studies on the mechanisms of neurulation in the chick: morphometric analysis of force distribution...

THE JOURNAL OF EXPERIMENTAL ZOOLOGY 244:425-436 (1987)

Studies on the Mechanisms of Neurulation in the Chick: Morphometric Analysis of Force Distribution Within the Neuroepithelium During Neural Tube Formation

ROBERT G. NAGELE, EDWARD HUNTER, KEVIN BUSH, AND HSIN-YI LEE Department ofPedLatrics, University ofMedicine and Dentistry of New Jerseyschool of Osteopathic Medicine, Camden, New Jersey 08103 (R. G. N.); Department of Biology, Rutgers University, Camdeq New Jersey 08102 (E.H., K.B., H.L)

ABSTRACT Changes in the shape of neuroepithelial cells, particularly apical constriction, are generally thought to play a major role in generating the driving forces for neural tube formation. Our previous study [Nagele and Lee (1987) J. Exp. Zool., 241:197-2051 has shown that, in the developing midbrain region of stage 8+ chick embryos, neuroepithelial cells showing the greatest degree of apical constriction are concentrated at sites of enhanced bending of the neuroepithelium (i.e., the floor and midlateral walls of developing neural tube), suggesting that driving forces resulting from apical constriction are concentrated at these sites during closure of the neural tube. In the present study, we have used morphometric methods to 1) measure regional variations in the degree of apical constriction and apical surface folding at selected regions along the anteroposterior axis of stage 8+ chick embryos, which closely resemble the various ontogenetic phases of neural tube formation, and 2) investigate how forces resulting from apical constriction are distributed within the neuroepithelium during transformation of the neural plate into a neural tube. Results show that, during neural tube formation, driving forces resulting from apical constriction are not distributed uniformly throughout the neuroepithelium but rather are concentrated sequentially at three distinct locations: 1) the floor (during transformation of the neural plate to a V-shaped neuroepithelium), 2) the midlateral walls (during transformation of the V-shaped neuroepithelium into a C-shaped neuroepithelium), and 3) the upper walls (during the transformation of the C-shaped neuroepithelium into a closed neural tube).

Neurulation is one of the earliest morphogenetic movements in the chick embryo. This process involves the formation of the neural plate as an ectodermal thickening, the elevation of its lateral edges as neural folds, and the curling over and fusion of neural folds to form a neural tube. These changes in the shape of the neuroepithelium are brought about by forces originating from within neuroepithelial cells (intrinsic forces) andor from outside the neuroepithelium (extrinsic forces). Examples of possible extrinsic forces are 1) mediad pushing forces (e.g., tension generated by the somites, perineural extra- cellular matrix, and expanding surface ecto- derm); 2) vertical (dorsoventral) stretching

forces (tension generated by pulling of the elongating notochord); and 3) longitudinal (craniocaudal) stretching forces (tension generated by elongation of the embryo and ex- pansion of the blastoderm over the yolk) (for discussion, see Desmond and Schoenwolf, '86). Evidence for the intrinsic origin of the driving forces was first shown by Holtfreter ('47) who observed that cells isolated from the amphibian neural plate retain their co- lumnar shape and even continue to elongate in culture. To date, most of the work on intrinsic forces has focused on the role of

0 1987 ALAN R. LISS, INC.

426 R.G. NAGELE ET AL.

cytoskeletal components in generating the observed changes in the shape of neuroepithelial cells. For example, Waddington and Perry ('66) and Messier and coworkers (Messier, '69; Messier and Auclair, '74) have demonstrated a close relationship between elongation of neuroepithelial cells and the presence of paraxially oriented microtubules. In addition, a number of studies have inves- tigated the possible role of apical constriction (presumably resulting from the contractile activities of apical microfilament bundles) in causing the observed bending of the neuroepithelium during neural tube formation (Baker and Schroeder, '67; Karfunkel, '72; Burnside, '73; Schroeder, '73; Nagele and Lee, '80; Lee and Nagele, '85b). Taken together, it is becoming increasingly clear that driving forces for neural tube formation are likely to be a result of a complex interplay of both intrinsic and extrinsic forces, the nature of which may vary somewhat among the major regions of the forming neural tube (forebrain, midbrain, hindbrain, and spinal cord). To get a clearer picture of how the neural tube closes, the separate contributions of each extrinsic and intrinsic factor should be stud- ied independently.

The present study focuses on the possible contribution of apical constriction of neuroepithelial cells to bending of the neuroepithelium during neural tube formation. Many neuroepithelial cells in the supranotochordal area become wedge-shaped before elevation of neural folds (Schoenwolf and Franks, '84; Schoenwolf, ,851, suggesting that apical constriction is not a passive consequence of bending of the neuroepithelium. A major unsolved problem is how forces arising from apical constriction of neuroepithelial cells are distributed (or applied) within the neuroepithelium during the complex sequence of morphogenetic movements leading to closure of the neural tube. Unfortunately, there is no method currently available to measure directly these forces in developing neuroepithelial cells. However, in our previous study Wagele and Lee, '871, we described a new method that, based on a combination of morphometry and computer-assisted image analysis, can reveal regional variations in the local degree of apical constriction within the neuroepithelium during neural tube formation. Since apical constriction is generally thought to be related to the activities of apical microfilament bundles, i t is possible that these measurements can also serve as an

indirect barometer of regional variations in the functional activity of apical microfilament bundles.

In the present study, we have extended our previous work and examined systematically the distribution of apical constriction-mediated forces within the neuroepithelium during the various phases of neural tube formation exhibited along the anteroposterior axis of stage 8+ chick embryos. Our main objective was to use the stage 8' embryo as a model system to determine how and when apical constriction-mediated forces might be applied within the neuroepithelium during the normal ontogeny of the neural tube.

MATERIALS AND METHODS

Fertile White Leghorn eggs were incu- bated at 375°C to obtain embryos at stage 8+ of development (Hamburger and Hamil- ton, '51). Embryos were isolated from the yolk and vitelline membrane. Some were fixed in Bouin's fluid, stained with Dela- fields hematoxylin, and kept as whole mounts. Others were processed for transmission electron microscopy and morphornetry as described below.

Transmission electron microscopy (TEM) Embryos were prepared for TEM as de-

scribed previously (Nagele and Lee, '80, '87). Thin, transverse sections through the developing neuroepithelium at selected locations along the embryo axis were cut with a dia- mond knife and mounted onto Formvar- coated slotted grids (1 x 2 mm single hole; Electron Microscopy Sciences, Fort Washing- ton, PA). This system facilitated visualiza- tion of the entire section without the interference of grid bars. Sections were con- trasted with uranyl acetate and lead citrate and stabilized by exposing briefly to a low intensity electron beam before examining at higher magnification with a Zeiss EM 109 electron microscope.

Computer-assisted image analyses All measurements were made with the aid

of a Zeiss Videoplan Computerized Image Analyzer interfaced with an IBM AT computer and equipped with a graphics tablet and all the necessary software for image analysis and for three-dimensional recon- struction of serial sections.

To document regional variations in the degree of apical surface folding and constriction, sections through the neuroepithelium

MECHANISMS OF NEURAL TUBE FORMATION 427

at selected locations (showing various phases of neural tube formation) along the axis of five stage 8' embryos were mounted onto Formvar-coated slotted grids. Each section was photographed at a magnification of x 1,100 in such a way that a composite (montage) photograph, composed of 15-30 individual printed photographs (20.3 x 25.4 cm), could be prepared at a final magnification of X4,OOO. To ensure a high degree of resolution for both measured parameters, the future luminal surface of the developing neuroepithelium was divided into a series of 10-pm wide zones (numbered consecutively on each side of the forming neural tube from the midline to the margin of the neuroepithelium) (Fig. 1B). Each montage was mounted on the graphics tablet attached to the Videoplan and the degree of apical surface folding and constriction exhibited by cells in each 10-pm wide zone within the neuroepithelium was determined as described previously (Nagele and Lee, '87) and outline briefly below.

Determination of regional variations in apical constriction

The contour length of the folded apical surfaces of neuroepithelial cells within each 10- pm wide zone was measured directly from composite photographs by tracing the apical surface topography with a stylus attached to the graphics tablet of the Videoplan. The stylus had a point-to-point resolution of 0.5 mm, and the graphics tablet had a spatial resolution of 60-pm. An apical surface folding factor (F) for each zone was calculated as follows:

apical contour length apical straight-line length

F =

The apical straight-line length was determined by drawing a straight line with the stylus across the width of each zone.

Determination of regional variations in apical surface folding

The method used was based on the fact that, on an average, cells with smaller apical diameter (i.e., a greater degree of apical constriction) exhibit smaller apical widths in thin, transverse sections. Thus, by measur- ing apical widths of individual cells making up each 10-pm wide zone, it is possible to calculate an apical constriction factor which reflects accurately the degree of apical constriction exhibited by cells in each zone. Since the degree of apical constriction is inversely

proportional to the measured apical width, an apical constriction factor was calculated as the reciprocal of the mean apical width of cells for each 10-pm wide zone.

RESULTS Ultrastructure of developing chick

neuroepithelial cells Since details of the ultrastructure of devel-

oping chick neuroepithelial cells have been presented elsewhere (e.g., Messier, '69; Kar- funkel, '72; Karfunkel, '74; Nagele and Lee, '79, '801, only features relevant to the present study will be described briefly.

The neuroepithelium of stage 8+ embryos shows great variation in the degree of bending along its length, ranging from a nearly flat neural plate just anterior to Hensen's node to a closed neural tube at the level of the midbrain (Fig. 1A). Throughout the forming neural tube, neuroepithelial cells are arranged into a pseudostratified epithelium (Fig. 1B). Adjacent cells are attached to one another at their apical ends by developing junctions (Fig. 1C). At the level of these junctions, microfilaments are organized into discrete bundles along the apical perimeter and anchored to the plasma membrane at regions of increased electron density. The straight- ness of these bundles and the "pulled-in" appearance of the plasma membrane at points of their attachment give the impres- sion that microfilament bundles are either under tension or are involved in generating tension.

Morphological comparison of neuroepithelial cells at different locations within the wall of the forming neural tube reveals consider- able variation in the degree of apical constriction and surface folding. For example, in many locations along the embryo axis, cells comprising the future floor exhibit a greater degree of apical constriction and surface folding than those found in the upper wall and future fusion areas (Nagele and Lee, '87) (cf. plates C,D, Fig. 1).

Regional variations in apical Constriction and surface folding in the neuroepithelium

during neural tube formation The neural plate

The neural plate region of stage 8+ embryos is located just anterior to Hensen's node and often shows slight bending along the midline (Fig. 2A). Figures 2B,C illustrate regional variations in the degree of apical constriction and surface folding of cells in the

MECHANISMS OF NEURAL TUBE FQEMATION 429

neural plate. Cells close to the midline possess a greater degree of apical constriction and surface folding than those located more laterally. In the former, the value representing the degree of apical constriction (apical constriction factor) reaches 0.42, which cor- responds to a mean apical width of 2.33 pm. In contrast, cells forming the lateral two- thirds of the neural plate show much less apical constriction and little or no regional variations in this parameter (i.e., the apical constriction factor remains close to 0.2, corresponding to a mean apical width of 5.0 pm). These results indicate that there are regional variations in the degree of apical constriction in the neural plate area of stage 8+ chick embryos. Apparently, neural plate cells located in the vicinity of the midline already have undergone a certain degree of apical constriction as a first step in uplifting the margins of the neural plate to form neural folds. This finding is supported by the obser- vation that the neural plate in stage 8+ chick embryos often is not completely flat (Fig. 2A). Patterns of regional variations in the degree of apical constriction and surface folding of neuroepithelial cells are similar throughout the neural plate region (Fig. 2B,C).

Fig. 1. General morphology of neuroepithelial cells in stage 8+ chick embryos. A) Dorsal view of a stage 8’ chick embryo. All phases of neural tube formation, from the flat neural plate (NP) just anterior to Hensen‘s node (arrow) to the closed neural tube at the level of the developing midbrain N), are represented in a single embryo. There are five pairs of well-defined somites at this stage of development. X38. B) Composite (montage) electron micrograph of a transverse section through the developing midbrain region of a stage 8+ embryo that was used to measure regional variations in the degree of apical surface folding and constriction. Each montage was printed at a final magnification of X4,OOO. The luminal surface of each contralateral neural fold was divided into a series of 10-pm wide zones which, starting at the midline and ending at the fusion area, were numbered consecutively. The location of a few of these zones is indicated by the numbers near the luminal surface. x 600. C ) Transmission electron micrograph showing the apical regions of neuroepithelial cells in the midlateral walls of the future midbrain. Microfilaments (MF) are organized into prominent bundles located just under the luminal surface. Apical cell surfaces are highly folded. L, lumen. x 10,000. D) Transmission electron micrograph showing the apical regions of neuroepithelial cells in the upper walls of the developing midbrain. Cells exhibit broader apical widths (less apical constridion) and smoother luminal surfaces than those located in the floor and midlateral walls (cf. C). Microfilament (MF) bundles are thinner and less conspicuous. L, lumen. x 10,000.

The V-shaped neuroepithelium Just anterior to the neural plate, the neu-

roepithelium bends along the midline to form a structure resembling the letter “V’ (Fig. 2D). This change in overall shape of the neuroepithelium is accompanied by an enhance- ment of regional variations in apical constriction and surface folding of neuroepithelial cells that were found to exist in the neural plate (Fig. 2E,F). Cells located in the immediate vicinity of the midline exhibit the greatest degree of apical constriction (apical constriction factor = 0.55, corresponding to a mean apical width of 1.82 pm) and surface folding (folding factor = 1.62). As in the neural plate region, the lateral two-thirds of the neuroepithelium exhibits little or no regional variations in either parameter. In fact, in terms of the degree of apical constriction and surface folding, cells comprising most of the uplifted walls of the V-shaped neuroepithelium are similar to those forming the lateral two-thirds of the neural plate (cf. Fig. 2, plates B,C and E,F).

The U-shaped neuroepithelium Just posterior to the developing hindbrain

of the stage 8+ chick embryo, the midlateral walls of the V-shaped neuroepithelium provide a locus of bending, resulting in a U- shaped structure (Fig. 3A). This transition is accompanied by a marked increase in the degree of apical constriction and surface folding of cells in the midlateral walls (Fig. 3B,C). For example, the mean apical constriction factor increases from 0.27 in the midlat- era1 walls of the V-shaped neuroepithelium (Fig. 2E) to 0.64 in the comparable region of the U-shaped neuroepithelium (Fig. 3B), corresponding to a local reduction in mean apical width from 3.7 pm to 1.6 pm. Surprisingly, despite this dramatic increase in apical constriction in the midlateral walls, the mean apical constriction factor of cells located close to the midline remains nearly identical to that in the V-shaped neuroepithelium (Fig. 3B). Furthermore, cells comprising the lateral third of the neuroepithelium, which later form the roof of the neural tube remain un- constricted (mean apical constriction factor = 0.18, corresponding to a mean apical width of 5.6 pm) and are comparable to those found in the lateral two-thirds of the neural plate. The pattern of regional variations in the degree of apical surface folding is similar to that of apical constriction (compare Fig. 3B,C). The mean apical surface folding factor


APICAL CONSTRICTION N E U W PUTE - STAGE a+

1

B

APICAL SURFACE FOLDING * G U R U PUE - STAGE e+

Fig. 2. Measurements of apical constriction and surface foldmg in the neural plate and V-shaped regions of the neuroepithelium of five stage 8+ embryos. A) Trans- verse section through the neural plate region. The neuroepithelium shows a slight bendmg along the midline and cells are arranged into a pseudostratified epithelium. X200. B,C) Graphs summarizing regional variations in the degree of apical constriction and surface folding in the neural plate region. Cells in the midline possess a greater degree of apical constriction and surface folding than those located more laterally. D) Trans- verse section through the V-shaped neuroepithelium of a stage 8+ embryo located just anterior to the neural plate. Bending occurs along the midline. The lateral two-thirds of the neuroepithelium remains flat as in the neural plate (cf. A). ~350. E,F) Graphs summarizing

APICAL SURFACE FOLDING v - W E D - n Y i F a+

regional variations in the degree of apical constriction and surface folding in the V-shaped neuroepithelium. Cells in the midline possess a greater degree of apical constriction and surface folding than those located more laterally. As in the neural plate region, the lateral two- thirds of the neuroepithelium exhibits little or no regional variations in apical constriction and surface folding. B,E) The scale for the ordinate is the apical Constriction factor (the reciprocal of the measured apical widths in micrometers. C,F) The scale for the ordinate is the folding factor which reflects the degree of apical surface folding. The abscissa is divided into a series of 10-pm wide zones which are numbered consecutively from the midline (floor) to the future area. Each data point represents the mean 5 SD of measurements taken at each zone from five embryos.


APICAL CONSTRICTION " - s * I P s o - SIACE e*

I

j

B

APICAL SURFACE FOLDING "- -SWLD - SI- a+

p 2.2

Fig. 3. Measurements of apical constriction and surface folding in the U- and C-shaped neuroepithelium of stage 8 'embryos. A) Transverse section through the U- shaped neuroepithelium of a stage S+ embryo. A locus of bending appears in the midlateral walls and bans- forms the V-shaped neuroepithelium to the U-shaped neuroepithelium. ~350. B,C) Graphs summarizing regional variations in the degree of apical constriction and surface folding in the U-shaped neuroepithelium. The transformation of the V- to U-shaped neuroepithelium is hallmarked by a dramatic increase in the degree of apical constriction and surface folding of cells in the midlateral walls. In contrast, cells located close to the midline remain unchanged and are comparable to those found at the midline of the V-shaped neuroepithelium. Cells comprising the lateral one-third of the neuroepithelium remain comparable to those found in the corresponding region of the neural plate. Dj Transverse section through the C-shaped neuroepithelium of a stage 8' embryo. The leading edges of neural folds are curled

APICAL SURFACE FOLDING c - s w m - STAGE a+

Q 3.2

towards each other. ~350. E,Fj Graphs summarizing regional variations in the degree of apical constriction and surface folding of neuroepithelial cells in the C- shaped neuroepithelium. The transformation of the U- to C-shaped neuroepithelium is accompanied by a marked increase in the degree of apical constriction and surface folding of cells located in the midlateral walls. In contrast, the degree of apical constriction and surface folding in the floor and upper walls of the C-shaped neuroepithelium increases only slightly over that of the U-shaped neuroepithelium. B,E) The scale for the ordinate is the apical constriction factor (the reciprocal of the measured apical widths in micrometers). C,F) The scale for the ordinate is the folding factor which reflects the degree of apical surface folding. The abscissa is divided into a series of 10-pm wide zones which are numbered consecutively from the midline (floor) to the fusion area. Each data point represents the mean SD of measurements taken at each zone from five embryos.

432 R.G. NA( 2ELE ET AL.

B

APICAL CON STR ICTI 0 N CLOSFD NEUPAL TUBE - sr*TE a+

I

APICAL SURFACE FOLDING CLDSED N E U W rdm - S * C E a+

z* ,

F 8 3 4 e b I f

2.3 i z2 4 I

reaches 1.97 in the midlateral walls of the U- shaped neuroepithelium, representing a more than two-fold increase in the degree of sup- face folding over the corresponding area in the V-shaped neuroepithelium. Thus, these data indicate that the transition of the V- to U-shaped neuroepithelium is accompanied by a marked increase in the degree of apical constridion and surface folding of cells comprising the midlateral walls, whereas those which are destined to form the floor and roof of the neural tube remain essentially unchanged during this phase of neural tube formation.

The C-shaped neuroepithelium In the hindbrain region of the stage 8+

chick embryo, the degree of bending in the midlateral walls is such that the shape of the neuroepithelium resembles the letter "C" (Fig. 3D). The transition of the U- to C-shaped neuroepithelium is accompanied by a further prominent increase in the degree of apical constriction and surface folding of neuroepithelial cells, especially in the midlat- era1 walls (Fig. 3E,F). In fact, the magnitudes of these parameters in the midlateral walls reach their highest values in the C-shaped neuroepithelium. For example, the mean apical constriction factor increases from 0.64 in the midlateral walls of the U-shaped neuroepithelium to 0.90 in the same area of the C- shaped neuroepithelium. This change corre- sponds to a reduction in the mean apical width of neuroepithelium cells from 1.6 pm to 1.10 pm. As was the case during the transition of the V- to U-shaped neuroepithelium, the degree of apical constriction and surface folding of cells in the floor and upper walls of the C-shaped neuroepithelium does not increase significantly over that of the U-shaped neuroepithelium. Furthermore, the pattern of regional variations in the degree of apical surface folding in the C-shaped neuroepithelium is similar to that of apical constriction.

Fig. 4. Measurements of apical constriction and surface folding of cells in the closed neural tube of five stage 8+ embryos. A) Transverse section through the closed neural tube at the level of the midbrain of a stage 8* embryo. ~350. B) Graph summarizing regional variations in the degree of apical costriction of neuroepithelial cells. The transformation of the C-shaped neuroepithelium into the closed neural tube is distinguished by a conspicuous increase in the degree of apical constriction exhibited by cells located throughout the upper walls. Cells situated close to the midline show a slight increase in the degree of apical constriction, while those in the midlateral walls exhibit a slight decrease in this parameter. The scale for the ordinate is the apical con-

striction factor (the reciprocal of the measured apical widths in micrometers). C) Graph summarizing regional variations in the degree of apical surface folding. Al- though the overall pattern for apical surface folding is nearly identical to that of the C-shaped neuroepithelium, the magnitude of the apical surface folding factor throughout the midbrain is considerably reduced. The scale for the ordinate is the folding factor which reflects the degree of apical surface folding. B,C) The abscissa is divided into a series of 10-pm wide zones which are numbered consecutively from the midline (floor) to the future fusion area. Each data point represents the mean k SD of measurements taken at each zone from five embryos.


The mean apical surface folding factor reaches 2.77 in the midlateral wall, representing a great increase in the degree of apical surface folding over the corresponding area of the U-shaped neuroepithelium. These results indicate that, during the transition of the U- to the C-shaped neuroepithelium, there is a dramatic increase in the degree of apical constriction and surface folding of cells located in the midlateral walls of the forming neural tube. In contrast, cells situated in the floor and lateral third (future roof) of the neuroepithelium do not undergo significant changes in these parameters during this phase of neural tube formation.

The closed neural tube Closure of the neural tube in the chick first

occurs in the future midbrain of the stage 8+ embryo (Fig. 4A). The transition of the C- shaped neuroepithelium in the hindbrain region to the closed neural tube in the midbrain is accompanied by local changes in the degree of apical constriction and surface folding, particularly in the floor and upper walls (Fig. 4B,C). In cells situated in the vicinity of the midline, the mean apical constriction factor increases from 0.53 in the C-shaped neuroepithelium to 0.65 in the closed neural tube, reflecting a moderate reduction in mean apical width from 1.89 pm to 1.54 pm. How- ever, compared to the C-shaped neuroepithelium, the most conspicuous change in the degree of apical constriction occurs in cells located throughout the upper walls. In fact, the mean apical constriction factor for cells comprising this region increases from 0.17 in the C-shaped neuroepithelium to 0.30 in the closed neural tube, corresponding to a reduction in mean apical width from 5.9 pm to 3.3 pm. In contrast, cells forming the midlateral walls exhibit slightly less apical constriction than those in the corresponding area of the C-shaped neuroepithelium. As in all other phases of neural tube formation in stage 8+ embryos, the pattern of regional variations in the degree of apical surface folding is similar to that of apical constriction. However, despite this similarity of patterns, the magnitude of the apical surface folding factor throughout the midbrain is surprisingly much reduced from that of the C-shaped neuroepithelium. Overall, results suggest that forces driving the initial contact of apposing neural folds arise from the upper walls of the forming neural tube.

DISCUSSION

Neurulation involves the gradual transformation of a flat neuroepithelial sheet (the neural plate) into a hollow cylinder (the neural tube). Over the past several decades, many hypotheses have been proposed to ac- count for the origin of the driving forces for this process (for review, see Karfunkel, '74; Schoenwolf, '82; Jacobson and Gordon, '76; Jacobson et al., '85; Gordon, '85; Campbell et al., '86; Jacobson et al., '86). Among these hypotheses, the original proposal of Baker and Schroeder ('67) and Schroeder ('73) that microfilaments constrict the apical ends of neuroepithelial cells by a "purse-string-like" mechanism to generate the wedge- (or bot- tle-) shaped appearance of neuroepithelial cells and provide the bulk of driving forces for uplifting of neural folds still seems to be viable. In support of this idea, numerous sub- sequent studies (e.g., Karfunkel, '72; Messier and Auclair, '74; Lee and Kalmus, '76; Moran, '76; Moran and Rice, '76; Lee and Nagele, '79; O'Shea and Kaufman, '80; Mor- riss-Kay, '81; Nagele et al., '81; Lee et al., '83; Lee and Nagele, '85a,b,c, '86) have shown that chemical agents (e.g., cytochalasins, local anesthetics, calcium agonists and antag- onists, and calmodulin inhibitors) known to interfere with microfilament-dependent cellular processes in many other developing sys- tems, inhibit apical constriction of neuroepithelial cells and neural tube closure and have a "smoothing effect" on apical surfaces of neuroepithelial cells. Although the con- tractility of apical microfilament bundles in neuroepithelial cells has never been demonstrated directly, the following morphological, immunocytochemical and circumstantial evidence strongly suggests that they are ca- pable of developing tension which in turn could serve as an effective source of driving forces for the observed apical constriction of neuroepithelial cells and neural tube formation: l) They are prominent in the apical ends of neuroepithelial cells as discrete bundles during apical constriction (Baker and Schroeder, '67; Sadler et al., '82; Lee et al., '83; Lee and Nagele, '85b; Lee and Nagele, '86; Nagele and Lee, '80, '87). These bundles usually are very straight and regions where they anchor to the plasma membrane often appear "pulled inward," suggesting that they exert constant tractional forces in the apical ends of neuroepithelial cells which could result in apical constriction. 2) They border the apical perimeter of neuroepithelial cells and are oriented in such a way that their contrac-


tion could conceivably bring about the observed apical constriction, apical surface folding, neural fold elevation, and neural tube closure (Baker and Schroeder, '67; Burnside, '73; Messier, '69; Karfunkel, '72, '74; Messier and Auclair, '74; O'Shea and Kaufman, '80; Nagele and Lee, '80; Lee and Nagele, '85b). 3) They are much more prominent (thicker in diameter and more electron dense) in regions of the neuroepithelium exhibiting bending than those in flatter regions such as the neural plate (Burnside, '73; Na- gele and Lee, '80; Lee and Nagele, '85b). 4) They contain the contractile protein, actin (as revealed by their ability to bind heavy meromyosin and indirect immunofluorescence studies) (Nagele and Lee, '80; Sadler et al., '82; Lee et al., '83; Lee and Nagele, '85b). Actin microfilaments have been shown to play a key role in many types of nonmuscle cell motility and the development of tractional forces such as those in stress fibers during spreading-out of cultured cells over the substratum (Pollard and Weihing, '74; Clarke and Spudich, '77; Groschel-Stewart, '80). 5) They contain myosin (as revealed by indirect immunofluorescence studies) (Lee et al., '83; Lee and Nagele, '85b). The coexis- tence of actin and myosin in apical microfilament bundles of neuroepithelial cells makes it almost inconceivable that they are not involved in motive force generation.

A major unsolved problem is how forces arising from changes in the shape of neuroepithelial cells, particularly apical constriction (presumably a result of the contractile activity of apical microfilament bundles), are applied within the neuroepithelium during neural tube formation. As part of our interest in this problem, we have used morphometry to determine regional variations in the rela- tive degree of apical constriction and surface folding at selected regions of the neuroepithelium of stage 8+ chick embryos exhibiting various phases of neural tube formation. These embryos serve as a good model system for the study of the biomechanical basis for neural tube formation because they exhibit a gradual transition of phases (ranging from the relatively flat neural plate to a closed neural tube) along their axes which are similar but not identical to the normal ontogenetic sequence. In the neural plate region, cells located close to the midline have folded apical surfaces and already have undergone a certain degree of apical constriction, apparently representing a first stage in the uplift-

ing of neural folds. In contrast, those forming the lateral two-thirds of the neural plate show smooth and broad apical ends. The transition of the flat neural plate to a V- shaped neuroepithelium is accompanied by an increase in the degree of apical constriction and surface folding of cells located in the midline, whereas those forming the lateral two-thirds of the neuroepithelium remain unchanged and comparable to those in the corresponding region of the neural plate. As- suming that the increased magnitude of these parameters is a consequence of increased contractile activity of apical microfilament bundles (Nagele and Lee, '87), these results suggest that the initial uplifting of neural folds is driven, at least in part, by microfilament-mediated forces arising from cells located close to the midline. During the next phase of neural tube formation, a locus of bending appears in the midlateral walls of the V-shaped neuroepithelium which brings apposing neural folds closer together in such a way that the overall shape resembles the letter "U." Surprisingly, despite the corresponding dramatic increase in the degree of apical constriction and surface folding of cells in the midlateral walls of the U-shaped neuroepithelium, cells situated close to the midline show no further significant increase in either parameter. This finding indicates that transformation of the V- to U-shaped neuroepithelium is driven primarily by cells located in the midlateral walls, whereas cells forming in the medial and lateral thirds of the neuroepithelium do not contribute to this phase of neural tube formation. Bending in the midlateral walls continues until the overall shape of the neuroepithelium resembles the letter "C." This transition is hallmarked by a further increase in the degree of apical surface folding and constriction. In fact, the magnitudes of these parameters reach their highest measured values in this region of the neuroepithelium. However, cells forming the medial and lateral thirds of the neuroepithelium do not appear to undergo significant changes in the apical morphology. This indicates that transformation of the U- to C- shaped neuroepithelium is brought about primarily by forces arising from cells located in the midlateral walls. In the midbrain region, where apposing neural folds approach each other and make contact, the most conspicuous increase in the degree of apical constriction and surface folding over that of the C-shaped neuroepithelium occurs in the up-


per walls. This finding suggests that forces, which contribute to the close apposition and initial contact of neural folds, originate largely from cells in the upper walls. How- ever, the degree of apical constriction exhibited by individual cells in the upper walls is markedly less than that shown by cells in the floor and midlateral walls of the forming neural tube. A possibility exists that, after the neural folds have been successfully uplifted and curled towards each other by forces originating in the floor and midlateral walls, much less force is required to mediate final contact of neural folds.

It is important to emphasize that the work presented here supports the idea that all neuroepithelial cells potentially contribute to shaping the neuroepithelium into a tube. However, the magnitude of their individual contributions appears to depend on their location within the wall of the neuroepithelium. It is interesting to note that, during the transition of the V- to U-shaped neuroepithelium, a marked increase in the degree of apical surface folding and constriction of cells immediately precedes bending of the neuroepithelium. This is in line with the ob- servations of Schoenwolf and Franks (‘84) and Schoenwolf (‘85) and essentially eliminates the possibility that the observed increases in the degree of apical constriction and surface folding of cells are passive consequences of bending of the neuroepithelium. There are two additional important points that need to be raised to put interpretation of these results in their proper perspective. First, the phases of neural tube formation that were the focus of this study were those exhibited along the axes of embryos at a single developmental stage (stage 8 9 . Consequently, we cannot be sure that temporal changes in the distribution of apical constriction-mediated forces reconstructed from these embryos are identical to those involved in the formation of the neural tube during stages 5-8 of development. Second, although we predict that the general temporal pattern of applied apical constriction-mediated forces described here is probably used throughout the neuroepithelium, it is likely that each major region of the forming neural tube (forebrain, midbrain, hindbrain, and spinal cord) has its own characteristic deviation from the general pattern. A more detailed examination of the distribution of apical constriction-mediated forces of each major region of the forming neural tube during stages 5-9 of

development would result in a better under- standing of neural tube formation. Third, local extrinsic forces originating from a number of different sources along the embryo axis may make an important but as yet indeterminate contribution to the observed changes in shape of the neuroepithelium during neural tube formation.

The present study further emphasizes that the transition of the neural plate to the closed neural tube is a complex morphogenetic event which involves a highly coordinated, sequential application of forces within the developing neuroepithelium. Our results indicate that the application of these forces occurs successively in the forming neural tube, first appearing in the future floor, fol- lowed by the midlateral walls, and lastly by the upper walls. The underlying basis for this medial-to-lateral sequence is unknown. However, a possibility exists that it may re- flect the fact that, during the formation of the neural plate, presumptive neuroepithelial cells lying immediately above the chordamesoderm (i.e., the putative neural in- ducer) appear to be slightly more advanced than those located more laterally. Conse- quently, a medial-to-lateral developmental gradient is already evident early in the formation of the neural plate. Further work will be needed to determine whether or not this is the case.

ACKNOWLEDGMENTS

We thank Gwen Harley, Nancy Driscoll, and Mary Kosciuk for their excellent techni- cal assistance. This project was supported by grants from the NIH (NS23200 and NS21730) and the Busch Fund of Rutgers University.

LITERATURE CJTED

Baker, P.C., and T.E. Schroeder (1967) Cytoplasmic fila- ments and morphogenetic movements in the amphibian neural tube. Dev. Biol., 15:432-450.

Burnside, B. (1973) Microtubules and microfilaments in amphibian neurulation. Am. Zool., 13:989-1006.

Campbell, L.R., D. Dayton, and G.S. Soh1 (1986) Neural tube defects: A review of human and animal studies on the etiology of neural tube defects. Teratology, 34: 17 1-187.

Clarke, M., and J.A. Spudich (1977) Nonmuscle contractile proteins: The role of actin and myosin in cell motility and shape determination. Annu. Rev. Biochem., 46t797-822.

Desmond, M.E., and G.C. Schoenwolf (1986) Evaluation of the roles of intrinsic and extrinsic forces in occlusion of the spinal neurocoel during rapid brain enlargement in the chick embryo. J. Embryol. Exp. Morphol., 97:25- 46.

Gordon, R. (1985) A review of the theories of vertebrate

436 R.G. NAGE

neurulation and their relationship to the mechanics of neural tube birth defects. J. Embryol. Exp. Morphol., 89229-255.

Groschel-Stewart, U. (1980) Immunocytochemistry of cytoplasmic contractile proteins. Int. Rev. Cytol., 65193- 254.

Hamburger, V., and H.L. Hamilton (1951) A series of normal stages in the development of the chick embryo. J. Morphol., 88:49-92.

Holtfreter, d. (1947) Changes of structure and the kinet- ics of differentiating embryonic cells. J. Morphol., 80157-91.

Jacobson, A.G., and R. Gordon (1976) Changes in the shape of the developing nervous system analysed ex- perimentally, mathematically, and by computer simu- lation. J. Exp. Zool., 1971191-246.

Jacobson, A.G., G.M. Odell, and G.F. Oster (1985) The cortical tractor model for epithelial folding: Applica- tion to the neural plate. In: Molecular Determination of Animal Form, UCLA Symposium on Molecular and Cellular Biology (new series), vol. 31, G.M. Edelman (ed.), Alan R. Liss, Inc., New York, pp. 143-167.

Jacobson, A.G., G. Oster, G.M. Odell, and L.Y. Cheng (1986) Neurulation and the cortical tractor model for epithelial folding. J. Embryol. Exp. Morphol., 96:19- 49.

Karfunkel, P. (1972) The activity of microtubules and microfilaments in neurulation in the chick. J. Exp. Zool., 181:289-302.

Karfunkel, P. (1974) The mechanisms of neural tube formation. Int. Rev. Cytol., 38:245-271.

Lee, H., and G.W. Kalmus (1976) Effects of cytochalasin B on the morphogenesis of explanted early chick embryos. Growth, 40:153-162.

Lee, H., R.G. Nagele, and N. Karasanyi (1978) Inhibition of neural tube closure by ionophore A23187 in chick embryos. Experientia, 341518-519.

Lee, H., and R.G. Nagele 11979) Neural tube closure defects caused by papaverine in explanted early chick embryos. Teratology, 201321-332.

Lee, H., M.C. Kosciuk, R.G. Nagele, and F.J. Roisen (1983) Studies on the mechanisms of neurulation in the chick Possible involvement of myosin in elevation of neural folds. J. Exp. Zool., 225r449-457.

Lee, H., and R.G. Nagele (1985a) Neural tube defects caused by local anesthetics in early chick embryos. Teratology, 32 I 119-127.

Lee, H., and R.G. Nagele (1985b) Studies on the mechanisms of neurulation in the chick Interrelationship of contractile proteins, microfilaments, and the shape of neuroepithelial cells. J . Exp. Zool., 235.205-215,

Lee, H., and R.G. Nagele (1985,) Possible involvement of calmoddin in apical constriction of neuroepithelial cells and elevation of neural folds in the chick. Exper- ientia, 42; 1186-1188.

Lee, H., and R.G. Natzele (1986) Toxic and teratoloeic

,LE ET AL.

0- ~ ~.

effects of verapamil %n early chick embryos. Teratol- ogy, 331203-211.

Messier, P.E. (1969) Effects of beta-mercaptoethanol on the fine structure of the neural plate cells of the chick embryo. J. Embryol. Ekp. Morphol., 21:309-329.

Messier, P.E., and C. Auclair (1974) Effects of cytochalasin B on interkinetic nuclear migration in the chick embryo. Dev. Biol., 36:218-223.

Moran, D. (1976) Scanning electron microscopic and flame spectrometry study on the role of Ca2+ in amphibian neurulation using papaverine inhibition and ionophore induction of morphogenetic movements. J. Exp. Zool., 198:409-416.

Moran, D., and R.W. Rice (1976) Action of papaverine and ionophore A23187 on neurulation. Nature, 261r497-499.

Morriss-Kay, G.M. (1981) Growth and development of patterns in the cranial neural epithelium of rat embryos during neurulation. J . Embryol. Exp. Morphol., 651225-241.

Nagele, R.G., and H. Lee (1979) Ultrastructural changes in cells associated with interkinetic nuclear migration in the developing chick neuroepithelium. J. Exp. Zool., 210189-106.

Nagele, R.G., and H. Lee (1980) Studies on the mechanisms of neurulation in the chick Microfilament-mediated changes in cell shape during uplifting of neural

Nagele, R.G., and H. Lee (1987) Studies on the mechanisms of neurulation in the chick: Morphometric analysis of the relationship between regional variations in cell shape and sites of motive force generation. J . Exp. Zool., 241:197-205.

Nagele, R.G., J.F. Pietrolungo, H. Lee, and F.J. Roisen (1981) Diazepam-induced neural tube defects in explanted early chick embryos. Teratol., 23t343-349.

O'Shea, K.S., and M.H. Kaufman (1980) Neural tube closure defects following in vitro exposure of mouse embryos to xylocaine. J. Exp. Zool., 214:235-238.

Pollard, T.D., and R.R. Weihing (1974) Actin and myosin and cell movement. CRC Crit. Rev. Biochem., 2:1-65.

Sadler, T.W., D. Greenberg, P. Coughlin, and L. Lessard (1982) Actin distribution patterns in the mouse neural tube during neurulation. Science, 215:172-174.

Schoenwolf, G.C. (1982) On the morphogenesis of the early rudiments of the developing central nervous system. Scanning Elec. Microsc. 1:289-308.

Schoenwolf, G.C., and M.V. Franks (1984) Quantitative analyses of changes in cell shapes during bending of the avian neural plate. Dev. Biol., 105:257-272.

Schoenwolf, G.C. (1985) Shaping and bending of the avian neuroepithelium: Morphometric analysis. Dev. Biol., I091 127-139.

Schoeder. T.E. (1973) Cell constriction: Contractile role

folds. J. EXP. ZOO^., 213:391-398.

of microfilaments in division and development. Am. Zool., 13:949-960.

Waddmgton, C.H., and M.M. Perry (1966) A note on the mechanism of cell deformation in the neural folds of the amphibian. Exp. Cell Res., 41:691-693.

studies on the mechanisms of neurulation in the chick: morphometric analysis of force distribution...

Documents