- Email: [email protected]
Development of the human trochlear nucleus: A morphometric study
Annals of Anatomy 193 (2011) 106–111
Contents lists available at ScienceDirect
Annals of Anatomy journal homepage: www.elsevier.de/aanat
Research article
Development of the human trochlear nucleus: A morphometric study Katsuyuki Yamaguchi ∗ , Koichi Honma Department of Pathology, Dokkyo University School of Medicine, 880 Kitakobayashi, Mibu, Tochigi 321-0293, Japan
a r t i c l e
i n f o
Article history: Received 6 March 2010 Received in revised form 6 October 2010 Accepted 11 October 2010
Keywords: Extraoculomotor nuclei Eye movements Fetus Klüver–Barrera’s method Oculomotor
a b s t r a c t Background: The trochlear nucleus, the smallest of the extraoculomotor nuclei, is unique or even curious, because the nerve roots emerge dorsally from the superior medullary velum after decussation. Little information is available on the developmental anatomy of this nucleus in humans. Design/subjects: We examined serial brain sections from 10 premature infants aged 20–39 weeks of gestation to document the histology and morphometry. Results: The trochlear nucleus was composed of three parts: the rostral tip, the main body, and the caudal division. The rostral tip was a rostral continuation of the main body, being closely related to the oculomotor nucleus; the main body was enveloped by a fibrous capsule; the caudal division was a small separate cluster of neurons in the medial longitudinal fasciculus or the root fibers with individual variations. Tigroid Nissl bodies first appeared at 28 weeks in presumed motoneurons. Various sizes of motoneurons were recognized; medium-sized to small motoneurons were preferentially accumulated in the rostral tip. Among the motoneurons, presumed non-motor neurons were infrequently scattered. Morphometric analysis showed that the nuclear volume exponentially increased with age, about 15 fold over 20–39 weeks, while the average profile area of the neurons linearly increased. Statistical analysis confirmed that cell area was smallest in the rostral tip among the three parts. Conclusion: Although the sample number is small in this study, it suggests that the human trochlear nucleus can be divided into three parts, and that the overall growth may be accelerated at about 30 weeks of gestation. © 2010 Elsevier GmbH. All rights reserved.
1. Introduction Eye movements are important for catching the clearest image of an object. With recent advances in obstetric medicine, fetal eye movements can be now visualized by means of real time ultrasonography, and the first slow changes in eye position have been documented at about 16 postmenstrual weeks (Prechtl and Nijhuis, 1983). Clinically, failure of the oculomotor system to develop has been reported in rare inherited diseases such as congenital dysinnervation disorders (Engle, 2007). To appreciate physiological, pathological, or clinical developmental data, it is necessary to have a precise knowledge of the developmental anatomy of extraoculomotor nuclei. The trochlear nucleus (nIV) innervating the superior oblique muscle, the smallest of these nuclei, is unique or even curious, because the nerve roots emerge dorsally from the superior medullary velum after decussation. In experimental animals, the neuronal cytoarchitecture of the nIV has been vigorously investi-
gated across various classes of the vertebrates, such as lampreys (Meléndez-Ferro et al., 2000), amphibians (Naujoks-Manteuffel ˜ et al., 1986; Munoz and González, 1995), reptiles (El Hassni et al., 2000), birds (Sohal et al., 1985), and mammals (Reis and Machado, 1981; Murphy et al., 1986; Sturrock, 1991; Büttner-Ennever et al., 2001; Eberhorn et al., 2006). Developmental anatomy of this nucleus was also reported by Sohal et al. (1985) and Sonntag and Fritzsch (1987). For humans, reports on the morphology of the nIV are largely limited to adults (Zaki, 1960; Vijayashankar and Brody, 1977; Olszewski and Baxter, 1982) or embryos (Pearson, 1943; Cooper, 1946; Cooper, 1947; Müller and O’Rahilly, 1988; Szyszka-Mróz, 1999), and developmental events have been less intensively studied during the later stages of fetal period. Here we report on the histology and morphometry of the nIV in preterm infants at 20–39 weeks of gestation (WG).
2. Material and methods 2.1. Tissues ∗ Corresponding author. Tel.: +81 282 87 2129; fax: +81 282 86 5171. E-mail addresses: [email protected], [email protected] (K. Yamaguchi). 0940-9602/$ – see front matter © 2010 Elsevier GmbH. All rights reserved. doi:10.1016/j.aanat.2010.10.006
Ten human brains were examined (Table 1). They were from premature infants at 20–39 WG. The infants died of various causes
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111 Table 1 Details of the material. Case no.
Age (weeksa )
Brain weight (g)
Clinical diagnosis
1 2 3 4 5 6 7 8 9 10
20 21 27 28 29 30 35 38 39 39
48 70 130 160 178 NRb 250 NRb 390 380
Medical termination Medical termination Septicemia Asphyxia Asphyxia Hydrops fetalis Esophageal atresia Diaphragmatic hernia Meconium aspiration Diaphragmatic hernia
a b
Gestational weeks. Not recorded.
shortly after birth, and their gestational ages were estimated from the first day of the mother’s last menstrual period. Removal of the brains at autopsy was approved by our institute’s board of ethics. Gross and microscopic examinations of the brain revealed that serious pathological changes including severe ischemic–hypoxic changes, massive hemorrhages, structural anomalies and injuries were absent. 2.2. Histology After fixation in 10% formalin, each brain was post-fixed in a 1:4 mixture of 5% potassium dichromate and 5% potassium chromate solutions for three weeks (Goto and Seki, 1980), embedded in celloidin and 30-m serial sections made. Every fifth or tenth section was stained according to the Klüver–Barrera (K–B) method (Klüver and Barrera, 1953). The remaining sections were stored in 70% ethanol; later some of them were stained with other staining methods, such as hematoxylin and eosin, Kultschitzky’s stain (a modified Weigert’s myelin stain), or silver impregnation. Immunohistochemical staining is desirable for definitive classification of neurons, but in this study, it was not possible to apply it to longstored celloidin sections. 2.3. Volumetry The nuclear volume was estimated using Cavalieri’s pointcounting method (Mouton, 2002). After attaching a square ocular grid to the eyepiece, we counted the intersections hitting the nuclear area on the K–B specimens. The volume (V) was calculated by the following formula: V (mm3 ) = a × P × d, where ‘a’ was the area of a unit square (0.0064 or 0.04 mm2 ), ‘P’ was the sum of the numbers of intersections for one side of the nucleus and ‘d’ was the distance between two adjacent slides (0.15 or 0.3 mm). 2.4. Neuronal profile area Neuronal profile areas were measured to make a quantitative assessment of the size of neurons. Because it is generally difficult to estimate cell volume accurately with conventional light microscopes, we decided to measure the profile (or sectional) area on the left side. Cell diameters have long been applied to morphometry, but it is laborious to determine and handle two variables (long and short axes) at the same time in a comparative analysis. Thus we felt a two-dimensional parameter (profile area) would produce the most accurate and efficient measurement for analysis. For the study, we selected three K–B specimens to compare the data from the following three parts of this nucleus: the rostral tip, the main body, and the caudal division (see below). For each specimen, all the neuronal profiles observed were drawn with a light microscope (Optiphot, Nikon, Tokyo, Japan) equipped with a drawing tube at a final magnification of 500×. The line drawings were digitized and
107
entered into a personal computer (HP Compaq DX 2000 ST, Hewlett Packard, Palo Alto, CA, USA) using a tablet (UD 1212R, Wacom, Saitama, Japan). Quantitative data was obtained from a program for image analysis (VM32, Rise Corporation, Sendai, Japan). Morphological determination of a neuron was made using the following criteria: (1) a relatively large cell body, (2) prominent and distinct nucleus surrounded by a basophilic perikaryon and (3) presence of single or multiple distinct nucleoli. Tissue shrinkage is generally inevitable during processing of histological preparations. However, because an accurate estimation of the extent of shrinkage was difficult in these specimens, we did not correct for shrinkage. After examining normality of distribution and equality of variance, we used the unpaired Welch’s (or Student’s) t-tests to compare the averages, and took P < 0.05 as significant. 3. Results 3.1. Three parts of the trochlear nucleus The nIV was identified as a compact mass of relatively large cells ventral to the central grey substance of the midbrain as early as 20 WG. Observations of serial sections revealed that the nucleus is comprised of two small parts, as well as the main body: the rostral tip and the caudal division. The rostral tip was a rostral continuation of the main body, being closely related to the most caudal part of the oculomotor nucleus (nIII) with a narrow cell-sparse zone (Fig. 1A). This part was constantly seen in our sample brains, and was mainly occupied by medium-sized to small neurons. The main body, the largest part of the nIV, was enveloped by a thick fibrous capsule, where relatively large neurons predominated (Fig. 1B). The caudal division was a small separate cluster of neurons in the medial longitudinal fasciculus, caudal to the main body. In two cases (Cases 4–5), it lay more laterally in the trajectory of root fibers (Fig. 1C). It showed some individual variations, as it appeared unilaterally or bilaterally, or was even absent in Case 2. No distinct differences were seen in appearance of the neuronal cytoarchitecture between the main body and the caudal division, except that the large neurons were somewhat elongated in the latter. 3.2. Neuronal types of trochlear nucleus Before 28 WG, the neurons were only divided into two types (large and small neurons), because their Nissl bodies were immature and rather fine even in the larger neurons. It was first possible to distinguish presumed motoneurons from other types of neurons (non-motor neurons) at 28 WG. Various sizes of motoneurons were recognized: from small to large motoneurons. The larger subtypes had a plump or ovoid cell body, and were generally conspicuous over the nucleus, except the rostral tip (Fig. 2A). The smaller subtypes had a relatively poor cytoplasm, but coarse Nissl bodies of tigroid type were visible (Fig. 2B). Among these motoneurons, presumed non-motor neurons were infrequently encountered. These presumed non-motor neurons were generally small, round or ovalshaped, and their Nissl bodies were fine and located peripherally (Fig. 2C). 3.3. Nuclear volume The caudal division was not included for the volume measurement, because this part was too small and variable. The nuclear volume increased about 15 fold with age during the examined period (20–39 WG). We independently measured the two sides as an in-specimen control. Measurements of the left trochlear nucleus (0.029–0.348 mm3 ) and right trochlear nucleus (0.02–0.336 mm3 ) showed very similar sizes and rates of increase. This dramatic
108
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111
Fig. 1. Three parts of the trochlear nucleus in a 28-week preterm infant (Case 4). (A) Rostral tip. Relatively small cells are densely accumulated. An asterisk indicates a cell-sparse zone between the oculomotor (III) and the trochlear (IV) nucleus. (B) Main body. Large neurons predominate. (C) Caudal division. A separate cluster of neurons (arrow). In this case it is somewhat laterally situated in the trochlear nerve roots (asterisk). CG, the central grey substance; D, dorsal; M, medial; RF, the reticular formation. Klüver–Barrera (K–B) stain. Scale bars: 200 m.
Fig. 2. Neuronal types of the trochlear nucleus in a 39-week preterm infant (Case 9). (A and B) Typical tigroid Nissl bodies are discernible in presumed motoneurons: larger (A) and smaller (B) subtype. (C) Nissl bodies are fine and located peripherally in a presumed non-motor neuron. K–B stain. Scale bars: 20 m.
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111
increase was best approximated to an exponential function curve on a scatter plot (Fig. 3A). 3.4. Neuronal profile area The average areas of the neuronal profiles were compared for the rostral tip, the main body and the caudal division. Averages from all the neurons, including presumed motor and non-motor types, showed a gradual or linear increase with the age over 20–39 WG for each part (Table 2). The averages were statistically smaller in the rostral tip, while they were not significantly different between the main body and the caudal division, except for in one case (Case 6). On a scatter plot, the regression line of the averages from the rostral tip was clearly below those from the other parts (Fig. 3B). The neuronal profile area was separately measured for the presumed motor and non-motor neurons from 28 WG onwards. The numerical percentages of presumed non-motor neurons to all neurons were estimated to be 7.7–16.5% in the rostral tip, 1.9–24.1% in the main body, and 0.0–17.6% in the caudal division; the averages were 10.9%, 12.7%, and 6.09%, respectively (Table 2). The averages of profile area showed a gradual or linear increase with the age for both motoneurons (Fig. 3C) and non-motor neurons (Fig. 3D). As shown in scatter plots, the regression lines of the averages from the rostral tip were below those from the other parts. 4. Discussion The nIV is one of the smallest motor nuclei in the brainstem, although the precise morphology has not been fully investigated in humans. This is partly because it is necessary to make a complete set of serial sections comprising at least the midbrain to adequately investigate the three-dimensional structure of this nucleus. While celloidin embedding is seldom used at present, we believe that it is indispensable for this purpose, as the brain shows considerable shrinkage after paraffin-embedding. The important findings of the present study can be summarized as follows: (1) two small parts of the nIV were distinguished from its main body (the rostral tip and the caudal division); (2) coarse, tigroid Nissl bodies were first visible at 28 WG in presumed motoneurons; (3) the caudal division appeared to have individual variations; (4) the nuclear volume exponentially increased with the age over 20–39 WG, while the average areas of neuronal profiles linearly increased; (5) statistical analysis confirmed that the average neuronal areas were smallest in the rostral tip. According to previous reports, the anlage of the human nIV first appears about 3–4 postovulatory weeks (Carnegie stage 13–14) within the mantle layer as a mass of postmitotic cells just caudal to the nIII (Pearson, 1943; Cooper, 1946, 1947; Müller and O’Rahilly, 1988; Szyszka-Mróz, 1999). There is a distinct gap between the two ˜ nuclei in most vertebrates (Naujoks-Manteuffel et al., 1986; Munoz and González, 1995; El Hassni et al., 2000), but it can be absent in some species (Reis and Machado, 1981). In humans, it has been described in early embryos (Szyszka-Mróz, 1999). We also noticed a narrow cell-sparse zone between the two nuclei in our samples. At least, the main body of nIV can be separated from the nIII in humans. Natural cell death or apoptosis of the trochlear neurons has been described in avian embryos (Sohal et al., 1985), but not in humans. We did not see any karyorrhectic cells that would suggest this phenomenon. The most rostral part of the nIV, the rostral tip, has not been previously described in humans to our knowledge. It should be noted that both of the motor and non-motor neurons are significantly smaller in this small part. Functional aspects remain unclear, but in experimental animals, a similar finding to ours has been reported by Büttner-Ennever and co-workers (Büttner-Ennever et al., 2001;
109
Eberhorn et al., 2006). They described a small region over the nIV, a dorsal cap, which is occupied by smaller motoneurons supplying the multiply innervated muscle fibers. The rostral tip observed here might be homologous to this structure in humans, but further studies are needed to know its precise function. The existence of the caudal division has already been reported. Pearson (1943) stated that it appeared in more than 20–30% of specimens, whereas Cooper (1946) denied the existence of this division. We observed it at a high frequency (90%), 9 of 10 sample brains, although it can appear with individual variations. As observed in Cases 4–5, it can be seen laterally arranged along the trajectory of nerve roots (Pearson, 1943). The exact origin of the caudal division remains unclear, but it could represent a remnant of migrating neurons, because our morphometric data on the neuronal size indicated little difference between these cells and those in the main body of the nucleus. We have observed various sizes of motoneurons in the nIV from 28 WG onwards. This study provided little information on the functional aspects, but the larger subtypes might be principal neurons supplying the singly innervated muscle fibers because of their predominance in the main body. However, the functions of the smaller subtypes are unknown. Some might supply the multiply innervated muscle fibers (Büttner-Ennever et al., 2001), as mentioned above, and others might serve as ␥-motoneurons innervating the intrafusal fibers of the muscle spindles (Lukas et al., 1994). Heterogeneity of the motoneurons has been reported in various species of the vertebrates. In rabbits, approximately 3% of the total motoneurons (smaller motoneurons) innervated the ipsilateral superior oblique muscle (Murphy et al., 1986), and in chick embryos, anterior and posterior cell groups were described in early trochlear neurons (Irving et al., 2002). Thus, it is possible that the nIV may not be a uniform structure. Little attention has been paid to the non-motor neurons of nIV. This is because they are relatively scarce in this nucleus. In contrast, their relative number is much higher in the abducens nucleus (nVI), as we estimated the numerical percentage of presumed nonmotor neurons to be about 55% in the nVI (unpublished data). This is in agreement with the fact that the nVI contains a number of internuclear neurons projecting to the nIII via the medial longitudinal fasciculus. Functions of the non-motor neurons of nIV remain unclear, although a specific population of small GABAergic neurons has been reported in lampreys (Meléndez-Ferro et al., 2000). Morphometric studies on the human nIV have been limited to adults so far. The total number of neurons in the adult nIV was estimated to be 1500 (Zaki, 1960) or between 1810 and 2400 (Vijayashankar and Brody, 1977). Vijayashankar and Brody examined 20 male brains at various postnatal ages and concluded that the number was relatively constant among them. This numerical stability has been also reported for mice (Sturrock, 1991). However, those authors only counted the motoneurons and ignored the non-motor neurons. Apart from their number, the size of a neuron is an important quantitative parameter. Our data suggests that the neuronal maturation may gradually progress after mid-gestation. Three-dimensional or volumetric analysis may be of great value to know how a structure develops as a whole. The present data indicates that the nIV may rapidly expand from nearly 30 WG. This may be coincident with the time when tigroid Nissl bodies were first observed in the presumed motoneurons of our samples. We have reported similar morphometric changes in other nuclei in developing human brains, including cerebellar nuclei (Yamaguchi et al., 1989) and the parvocellular red nucleus (Yamaguchi and Goto, 2008), demonstrating an exponential increase of the volume and a linear increase of the neuronal profile area. The latter was also observed in the gigantocellular reticular nucleus (Yamaguchi et al., 1994) and the magnocellular red nucleus (Yamaguchi and Goto, 2006). This suggests that an exponential
110
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111
Table 2 Average profile areas of motor and non-motor neurons. Case no.
Rostral tip All
1 2 3 4 5 6 7 8 9 10
Main body Motor
N
Area
N
178 216 137 103 149 91 154 109 95 119
115.3 137.3 135.1 194.2 144.5 203.8 289.3 372.4 248.2 237.8
– – – 93 132 84 140 91 86 104
Caudal division
Non-motor
All
Area
N
Area
N
Area
N
Motor
206.7 153.4 213.4 308.3 420.8 263.8 261.2
– – – 10 17 7 14 18 9 15
78.6 74.8 89.1 99.8 127.8 98.9 75.4
105 184 108 112 103 196 139 83 135 80
147.6 181.6 185.7 265.8 251.3 372.2 367.5 447.2 330.9 323.6
– – – 103 101 180 119 63 112 68
Non-motor
All
Area
N
Area
N
Area
N
Motor
281.8 254.3 392.4 410.1 537.5 373.5 360.8
– – – 9 2 16 20 20 23 12
83.4 99.9 144.9 113.6 162.9 123.2 112.7
14 – 20 22 13 27 18* 17 11 26*
129.2 – 182.3 279.9 246.5 293.5 380.2 447.1 354.3 294.9
– – – 21 12 26 18 14 10 26
Non-motor Area
N
290.7 257.2 299.4 380.2 525.5 375.8 294.9
– – – 1 1 1 0 3 1 0
Area
53.1 117.9 140.8 – 81.2 139.0 –
The data was from the left trochlear nucleus, except for in the caudal divisions of Cases 7 and 10 (*). (m2 ).
Fig. 3. Scatter plots of the relationship between morphometric data and age. (A) Nuclear volume. Regression equations: y = 0.0029 exp (0.1201x) for right nucleus (dotted line), and y = 0.0026 exp (0.123 x) for left nucleus (solid line). (B) Average profile area for all neurons. Regression equations: y = 9.5526x − 84.519 for rostral tip (dotted line), y = 11.598x − 67.573 for main body (solid line), and y = 12.732x − 113.42 for caudal division (broken line). (C) Average profile area for motoneurons. Regression equations: y = 12.192x − 153.45 for rostral tip (dotted line), y = 11.803x − 28.404 for main body (solid line), and y = 11.616x − 50.243 for caudal division (broken line). (D) Average profile area for non-motor neurons. Regression equations: y = 1.9479x + 25.828 for rostral tip (dotted line), and y = 2.5361x + 33.858 for main body (solid line).
increase occurs in the neuropil due to axonal and dendritic branching which does not require a dramatic expansion of the somatic size to support it. In conclusion, although the sample number is small in this study, it suggests that the human trochlear nucleus can be divided into three parts, and that the overall growth may be accelerated at about 30 weeks of gestation. Acknowledgment We greatly appreciate the support and advice of Dr Nobuhide Masawa, Professor of Pathology at Dokkyo University School of Medicine.
References Büttner-Ennever, J.A., Horn, A.K.E., Scherberger, H., D’Ascanio, P., 2001. Motoneurons of twitch and nontwitch extraocular muscle fibers in the abducens, trochlear, and oculomotor nuclei of monkeys. J. Comp. Neurol. 438, 318– 335. Cooper, E.R.A., 1946. The development of the nuclei of the oculomotor and trochlear nerves (somatic efferent column). Brain 69, 20–57. Cooper, E.R.A., 1947. The trochlear nerve in the human embryo and fetus. Br. J. Ophthalmol. 31, 257–275. Eberhorn, A.C., Büttner-Ennever, J.A., Horn, A.K.E., 2006. Identification of motoneurons supplying multiply- or singly-innervated extraocular muscle fibers in the rat. Neuroscience 137, 891–903. Engle, E.C., 2007. Oculomotility disorders arising from disruptions in brainstem motor neuron development. Arch. Neurol. 64, 633– 637.
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111 El Hassni, M., Bennis, M., Rio, J.P., Repérant, J., 2000. Localization of motoneurons innervating the extraocular muscles in the chameleon (Chomaeleo chameleon). Anat. Embryol. 201, 63–74. Goto, N., Seki, Y., 1980. Influences on findings of sections stained with luxol fast blue and cresyl violet method after differential fixations: with special reference to the findings of nerve cells after a new fixation. Shinkei Naika 12, 358–366 (in Japanese). Irving, C., Malhas, A., Guthrie, S., Mason, I., 2002. Establishing the trochlear motor axon trajectory: role of the isthmic organizer and Fgf8. Development 129, 5389–5398. Klüver, H., Barrera, E., 1953. A method for the combined staining of cells and fibers in the nervous system. J. Neuropathol. Exp. Neurol. 12, 400–403. Lukas, J.R., Aigner, M., Blumer, R., Heinzl, H., Mayr, R., 1994. Number and distribution of neuromuscular spindles in human extraocular muscles. Invest. Ophthalmol. Vis. Sci. 35, 4317–4327. Meléndez-Ferro, M., Pérez-Costas, E., González, M.J., Pombal, M.A., Anadón, R., Rodicio, M.C., 2000. GABA-immunoreactive internuclear neurons in the ocular system of lampreys. Brain Res. 855, 150–157. Mouton, P.R., 2002. Principles and Practices of Unbiased Stereology. An Introduction for Biologists. Johns Hopkins University Press, Baltimore and London. Müller, F., O’Rahilly, R., 1988. The first appearance of the future cerebral hemispheres in the human embryo at stage 14. Anat. Embryol. 177, 495–511. ˜ Munoz, M., González, A., 1995. The trochlear nucleus of the frog Rana ridibunda: localization, morphology and ultrastructure of identified motoneurons. Brain Res. Bull. 36, 433–441. Murphy, E.H., Garone, M., Tashayyod, D., Baker, R.B., 1986. Innervation of extraocular muscles in the rabbit. J. Comp. Neurol. 254, 78–90. Naujoks-Manteuffel, C., Manteuffel, G., Himstedt, W., 1986. Localization of motoneurons innervating the extraocular muscles in Salamandra salasmandra L. (amphibia, urodela). J. Comp. Neurol. 254, 133–141.
111
Olszewski, J., Baxter, D., 1982. Cytoarchitecture of the Human Brain Stem, second edition. Karger, Basel. Pearson, A.A., 1943. The trochlear nerve in human fetuses. J. Comp. Neurol. 78, 29–43. Prechtl, H.F.R., Nijhuis, J.G., 1983. Eye movements in the human fetus and newborn. Behav. Brain Res. 10, 119–124. Reis, F.P., Machado, L.M.S., 1981. Cytoarchitecture of the oculomotor nuclear complexes and of the trochlear nucleus in the marmoset (Callithrix jacchus). Acta Anat. 111, 289–295. Sohal, G.S., Knox, T.S., Allen jr, J.C., Arumugam, T., Campbell, L.R., Yamashita, T., 1985. Development of the trochlear nucleus in quail and comparative study of the trochlear nucleus, nerve, and innervation of the superior oblique muscle in quail, chick, and duck. J. Comp. Neurol. 239, 227–236. Sonntag, R., Fritzsch, B., 1987. The development of the amphibian trochlear nucleus. An HRP study. Neurosci. Lett. 77, 143–148. Sturrock, R.R., 1991. Stability of motor neuron number in the oculomotor and trochlear nuclei of the ageing mouse brain. J. Anat. 174, 125–129. Szyszka-Mróz, J., 1999. Development of the trochlear nerve in the human embryos. Folia Morphol. (Warsz) 58, 307–313. Vijayashankar, A., Brody, H., 1977. Aging in the human brain stem. A study of the nucleus of the trochlear nerve. Acta Anat. 99, 169–172. Yamaguchi, K., Goto, N., Yamamoto, T.Y., 1989. Development of human cerebellar nuclei. Morphometric study. Acta Anat. 136, 61–68. Yamaguchi, K., Goto, N., Honma, K., 1994. Development of the human gigantocellular reticular nucleus: a morphometric study. Acta Anat. 150, 191–197. Yamaguchi, K., Goto, N., 2006. Development of the human magnocellular red nucleus: a morphological study. Brain Dev. 28, 431–435. Yamaguchi, K., Goto, N., 2008. Development of the human parvocellular red nucleus. A morphological study. Dev. Neurosci. 30, 325–330. Zaki, W., 1960. Le nerf pathétique chez l’homme. Etude relative à son origine, son trajet intracérébral et à sa structure. Arch. Anat. Histol. Embryol. 43, 105–120.
Contents lists available at ScienceDirect
Annals of Anatomy journal homepage: www.elsevier.de/aanat
Research article
Development of the human trochlear nucleus: A morphometric study Katsuyuki Yamaguchi ∗ , Koichi Honma Department of Pathology, Dokkyo University School of Medicine, 880 Kitakobayashi, Mibu, Tochigi 321-0293, Japan
a r t i c l e
i n f o
Article history: Received 6 March 2010 Received in revised form 6 October 2010 Accepted 11 October 2010
Keywords: Extraoculomotor nuclei Eye movements Fetus Klüver–Barrera’s method Oculomotor
a b s t r a c t Background: The trochlear nucleus, the smallest of the extraoculomotor nuclei, is unique or even curious, because the nerve roots emerge dorsally from the superior medullary velum after decussation. Little information is available on the developmental anatomy of this nucleus in humans. Design/subjects: We examined serial brain sections from 10 premature infants aged 20–39 weeks of gestation to document the histology and morphometry. Results: The trochlear nucleus was composed of three parts: the rostral tip, the main body, and the caudal division. The rostral tip was a rostral continuation of the main body, being closely related to the oculomotor nucleus; the main body was enveloped by a fibrous capsule; the caudal division was a small separate cluster of neurons in the medial longitudinal fasciculus or the root fibers with individual variations. Tigroid Nissl bodies first appeared at 28 weeks in presumed motoneurons. Various sizes of motoneurons were recognized; medium-sized to small motoneurons were preferentially accumulated in the rostral tip. Among the motoneurons, presumed non-motor neurons were infrequently scattered. Morphometric analysis showed that the nuclear volume exponentially increased with age, about 15 fold over 20–39 weeks, while the average profile area of the neurons linearly increased. Statistical analysis confirmed that cell area was smallest in the rostral tip among the three parts. Conclusion: Although the sample number is small in this study, it suggests that the human trochlear nucleus can be divided into three parts, and that the overall growth may be accelerated at about 30 weeks of gestation. © 2010 Elsevier GmbH. All rights reserved.
1. Introduction Eye movements are important for catching the clearest image of an object. With recent advances in obstetric medicine, fetal eye movements can be now visualized by means of real time ultrasonography, and the first slow changes in eye position have been documented at about 16 postmenstrual weeks (Prechtl and Nijhuis, 1983). Clinically, failure of the oculomotor system to develop has been reported in rare inherited diseases such as congenital dysinnervation disorders (Engle, 2007). To appreciate physiological, pathological, or clinical developmental data, it is necessary to have a precise knowledge of the developmental anatomy of extraoculomotor nuclei. The trochlear nucleus (nIV) innervating the superior oblique muscle, the smallest of these nuclei, is unique or even curious, because the nerve roots emerge dorsally from the superior medullary velum after decussation. In experimental animals, the neuronal cytoarchitecture of the nIV has been vigorously investi-
gated across various classes of the vertebrates, such as lampreys (Meléndez-Ferro et al., 2000), amphibians (Naujoks-Manteuffel ˜ et al., 1986; Munoz and González, 1995), reptiles (El Hassni et al., 2000), birds (Sohal et al., 1985), and mammals (Reis and Machado, 1981; Murphy et al., 1986; Sturrock, 1991; Büttner-Ennever et al., 2001; Eberhorn et al., 2006). Developmental anatomy of this nucleus was also reported by Sohal et al. (1985) and Sonntag and Fritzsch (1987). For humans, reports on the morphology of the nIV are largely limited to adults (Zaki, 1960; Vijayashankar and Brody, 1977; Olszewski and Baxter, 1982) or embryos (Pearson, 1943; Cooper, 1946; Cooper, 1947; Müller and O’Rahilly, 1988; Szyszka-Mróz, 1999), and developmental events have been less intensively studied during the later stages of fetal period. Here we report on the histology and morphometry of the nIV in preterm infants at 20–39 weeks of gestation (WG).
2. Material and methods 2.1. Tissues ∗ Corresponding author. Tel.: +81 282 87 2129; fax: +81 282 86 5171. E-mail addresses: [email protected], [email protected] (K. Yamaguchi). 0940-9602/$ – see front matter © 2010 Elsevier GmbH. All rights reserved. doi:10.1016/j.aanat.2010.10.006
Ten human brains were examined (Table 1). They were from premature infants at 20–39 WG. The infants died of various causes
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111 Table 1 Details of the material. Case no.
Age (weeksa )
Brain weight (g)
Clinical diagnosis
1 2 3 4 5 6 7 8 9 10
20 21 27 28 29 30 35 38 39 39
48 70 130 160 178 NRb 250 NRb 390 380
Medical termination Medical termination Septicemia Asphyxia Asphyxia Hydrops fetalis Esophageal atresia Diaphragmatic hernia Meconium aspiration Diaphragmatic hernia
a b
Gestational weeks. Not recorded.
shortly after birth, and their gestational ages were estimated from the first day of the mother’s last menstrual period. Removal of the brains at autopsy was approved by our institute’s board of ethics. Gross and microscopic examinations of the brain revealed that serious pathological changes including severe ischemic–hypoxic changes, massive hemorrhages, structural anomalies and injuries were absent. 2.2. Histology After fixation in 10% formalin, each brain was post-fixed in a 1:4 mixture of 5% potassium dichromate and 5% potassium chromate solutions for three weeks (Goto and Seki, 1980), embedded in celloidin and 30-m serial sections made. Every fifth or tenth section was stained according to the Klüver–Barrera (K–B) method (Klüver and Barrera, 1953). The remaining sections were stored in 70% ethanol; later some of them were stained with other staining methods, such as hematoxylin and eosin, Kultschitzky’s stain (a modified Weigert’s myelin stain), or silver impregnation. Immunohistochemical staining is desirable for definitive classification of neurons, but in this study, it was not possible to apply it to longstored celloidin sections. 2.3. Volumetry The nuclear volume was estimated using Cavalieri’s pointcounting method (Mouton, 2002). After attaching a square ocular grid to the eyepiece, we counted the intersections hitting the nuclear area on the K–B specimens. The volume (V) was calculated by the following formula: V (mm3 ) = a × P × d, where ‘a’ was the area of a unit square (0.0064 or 0.04 mm2 ), ‘P’ was the sum of the numbers of intersections for one side of the nucleus and ‘d’ was the distance between two adjacent slides (0.15 or 0.3 mm). 2.4. Neuronal profile area Neuronal profile areas were measured to make a quantitative assessment of the size of neurons. Because it is generally difficult to estimate cell volume accurately with conventional light microscopes, we decided to measure the profile (or sectional) area on the left side. Cell diameters have long been applied to morphometry, but it is laborious to determine and handle two variables (long and short axes) at the same time in a comparative analysis. Thus we felt a two-dimensional parameter (profile area) would produce the most accurate and efficient measurement for analysis. For the study, we selected three K–B specimens to compare the data from the following three parts of this nucleus: the rostral tip, the main body, and the caudal division (see below). For each specimen, all the neuronal profiles observed were drawn with a light microscope (Optiphot, Nikon, Tokyo, Japan) equipped with a drawing tube at a final magnification of 500×. The line drawings were digitized and
107
entered into a personal computer (HP Compaq DX 2000 ST, Hewlett Packard, Palo Alto, CA, USA) using a tablet (UD 1212R, Wacom, Saitama, Japan). Quantitative data was obtained from a program for image analysis (VM32, Rise Corporation, Sendai, Japan). Morphological determination of a neuron was made using the following criteria: (1) a relatively large cell body, (2) prominent and distinct nucleus surrounded by a basophilic perikaryon and (3) presence of single or multiple distinct nucleoli. Tissue shrinkage is generally inevitable during processing of histological preparations. However, because an accurate estimation of the extent of shrinkage was difficult in these specimens, we did not correct for shrinkage. After examining normality of distribution and equality of variance, we used the unpaired Welch’s (or Student’s) t-tests to compare the averages, and took P < 0.05 as significant. 3. Results 3.1. Three parts of the trochlear nucleus The nIV was identified as a compact mass of relatively large cells ventral to the central grey substance of the midbrain as early as 20 WG. Observations of serial sections revealed that the nucleus is comprised of two small parts, as well as the main body: the rostral tip and the caudal division. The rostral tip was a rostral continuation of the main body, being closely related to the most caudal part of the oculomotor nucleus (nIII) with a narrow cell-sparse zone (Fig. 1A). This part was constantly seen in our sample brains, and was mainly occupied by medium-sized to small neurons. The main body, the largest part of the nIV, was enveloped by a thick fibrous capsule, where relatively large neurons predominated (Fig. 1B). The caudal division was a small separate cluster of neurons in the medial longitudinal fasciculus, caudal to the main body. In two cases (Cases 4–5), it lay more laterally in the trajectory of root fibers (Fig. 1C). It showed some individual variations, as it appeared unilaterally or bilaterally, or was even absent in Case 2. No distinct differences were seen in appearance of the neuronal cytoarchitecture between the main body and the caudal division, except that the large neurons were somewhat elongated in the latter. 3.2. Neuronal types of trochlear nucleus Before 28 WG, the neurons were only divided into two types (large and small neurons), because their Nissl bodies were immature and rather fine even in the larger neurons. It was first possible to distinguish presumed motoneurons from other types of neurons (non-motor neurons) at 28 WG. Various sizes of motoneurons were recognized: from small to large motoneurons. The larger subtypes had a plump or ovoid cell body, and were generally conspicuous over the nucleus, except the rostral tip (Fig. 2A). The smaller subtypes had a relatively poor cytoplasm, but coarse Nissl bodies of tigroid type were visible (Fig. 2B). Among these motoneurons, presumed non-motor neurons were infrequently encountered. These presumed non-motor neurons were generally small, round or ovalshaped, and their Nissl bodies were fine and located peripherally (Fig. 2C). 3.3. Nuclear volume The caudal division was not included for the volume measurement, because this part was too small and variable. The nuclear volume increased about 15 fold with age during the examined period (20–39 WG). We independently measured the two sides as an in-specimen control. Measurements of the left trochlear nucleus (0.029–0.348 mm3 ) and right trochlear nucleus (0.02–0.336 mm3 ) showed very similar sizes and rates of increase. This dramatic
108
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111
Fig. 1. Three parts of the trochlear nucleus in a 28-week preterm infant (Case 4). (A) Rostral tip. Relatively small cells are densely accumulated. An asterisk indicates a cell-sparse zone between the oculomotor (III) and the trochlear (IV) nucleus. (B) Main body. Large neurons predominate. (C) Caudal division. A separate cluster of neurons (arrow). In this case it is somewhat laterally situated in the trochlear nerve roots (asterisk). CG, the central grey substance; D, dorsal; M, medial; RF, the reticular formation. Klüver–Barrera (K–B) stain. Scale bars: 200 m.
Fig. 2. Neuronal types of the trochlear nucleus in a 39-week preterm infant (Case 9). (A and B) Typical tigroid Nissl bodies are discernible in presumed motoneurons: larger (A) and smaller (B) subtype. (C) Nissl bodies are fine and located peripherally in a presumed non-motor neuron. K–B stain. Scale bars: 20 m.
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111
increase was best approximated to an exponential function curve on a scatter plot (Fig. 3A). 3.4. Neuronal profile area The average areas of the neuronal profiles were compared for the rostral tip, the main body and the caudal division. Averages from all the neurons, including presumed motor and non-motor types, showed a gradual or linear increase with the age over 20–39 WG for each part (Table 2). The averages were statistically smaller in the rostral tip, while they were not significantly different between the main body and the caudal division, except for in one case (Case 6). On a scatter plot, the regression line of the averages from the rostral tip was clearly below those from the other parts (Fig. 3B). The neuronal profile area was separately measured for the presumed motor and non-motor neurons from 28 WG onwards. The numerical percentages of presumed non-motor neurons to all neurons were estimated to be 7.7–16.5% in the rostral tip, 1.9–24.1% in the main body, and 0.0–17.6% in the caudal division; the averages were 10.9%, 12.7%, and 6.09%, respectively (Table 2). The averages of profile area showed a gradual or linear increase with the age for both motoneurons (Fig. 3C) and non-motor neurons (Fig. 3D). As shown in scatter plots, the regression lines of the averages from the rostral tip were below those from the other parts. 4. Discussion The nIV is one of the smallest motor nuclei in the brainstem, although the precise morphology has not been fully investigated in humans. This is partly because it is necessary to make a complete set of serial sections comprising at least the midbrain to adequately investigate the three-dimensional structure of this nucleus. While celloidin embedding is seldom used at present, we believe that it is indispensable for this purpose, as the brain shows considerable shrinkage after paraffin-embedding. The important findings of the present study can be summarized as follows: (1) two small parts of the nIV were distinguished from its main body (the rostral tip and the caudal division); (2) coarse, tigroid Nissl bodies were first visible at 28 WG in presumed motoneurons; (3) the caudal division appeared to have individual variations; (4) the nuclear volume exponentially increased with the age over 20–39 WG, while the average areas of neuronal profiles linearly increased; (5) statistical analysis confirmed that the average neuronal areas were smallest in the rostral tip. According to previous reports, the anlage of the human nIV first appears about 3–4 postovulatory weeks (Carnegie stage 13–14) within the mantle layer as a mass of postmitotic cells just caudal to the nIII (Pearson, 1943; Cooper, 1946, 1947; Müller and O’Rahilly, 1988; Szyszka-Mróz, 1999). There is a distinct gap between the two ˜ nuclei in most vertebrates (Naujoks-Manteuffel et al., 1986; Munoz and González, 1995; El Hassni et al., 2000), but it can be absent in some species (Reis and Machado, 1981). In humans, it has been described in early embryos (Szyszka-Mróz, 1999). We also noticed a narrow cell-sparse zone between the two nuclei in our samples. At least, the main body of nIV can be separated from the nIII in humans. Natural cell death or apoptosis of the trochlear neurons has been described in avian embryos (Sohal et al., 1985), but not in humans. We did not see any karyorrhectic cells that would suggest this phenomenon. The most rostral part of the nIV, the rostral tip, has not been previously described in humans to our knowledge. It should be noted that both of the motor and non-motor neurons are significantly smaller in this small part. Functional aspects remain unclear, but in experimental animals, a similar finding to ours has been reported by Büttner-Ennever and co-workers (Büttner-Ennever et al., 2001;
109
Eberhorn et al., 2006). They described a small region over the nIV, a dorsal cap, which is occupied by smaller motoneurons supplying the multiply innervated muscle fibers. The rostral tip observed here might be homologous to this structure in humans, but further studies are needed to know its precise function. The existence of the caudal division has already been reported. Pearson (1943) stated that it appeared in more than 20–30% of specimens, whereas Cooper (1946) denied the existence of this division. We observed it at a high frequency (90%), 9 of 10 sample brains, although it can appear with individual variations. As observed in Cases 4–5, it can be seen laterally arranged along the trajectory of nerve roots (Pearson, 1943). The exact origin of the caudal division remains unclear, but it could represent a remnant of migrating neurons, because our morphometric data on the neuronal size indicated little difference between these cells and those in the main body of the nucleus. We have observed various sizes of motoneurons in the nIV from 28 WG onwards. This study provided little information on the functional aspects, but the larger subtypes might be principal neurons supplying the singly innervated muscle fibers because of their predominance in the main body. However, the functions of the smaller subtypes are unknown. Some might supply the multiply innervated muscle fibers (Büttner-Ennever et al., 2001), as mentioned above, and others might serve as ␥-motoneurons innervating the intrafusal fibers of the muscle spindles (Lukas et al., 1994). Heterogeneity of the motoneurons has been reported in various species of the vertebrates. In rabbits, approximately 3% of the total motoneurons (smaller motoneurons) innervated the ipsilateral superior oblique muscle (Murphy et al., 1986), and in chick embryos, anterior and posterior cell groups were described in early trochlear neurons (Irving et al., 2002). Thus, it is possible that the nIV may not be a uniform structure. Little attention has been paid to the non-motor neurons of nIV. This is because they are relatively scarce in this nucleus. In contrast, their relative number is much higher in the abducens nucleus (nVI), as we estimated the numerical percentage of presumed nonmotor neurons to be about 55% in the nVI (unpublished data). This is in agreement with the fact that the nVI contains a number of internuclear neurons projecting to the nIII via the medial longitudinal fasciculus. Functions of the non-motor neurons of nIV remain unclear, although a specific population of small GABAergic neurons has been reported in lampreys (Meléndez-Ferro et al., 2000). Morphometric studies on the human nIV have been limited to adults so far. The total number of neurons in the adult nIV was estimated to be 1500 (Zaki, 1960) or between 1810 and 2400 (Vijayashankar and Brody, 1977). Vijayashankar and Brody examined 20 male brains at various postnatal ages and concluded that the number was relatively constant among them. This numerical stability has been also reported for mice (Sturrock, 1991). However, those authors only counted the motoneurons and ignored the non-motor neurons. Apart from their number, the size of a neuron is an important quantitative parameter. Our data suggests that the neuronal maturation may gradually progress after mid-gestation. Three-dimensional or volumetric analysis may be of great value to know how a structure develops as a whole. The present data indicates that the nIV may rapidly expand from nearly 30 WG. This may be coincident with the time when tigroid Nissl bodies were first observed in the presumed motoneurons of our samples. We have reported similar morphometric changes in other nuclei in developing human brains, including cerebellar nuclei (Yamaguchi et al., 1989) and the parvocellular red nucleus (Yamaguchi and Goto, 2008), demonstrating an exponential increase of the volume and a linear increase of the neuronal profile area. The latter was also observed in the gigantocellular reticular nucleus (Yamaguchi et al., 1994) and the magnocellular red nucleus (Yamaguchi and Goto, 2006). This suggests that an exponential
110
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111
Table 2 Average profile areas of motor and non-motor neurons. Case no.
Rostral tip All
1 2 3 4 5 6 7 8 9 10
Main body Motor
N
Area
N
178 216 137 103 149 91 154 109 95 119
115.3 137.3 135.1 194.2 144.5 203.8 289.3 372.4 248.2 237.8
– – – 93 132 84 140 91 86 104
Caudal division
Non-motor
All
Area
N
Area
N
Area
N
Motor
206.7 153.4 213.4 308.3 420.8 263.8 261.2
– – – 10 17 7 14 18 9 15
78.6 74.8 89.1 99.8 127.8 98.9 75.4
105 184 108 112 103 196 139 83 135 80
147.6 181.6 185.7 265.8 251.3 372.2 367.5 447.2 330.9 323.6
– – – 103 101 180 119 63 112 68
Non-motor
All
Area
N
Area
N
Area
N
Motor
281.8 254.3 392.4 410.1 537.5 373.5 360.8
– – – 9 2 16 20 20 23 12
83.4 99.9 144.9 113.6 162.9 123.2 112.7
14 – 20 22 13 27 18* 17 11 26*
129.2 – 182.3 279.9 246.5 293.5 380.2 447.1 354.3 294.9
– – – 21 12 26 18 14 10 26
Non-motor Area
N
290.7 257.2 299.4 380.2 525.5 375.8 294.9
– – – 1 1 1 0 3 1 0
Area
53.1 117.9 140.8 – 81.2 139.0 –
The data was from the left trochlear nucleus, except for in the caudal divisions of Cases 7 and 10 (*). (m2 ).
Fig. 3. Scatter plots of the relationship between morphometric data and age. (A) Nuclear volume. Regression equations: y = 0.0029 exp (0.1201x) for right nucleus (dotted line), and y = 0.0026 exp (0.123 x) for left nucleus (solid line). (B) Average profile area for all neurons. Regression equations: y = 9.5526x − 84.519 for rostral tip (dotted line), y = 11.598x − 67.573 for main body (solid line), and y = 12.732x − 113.42 for caudal division (broken line). (C) Average profile area for motoneurons. Regression equations: y = 12.192x − 153.45 for rostral tip (dotted line), y = 11.803x − 28.404 for main body (solid line), and y = 11.616x − 50.243 for caudal division (broken line). (D) Average profile area for non-motor neurons. Regression equations: y = 1.9479x + 25.828 for rostral tip (dotted line), and y = 2.5361x + 33.858 for main body (solid line).
increase occurs in the neuropil due to axonal and dendritic branching which does not require a dramatic expansion of the somatic size to support it. In conclusion, although the sample number is small in this study, it suggests that the human trochlear nucleus can be divided into three parts, and that the overall growth may be accelerated at about 30 weeks of gestation. Acknowledgment We greatly appreciate the support and advice of Dr Nobuhide Masawa, Professor of Pathology at Dokkyo University School of Medicine.
References Büttner-Ennever, J.A., Horn, A.K.E., Scherberger, H., D’Ascanio, P., 2001. Motoneurons of twitch and nontwitch extraocular muscle fibers in the abducens, trochlear, and oculomotor nuclei of monkeys. J. Comp. Neurol. 438, 318– 335. Cooper, E.R.A., 1946. The development of the nuclei of the oculomotor and trochlear nerves (somatic efferent column). Brain 69, 20–57. Cooper, E.R.A., 1947. The trochlear nerve in the human embryo and fetus. Br. J. Ophthalmol. 31, 257–275. Eberhorn, A.C., Büttner-Ennever, J.A., Horn, A.K.E., 2006. Identification of motoneurons supplying multiply- or singly-innervated extraocular muscle fibers in the rat. Neuroscience 137, 891–903. Engle, E.C., 2007. Oculomotility disorders arising from disruptions in brainstem motor neuron development. Arch. Neurol. 64, 633– 637.
K. Yamaguchi, K. Honma / Annals of Anatomy 193 (2011) 106–111 El Hassni, M., Bennis, M., Rio, J.P., Repérant, J., 2000. Localization of motoneurons innervating the extraocular muscles in the chameleon (Chomaeleo chameleon). Anat. Embryol. 201, 63–74. Goto, N., Seki, Y., 1980. Influences on findings of sections stained with luxol fast blue and cresyl violet method after differential fixations: with special reference to the findings of nerve cells after a new fixation. Shinkei Naika 12, 358–366 (in Japanese). Irving, C., Malhas, A., Guthrie, S., Mason, I., 2002. Establishing the trochlear motor axon trajectory: role of the isthmic organizer and Fgf8. Development 129, 5389–5398. Klüver, H., Barrera, E., 1953. A method for the combined staining of cells and fibers in the nervous system. J. Neuropathol. Exp. Neurol. 12, 400–403. Lukas, J.R., Aigner, M., Blumer, R., Heinzl, H., Mayr, R., 1994. Number and distribution of neuromuscular spindles in human extraocular muscles. Invest. Ophthalmol. Vis. Sci. 35, 4317–4327. Meléndez-Ferro, M., Pérez-Costas, E., González, M.J., Pombal, M.A., Anadón, R., Rodicio, M.C., 2000. GABA-immunoreactive internuclear neurons in the ocular system of lampreys. Brain Res. 855, 150–157. Mouton, P.R., 2002. Principles and Practices of Unbiased Stereology. An Introduction for Biologists. Johns Hopkins University Press, Baltimore and London. Müller, F., O’Rahilly, R., 1988. The first appearance of the future cerebral hemispheres in the human embryo at stage 14. Anat. Embryol. 177, 495–511. ˜ Munoz, M., González, A., 1995. The trochlear nucleus of the frog Rana ridibunda: localization, morphology and ultrastructure of identified motoneurons. Brain Res. Bull. 36, 433–441. Murphy, E.H., Garone, M., Tashayyod, D., Baker, R.B., 1986. Innervation of extraocular muscles in the rabbit. J. Comp. Neurol. 254, 78–90. Naujoks-Manteuffel, C., Manteuffel, G., Himstedt, W., 1986. Localization of motoneurons innervating the extraocular muscles in Salamandra salasmandra L. (amphibia, urodela). J. Comp. Neurol. 254, 133–141.
111
Olszewski, J., Baxter, D., 1982. Cytoarchitecture of the Human Brain Stem, second edition. Karger, Basel. Pearson, A.A., 1943. The trochlear nerve in human fetuses. J. Comp. Neurol. 78, 29–43. Prechtl, H.F.R., Nijhuis, J.G., 1983. Eye movements in the human fetus and newborn. Behav. Brain Res. 10, 119–124. Reis, F.P., Machado, L.M.S., 1981. Cytoarchitecture of the oculomotor nuclear complexes and of the trochlear nucleus in the marmoset (Callithrix jacchus). Acta Anat. 111, 289–295. Sohal, G.S., Knox, T.S., Allen jr, J.C., Arumugam, T., Campbell, L.R., Yamashita, T., 1985. Development of the trochlear nucleus in quail and comparative study of the trochlear nucleus, nerve, and innervation of the superior oblique muscle in quail, chick, and duck. J. Comp. Neurol. 239, 227–236. Sonntag, R., Fritzsch, B., 1987. The development of the amphibian trochlear nucleus. An HRP study. Neurosci. Lett. 77, 143–148. Sturrock, R.R., 1991. Stability of motor neuron number in the oculomotor and trochlear nuclei of the ageing mouse brain. J. Anat. 174, 125–129. Szyszka-Mróz, J., 1999. Development of the trochlear nerve in the human embryos. Folia Morphol. (Warsz) 58, 307–313. Vijayashankar, A., Brody, H., 1977. Aging in the human brain stem. A study of the nucleus of the trochlear nerve. Acta Anat. 99, 169–172. Yamaguchi, K., Goto, N., Yamamoto, T.Y., 1989. Development of human cerebellar nuclei. Morphometric study. Acta Anat. 136, 61–68. Yamaguchi, K., Goto, N., Honma, K., 1994. Development of the human gigantocellular reticular nucleus: a morphometric study. Acta Anat. 150, 191–197. Yamaguchi, K., Goto, N., 2006. Development of the human magnocellular red nucleus: a morphological study. Brain Dev. 28, 431–435. Yamaguchi, K., Goto, N., 2008. Development of the human parvocellular red nucleus. A morphological study. Dev. Neurosci. 30, 325–330. Zaki, W., 1960. Le nerf pathétique chez l’homme. Etude relative à son origine, son trajet intracérébral et à sa structure. Arch. Anat. Histol. Embryol. 43, 105–120.