Surface Modifications of Neural Epithelial Cells During Formation of the Neural Tube in the Rat Embryo

/. Embryol. exp. Morph. Vol. 28, 2, pp. 437-448, 1972 437 Printed in Great Britain

Surface modifications of neural epithelial cells during formation of the neural tube in the rat embryo

By BRUCE G. FREEMAN1 From the Department of Anatomy, The University of Tennessee Medical Units, Memphis, Tennessee

SUMMARY The apical (juxtaluminal) ends of the neural epithelial cells of rat embryos were examined using light and electron microscopy during varying stages of neural tube formation. At the neural-plate stage the apical surfaces exhibit numerous microvilli. At the presomite neurula stage the microvilli are longer and more irregular. Filaments of approximately 40-60 A diameter appear in the apical cytoplasm. By the neural-groove stage, cytoplasmic protrusions containing various organelles have begun to appear. Apical filaments are present. At the beginning of closure the apical surfaces are characterized by large, irregular protrusions that are still associated with apical filaments. Finally, at the time of neural closure, the apical protrusions as well as the apical filaments have disappeared and the apical surfaces of the neural epithelial cells are relatively smooth. These observations bear out the proposal that contraction of the apical filaments is responsible for the folding of the neural plate and the production of apical protrusions.

INTRODUCTION It is generally agreed that certain congenital abnormalities of the central nervous system (exencephaly, anencephaly, myeloschisis) are due to a failure of normal formation of the neural tube. In spite of the fact that a large number of chemicals and drugs have been used to cause abnormal neural development (Kalter, 1968), there is little or no consensus on the factors responsible for normal neurulation in many species. Since the process of neurulation is funda- mental to the development of the central nervous system, it would be highly desirable to have as much information about it as possible. The following investigation was undertaken to study, at the fine structural level, the normal morphology of the embryonic rat neural epithelial cells during neurulation in order to gain some insight as to the mechanism of neural tube formation under normal conditions. Structural modifications of the apical ends of cells undergoing neurulation or 1 Author's address: Department of Anatomy, Case Western Reserve University School of Medicine, 2119 Abington Road, Cleveland, Ohio 44016, U.S.A. 438 B. G. FREEMAN neurulation-like movements have been described by a number of authors. Balinsky (1961) was among the first to report protrusions from the apical ends of neural epithelial cells during neurulation in frog embryos. Since then, Baker & Schroeder (1967) and Schroeder (1970) noted' apical protrusions' in neurulating amphibian cells, Wrenn & Wessells (1969) noted 'finger-like projections' in invaginating mouse lens, and Pearce & Zwaan (1970) noted 'apical protusions' in invaginating chick lens. However, to this date, the changes seen in the apical (juxtaluminal) surfaces of the neural epithelial cells of the rat during neurulation have not been described. This study will deal with the observed changes in the apical ends of the neural epithelial cells of the rat during formation of the neural tube. The possible significance of these changes in the mechanisms of closure will be discussed.

MATERIALS AND METHODS Sprague-Dawley rats were obtained from Zivic-Miller Laboratories, Allison Park, Pa., at varying days of pregnancy. Both uterine horns were removed under ether anesthesia and transferred to Tyrode's solution. Embryos were removed under Tyrode's and staged according to Witschi (1956). The embryos were then fixed in toto in 4 % glutaraldehyde or in 2 % OsO4, both buffered to pH 7-5 with 0-2 M cacodylate. After 2-4 h of fixation the embryos fixed in glutaraldehyde were washed for an equivalent amount of time in buffer and postosmicated in 2 % osmium tetroxide buffered to pH 7-5 with 0-2 M cacodylate. The embryos were then dehydrated in an ascending series of concentrations of methanol, passed through propylene oxide, and embedded in Epon 812. Thick

FIGURE 1 Fig. 1. Light micrographs of transverse sections of embryos at Witschi stages 12,13, 14, 15, and 16. All embryos were embedded in Epon and sectioned in the transverse plane at levels approximating one-half of the length of the embryo in stages 12 and 13 and approximately mid- to high-thoracic in stages 14,15, and 16. Sections for electron microscopy were taken from the same levels. (A) Stage 12 (primitive streak), day 9 of gestation. PRO = proamniotic cavity; PNE = primitive neural epithelium; END = endoderm; glutaraldehyde-osmium fixation, x 120. (B) Stage 13 (presomite neurula), day 9-5 of gestation. NG = neural groove; NEP = neural epithelium; arrows = mitotic figures; osmium fixation, x 225. (C) Stage 14 (1-4 somites), day 10 of gestation. NG = neural groove; NEP = neural epithelium; AP = apical protrusions; arrows = mitotic figures; glutaraldehyde-osmium fixation, x 225. (D) Stage 15 (5-12 somites), day 10-5 of gestation. NG = neural groove; NEP = neural epithelium; AP = apical protrusions; arrows = mitotic figures, glutaraldehyde-osmium fixation, x 225. (E) Stage 16 (13-20 somites), day 11 of gestation. NEP = neural epithelium; osmium fixation, x 200. Mammalian neurulation 439 440 B. G. FREEMAN

Fig. 2. Higher magnification view of a transverse section through the apical ends of primitive neural ectoderm cells at stage 12. L = lumen; MV = microvilli; P = plasma membrane vesicles; M = mitochondria; JC = junctional complex; glutaraldehyde-osmium fixation, x 20000. and thin transverse sections from approximately half-way through the neurula, neural plate, neural groove and high thoracic levels in older embryos were cut on a Sorval MT1 ultramicrotome fitted with a diamond knife. Thick (0-5-1 /an) sections of whole embryos were made and stained with Mallory azure II- methylene blue for purposes of orientation. Thin sections were floated on distilled water, picked up on 150-mesh carbon- coated grids and contrasted with uranyl acetate and lead citrate. Specimens were examined in an RCA EMU 3F electron microscope equipped with a heated objective aperture or an Hitachi HU 11A electron microscope, both operated at 50 kV. Micrographs were made on prepumped Cronar, Ortholitho, Type A sheet film at original magnifications of 5000-20000 and photographically enlarged up to 4 times. Mammalian neurulation 441

Fig. 3. (A) View of apical ends of neural epithelial cells in transverse section at stage 13. L = lumen; MV — microvilli; P = plasma membrane vesicles; M = mitochondria; JC = junctional complex; W = mitochondrial whorl; glutaraldehyde- osmium fixation, x 200000. (B) Higher magnification view of apical ends of neural epithelial cells in transverse section at stage 13. L = lumen; JC = junctional complex; F = apical filaments; osmium fixation, x 40000. 442 B. G. FREEMAN

Fig. 4. Transverse section through apical ends of neural epithelial cells at stage 14. L = lumen; MV = microvilli; M = mitochondria; JC = junctional complex; AP = apical protrusion; F = apical filaments; glutaraldehyde-osmium fixation, x 20000.

In all, 17 dams were used to provide a minimum of three dams for each stage of development. A minimum of three embryos were examined from each dam for this investigation. RESULTS The changes in the appearance of the neural epithehum during formation of the neural tube are evident in light micrographs taken from midneural-plate sections in younger embryos to approximately midthoracic levels in older embryos (Fig. 1A-E). At stage 12 the cells are arranged in a low pseudostratifled columnar epithelium (Fig. 1 A). They contain numerous free ribosomes as well as ribosomal aggre- Mammalian neurulation 443

Fig. 5. Transverse section through apical ends of neural epithelial cells at stage 15. AP = apical protrusion; L = lumen; M = mitochondria; JC = junctional complex; F = apical filaments; glutaraldehyde-osmium fixation, x 20000. gates. Mitochondria are numerous. The apical surface is seen to be quite irregular and to exhibit numerous microvilli. Some profiles of shed plasma membranes can be seen in the lumen (Fig. 2). By stage 13 the neural groove has already formed and the neural epithelial cells have become somewhat taller (Fig. 1B). The apical surfaces of the neural epithelial cells exhibit numerous microvilli, some of which contain filaments that continue into the cytoplasm (Fig. 3 A). Junctional complexes are present and well developed. The cytoplasm contains large numbers of polysomes but relatively few profiles of granular endoplasmic reticulum. Mitochondria are numerous and a few contain 'membranous whorls' resembling those described by Jurand & Yamada (1967) in degenerating mitochondria. 20 EMB 28 444 B. G. FREEMAN

Fig. 6. Transverse section through apical ends of neural epithelial cells at stage 16. L = lumen; JC = junctional complex; EX = small extrusion; C = centriole; CI = developing cilium; glutaraldehyde-osmium fixation, x 20000.

Beginning at stage 13, a system of filaments, approximately 40-60 A in diameter, appears in the apical cytoplasm (Fig. 3A, B). These filaments are usually seen to be associated with the junctional complexes of the neural epithelial cells. At stage 14 the neural groove has deepened and the neural folds have begun to approximate each other (Fig. 1C). The apical ends of the neural epithelial cells have undergone observable changes. These include a decrease in the number of microvilli and the appearance of protrusions of the apical cytoplasm into the presumptive lumen (Figs. 1C, 4). These protrusions appear as small buds containing cytoplasmic matrix and ribosomes or as large 'blebs' containing cytoplasmic matrix, ribosomes and mitochondria (Fig. 4). The cytoplasmic Mammalian neurulation 445 protrusions are usually associated with rather complex arrangements of junctional complexes and apical filaments (Fig. 4). By stage 15 the neural tube has usually closed in low cervical and high thoracic levels, although Fig. ID shows a neural tube that is slightly open. At this stage the apical surfaces of the neural epithelial cells are almost completely devoid of microvilli. Moreover, the number of small apical protrusions has decreased. However, the number of large cyoplasmic protrusions, or 'blebs', has greatly increased and the luminal surface now exhibits a highly irregular surface (Figs. 1C, 5). The protrusions of the neural epithelial cells contain more cytoplasm and mitochondria than those seen at stage 14. Junctional complexes and apical filaments are quite prominent. Finally, at stage 16, the neural tube has closed completely (Fig. IE). At this stage the apical ends of the neural epithelial cells have flattened and the luminal surface appears to be smooth (Fig. 6). The apical ends of the cells are smooth and rounded, and the apical protrusions typical of stage-15 embryos (Fig. 5) have disappeared. A few small cytoplasmic extrusions can occasionally be seen. Centrioles and developing cilia are seen in increasing numbers (Fig. 6).

DISCUSSION Many theories concerning the mechanism of neurulation have been proposed. However, sequential ultrastructural analyses of neurulation and related morphogenetic processes (lens invagination, otic vesicle formation, gastrulation, etc.) are relatively rare. The filaments seen in the apical cytoplasm of rat neural epithelial cells appear to be morphologically identical to those described in amphibian neuralepithelial cells (Balinsky, 1961; Baker & Schroeder, 1967; Schroeder, 1970). Balinsky (1961) first reported some of the ultrastructural aspects of neurulation in the frog. He noted the presence of a 'rather thin electron-dense layer' in the apical cytoplasm of the neural epithelial cells. Moreover, he found that this layer was not present in the open neural plate nor in the completed neural tube. It was Balinsky's (1961) contention that this 'electron-dense layer' was made up of fine filaments that were contractile and that their contraction was responsible for folding of the neural plate and the production of apical protrusions. Baker & Schroeder (1967) also noted the presence of filaments in the apical cytoplasm of neural epithelial cells of neurulating tree frog and toad embryos. These authors also suggested these 'apical filaments' were contractile and that the filaments act as a 'purse string' to cause folding of the neural plate. The observations of Schroeder (1970) in Xenopus strongly indicate that whereas the apical filaments are contractile, they probably operate only at the beginning of neurulation, whereas the remainder of the process is due to adhesions of the neural plate cells to the notochord as well as the participation of 'extrinsic forces', such as elongation of myotome cells and movement of the epidermis. 29-2 446 B. G. FREEMAN There is also evidence that apical filaments operate in the folding of epithelia in mammals. Wrenn & Wessells (1969) reported a system of fine filaments in the apical cytoplasm of invaginating mouse lens epithelium. These authors also feel that these filaments are contractile and operate to form the lens cup via a 'purse-string' mechanism. Karfunkel (1971) has shown that the treatment of neurulating embryos of Xenopus with the antimitotic drug vinblastine severely inhibits neurulation to this form. Indeed, the embryos appear externally to have completed neurulation, but Karfunkel's (1971) micrographs clearly show that there is no neural tube, but merely a mass of neural epithelial cells that have rounded up. Moreover, these cells have lost 'the overwhelming majority of the 60 A microfilaments...' as well as all of their microtubules (Karfunkel, 1971). This report would seem to provide further evidence for the involvement of apical filaments in the neurula- tory process. Recently Burnside (1971) has provided further support for the hypothesis that apical filaments have a contractile function in neurulation. Burnside (1971) has shown that the thickness of the bundles of apical filaments increases during neurulation while the circumference of the apical end of the neural epithelial cells diminishes. Her measurements show that during the neurulation process the length of the filament bundles does not change. Moreover, this author noted that apical protrusions and convolutions in the lateral cell contacts appear at this time. Based on this evidence and certain biochemical data that suggest that these filaments resemble actin (Ishikawa, Bischoff & Holtzer, 1969), Burnside (1971) proposes that these apical filaments of the 50-70 A diameter variety actively contract to produce folding of the neural epithelium and do so via a sliding filament mechanism. Recently, Wessells et al. (1971) have reported on the relationship between the occurrence of these filaments and certain biological processes, such as cyto- kinesis, cell movement, tubular gland formation, invagination during gastrulation, and others. Using colchicine and a drug known as cytochalasin B, Wessells et al. (1971) have shown that these filaments are not functionally related to microtubules. The size and number of the apical protrusions seen in this study can also be correlated with the degree of folding of the neural epithelium, thus providing indirect evidence that they are caused by contraction of the apical filaments resulting in the extrusion of apical cytoplasm. Apical protrusions seem to be constantly appearing structures in virtually any embryonic epithelium undergoing folding. Their presence can be correlated with the presence of apical filaments in neurulating amphibian embryos (Balinsky, 1961; Baker & Schroeder, 1967; Schroeder, 1970), in the formation of the mammalian pancreas (Wessells & Evans, 1968) and in the mammalian lens (Wrenn & Wessells, 1969). On the other hand, Langman & Welch (1966) describe the presence of 'bleb- like cytoplasmic protrusions' of the neural epithelial cells lining the diencephalon Mammalian neurulation 447 in rat embryos made exencephalic by maternal hypervitaminosis A. Their 'bleb-like protrusions' closely resemble those seen in Fig. ID. Langman & Welch (1966) also noted that these protrusions were seen in 'younger embryos, but not after day 14'. Since neurulation in the rat occurs from day 9-5 to day 11, and since these authors presented no micrographs of control embryos, the possibility that these cytoplasmic protrusions are normally occurring structures in the diencephalon at this stage must still be entertained. Based on the above data and their striking morphological similarity to other neurulating systems, it is proposed that neurulation in the rat embryo closely resembles that in amphibian embryos. Specifically, these data support the conclusion that the 40-60 A filaments seen in the apical cytoplasm of the rat neural epithelial cells are the agents responsible for the folding of the neural plate. The apical protrusions probably result from apical cytoplasm being squeezed out from between the junctional complexes when the apical filaments contract. Continued contraction of the apical filaments produces deeper folding, finally leading to closure of the neural groove.

The author is deeply grateful to Dr G. Gordon Robertson and to Dr James F. Reger for their interest and helpful criticism in the preparation of this work. This work is part of a dissertation submitted to the Graduate School-Medical Sciences of the University of Tennessee in partial fulfillment of the requirements for the degree of Doctor of Philosophy. This investigation was supported by PHS Training Grant No. 5 T01-GM00202 from the National Institute of General Medical Sciences.

REFERENCES BAKER, P. C. & SCHROEDER, T. E. (1967). Cytoplasmic filaments and morphogenetic movements in the amphibian neural tube. Devi Biol. 15, 432-450. BALINSKY, B. I. (1961). Ultrastructural mechanisms of gastrulation and neurulation. Symp. on Germ Cells and Development, Inst. Intern. Embryol. Fondaz. Pavia: A. Baselli. BURNSIDE, B. (1971). Microtubules and microfilaments in newt neurulation. Devi Biol. 26, 416-441. ISHIKAWA, H., BISCHOFF, R. & HOLTZER, H. (1969). Formation of arrowhead complexes with heavy meromyosin in a variety of cell types. /. Cell Biol. 43, 312-328. JURAND, A. & YAMADA, T. (1967). Elimination of mitochondria during Wolffian lens regenera- tion. Expl Cell Res. 46, 636-638. KARFUNKEL, P. (1971). The role of microtubules and microfilaments in neurulation in Xenopus. Devi Biol. 25, 30-56. KALTER, H. (1968). Teratology of the Central Nervous System. Chicago: University of Chicago Press. LANGMAN, J. & WELCH, G. W. (1966). Effect of vitamin A on development of the central nervous system. /. comp. Neurol. 128, 1-16. PEARCE, T. L. & ZWAAN, J. (1970). A light and electron microscopic study of cell behavior and microtubules in the embryonic chicken lens using Colcemid. J. Embryol. exp. Morph. 23, 491-507. SCHROEDER, T. E. (1970). Neurulation in Xenopus laevis. An analysis and model based on light and electron microscopy. /. Embryol. exp. Morph. 23, 427-462. WESSELLS, N. K. & EVANS, J. (1968). Ultrastructural studies of early morphogenesis and cytodifferentiation in the embryonic mammalian pancreas. Devi Biol. 17, 412-446. 448 B. G. FREEMAN WESSELLS, N. K., SPOONER, B. S., ASH, J. F., BRADLEY, M. O., LUDUENA, M. A., TAYLOR, E. L., WRENN, J. T. & YAMADA, K. F. (1971). Microfilaments in cellular and development processes. Science, N.Y. 171, 135-143. WITSCHI, E. (1956). Development of Vertebrates. Philadelphia: W. B. Saunders Co. WRENN, J. T. & WESSELLS, N. K. (1969). An ultra-structural study of lens invagination in the mouse. /. exp. Zool. Ill, 359-368.

(Manuscript received 13 March 1972)