studied combining with both [3H]thymidine autoradiography and electron microscopy by Rakic23"e4. In those studies, the migration of neurons from rhombic lip to the gray nucleus of the pons in the brain stem did not take a gliophilic mode, but a neurophilic mode which was an exceptional pattern in the central nervous system. Altman and Bayer 3 reported that pontine neurons arose later than those of the inferior olive originating from the ne~roepithelium of the rhombic lip of the fourth ventricle. The convoluted form of the inferior olivary nucleus is established during the period from 15 to 22WG. Cells destined for the pontine nuclei arise and differentiate somewhat later than the cells of olivary nuclei ~3. They do not appear to differentiate until they reach their final destination adjacent to the longitudinally oriented fibers of the cerebral peduncle. At the beginning of the 6th lunar month, the genesis of neurons in the rhombic lip and their migration outward gradually cease. The corpus pontobulbare disappears in this stage. Further development in the formation of the pons and medulla oblongata derives from the growth of intrinsic and extrinsic fiber systems, growth of cell bodies and dendrites, gliogenesis, and myelination t3. Mihailoff et al. t9 demonstrated in the rat that an intense wave of synaptic proliferation was observed during the postnatal period extending from day 6 to 13 after birth. The results of our study can be summarized as follows: (1) the pontine nuclear volume developed in 4 steps: 16WG, 21-32WG, 35WG-2M and 63Y, (2) the pontine neuronal number did not indicate the developmental stage, (3) the pontine neuronal sizes developed in and after 27WG in three steps: 27-32WG, 35WG-2M and 63Y, (4) large neurons appeared after 32WG and developed abruptly from the view point of areal distribution and circularity ratio. Although the total estimated number of pontine neurons was around 3.0 x 107 except for the infant of 32WG, the volumes of the nuclei increased and the neuronal areas were widely distributed in three steps: 27-32WG, 35WG-2M and 63Y. At 63Y, the nuclear volume remarkably increased and the neuronal areas were widely distributed. This suggests that postnatal development may be also indispensable for the pontine nuclei. Qualitatively, the Nissl bodies increased in amount and with the development of cell processes, the
19 configuration of neuronal somata gradually changed in shape from round to oval, triangular or multipolar. This was confirmed by quantitation showing a gradual decrease of neuronal circularity ratios with age. These results suggest that the development of pontine nuclei may accelerate after 32WG and more rapidly at 2M. Ontogenetically, the corticopontocerebeilar system develops early in the human brain but is far behind the other parts of the brainstem phylogenetically except for the pyramidal system. This system was investigated in the rat and found to develop primarily during the postnatal period 2. At the end of the 2rid lunar month, the ventricular zone of the 4th ventricle is already exhausted, and virtually all the neurons of the cranial nerve nuclei have been generated t3. Comparing our results with other data from hypoglossal nucleus 2t, motor trigeminal nucleus t° and mesencephalic trigeminal nucleus n, it appears that pontine nuclei develop during later stages of gestation than that of motor or sensory cranial nuclei. Although our data suggest that postnatal changes may occur, the number of subjects in this study was insufficient to examine adequately the growth phases of postnatal maturation. Thus, it will be necessary to conduct further developmental studies on postnatal brains using similar procedures. The existence of large neurons forming small groups within the pontine gray matter has not been reported. Large neurons were more clearly observed in perinatai brains compared to the adult brain. The changes in size of large neuron groups suggests that the perinatal period is also important in the development of human pontine nuclei. However, these data must be expanded after careful observation at wider collections of brain specimens including more postnatal infants. REFERENCES 1 Abercrombie, M., Estimation of nuclear population from microtome sections, Anat. Rec., 94 (1946) 239-247. 2 Adams, C.E., Mihailoff, G.A. and Woodward, D.J., A transient component of the developing corticospinal tract arises in visual cortex, Neurosci. Lett., 36 (1983) 243-248. 3 Altman, J. and Bayer, S.A., Prenatal development of the cerebellar system in the rat. II. Cytogenesis and histogenesis of the inferior live, pontine gray, and the precerebellar reticular nuclei, J. Comp. Neurol., 1979 (1978) 49-76. 4 Essick, C.R., The corpus ponto-bulbare -- a hitherto undescribed nuclear mass in the human hindbrain, Am. J. Anat., 7 (1907) 119-135. 5 Essick, C.R., On the embryology of the corpus ponto-bulbare and its relationship to the development of the pons, Anat. Rec., 3 (1909) 254-257. 6 Essick, C.R., The development of the nuclei pontis and the nucleus arcuatus in man, Am. J. Anat., 13 (1912) 25-54. 7 Goto, N., Discriminative staining methods for the nervous system: luxol fast blue-periodic acid Schiff-hematoxylin triple stain and subsidiary staining methods, Stain Technol., 62 (1987) 305-315. 8 Graybiel, A.M., Ragsdale, Jr., C.W., Yoneoka, E.S., et al., An immunohistochemical study of enkephalins and other neuropeptides in the striatum of the cat with evideace that the opiate
The difference in neuronal size may be dependent upon the lengths of dendrites. This assumption is supported by the results of Jones 14't5 who demonstrated that the neuronal size in the sensory-motor cortex was related to the terminal distribution of fibers and that large cells had long apical dendritic projections. The presence of large neurons in other nuclei, such as the striatum 25 or putamen 2°, has been also reported. Kemp and Powell ~6 classified neurons in the cat striatum into six groups: one group of large cells, four groups of medium-sized cells, and one group of small cells. Graybiel s reported that large neurons in the rat striatum are considered to be intrinsic neurons containing acetylcholine as a neurotransmitter. The determination of neurochemical differences among the pontine nuclei should be evaluated employing other methods, such as immunohistological staining. Studies on human brain development are by necessity limited. However, we have found that limited material can be maximally utilized using the serial section approach employing the method of secondary fixation with chromic acid, celloidin embedding and computer-assisted electronic planimeter combined with microscopic examination.
Acknowledgments. We are grateful to Prof. K. Maekawa (De. partment of Pediatrics, Jikei University School of Medicine) for his encouragement and to Prof. L.E. Becker (Prof. and Head, Division of Neuropathology, University of Toronto, Canada) for his useful comments and suggestions toward the draft of this manuscript. We are also grateful to the followingcolleagues whose assistance helped bring this investigation to completion: Ms. R, Takagi, neurohistological technology; Dr. A. Okada, computer programming; and Dr. A. Komatsu who provided us with some of the ~ormal brains obtained on pathological postmortem examinations.
peptides are arranged to form mosaic patterns in register with the compartments visible by acetylcholinesterase staining, Neu. roscience, 6 (1981) 377-397. 9 Gundersen, H.J.G. and Jensen, E.B., The efficiency of systematic sampling in stereology and its prediction, J. Microsc., 147 (1987) 229-263. 10 Hamano, S., Goto, N. and Nara, T., Development of the human motor trigeminal nucleus: a morphometric study, Pediatr. Neurosci., 14 (1988) 230-235. 11 Hamano, S., Goto, N., Nara, T., Yamaguchi, K. and Maekawa, K, Development of the human mesencephalic trigeminal nuclzus: a Morphometric Study, Dev. Med. Child Neurol., 32 (1990) 621-628. 12 Harkmark, W., Cell migrations from the rhombic lip to the inferior olive, the nucleus raphe and the pons. A morphological and experimental investigation on chick embryos, J. Comp. Neurol., 100 (1954) 115-209. 13 Haymaker, W. and Adams, R.D. (Eds.), Histology and Histo. pathology of the Nervous System, C.C. Thomas, Springfield. 14 Jones, E.G., Coulter, J.D., Burton, H. and Porter, R., Cells of origin and terminal distribution of corticostriatal fibers arising in the sensory-motor cortex of monkeys, J. Comp. Neurol., 173 (1977) 53-80. 15 Jones, E.G. and Wise, S.P., Size, laminar and columnar distribution of efferent cells in the sensory-motor cortex of monkeys,
20 J. Comp. Neurol., 175 (1977)391-438. 16 Kemp, J,M. and Poweli, T.P.S., The structure of the caudate nucleus of the cat: light and electron microscopy, Phil. Trans. R. Soc. Set. B., 262 (1971) 383-401. 17 Kojima, T., Circularity ratio, Okaiimas Folia Anat. lpn., 48 (1971) 153-162. 18 Langman, J. and Haden, C.C., Formalin and migration of neuroblasts in the spinal cord of chick embryo, J. Comp. Neurol., 138 (1970). 19 Mihailoff, G.A., Adams, C.E. and Woodward, D.J., An autoradiographic study of the postnatal development of sensorimotor and visual components of the coorticopontine system, J. Comp. Neurol., 222 (1984) 116-127. 20 Nakae, Y., Goto, N. and Nara, T., Development of the human putamen: a morphometric study, Acta Anat., 137 (1990) 272277. 21 Nara, T., Goto, N. and Yamaguchi, K., Development of the
22
23 24 25 26
Human Hypoglossal Nucleus, Dev. Neurosci., 11 (1989) 212220. Nornes, H.O. and Das, G.D., Temporal pattern of neurogenesis in spinal cord of rat. I. An autoradiographic study -- Time and sites of origin and migration and settling patterns of neuroblasts, Brain Res., 73 (1974) 121-138. Rakic, P., Contact regulation of neuronal migration. In G.M.E. Edelman and J.P. Tiery, (Eds.), The Cell in Contact, Wiley, New York, pp. 67-91. Rakic, P., Principles of neural cell migration, Experientia, 46 (1990) 882-891. Schr6der, K.E, Hopf, A. and Thorner, G., Morphometrischstatische Strukturanalysen des Striatum, Pallidum, und Nucleus Subthalamicus bein Menschen,/. Hirnforsch, 16 (!975) 333-350. Taber and Pierce, E., Histogenesis of the nuclei griseum pontis, corporis pontobulbaris and reticularis tegmenti pontis (Bechterew) in the mouse, I. Comp. Neurol,, 126 (1966) 219-240.