ABSTRACT
The localization of alkaline phosphatase activity in the lumbosacral region of the developing spinal cord was studied in 9·5-to 17·5-day mouse embryos. The activity was uniformly distributed in the pseudostratified neuroepithelium of the 9·5-day cord. In the 11-5-day cord in which the lateral motor columns were being formed, the enzymatic activity was localized in the ventrolateral sector of the cord. The enzyme-positive ventricular cells tended to be located medially whereas radially oriented enzyme-positive processes extended into the marginal layer. The 13·5-day cord displayed a similar distribution pattern, but there were many more radial processes and the enzyme-positive cells had spread laterally. Close apposition between the processes and the ventricular cells was observed. By 15·5 and 17·5 days, when the intermediate layer was fully developed and the ventricular layer had regressed to a thin ependyma, the activity had become diffusely located in the ventral half of the cord. The enzyme-positive cells and processes became less conspicuous. The silver-stained processes in the cord were found to be organized in an entirely different pattern from that of the enzyme-positive processes, suggesting that the enzyme-positive processes were not neuronal processes. The enzymatic activity found in the developing spinal cord may be associated with the migration of neuroblasts along the radially aligned processes.
INTRODUCTION
The early embryonic neural tube is made up of pseudostratified epithelial cells (Hernan & Kaufman, 1966; Sturrock, 1981a) whose nuclei undergo intermitotic migration (Smart, 1972). The formation of the basic layers of the spinal cord begins at day 10-−11 with the appearance of the ventricular, intermediate and marginal layers (Sturrock, 1981b; Boulder Committee, 1970). The ventricular layer forms the germinal tissue from which neuroblasts and glioblasts of the foetal cord are generated (Smart, 1972; Nemar, Sakla & Mahran, 1974; Wentworth & Hinds, 1978; Sturrock, 1981a). The earliest neurones that appear on day 9 are of medium size and reside in the lateral and the ventral horns (McConnell, 1981). Motor neurones emerging on day 10 and day 11 migrate from the basal plate to the adjacent intermediate layer. Those neurones formed on day 12−14 come from the alar plate and settle in the dorsal portion of the intermediate layer (Nornes & Carry, 1978). Some of the earliest neurones from the alar plate send out circumferential axons which are the forerunners of commissural fibres (Holley, 1982). The mechanism whereby this pattern of histogenesis is brought about in the developing cord is not fully known. However, it has been proposed that the organization of layers of neural cells and the formation of neuronal connections are dependent on the information perceived by the neuroblasts via contact guidance with the pre-existing fascicles of cell processes (Sidman & Rakic, 1973; Henrikson & Vaughn, 1974; Rakic, 1978; Holley, Nornes & Morita, 1982).
The histogenesis of the spinal cord has been correlated with changes in the biochemical profiles of the various cell types involved in this developmental process. For example, the formation of morphologically discernible neurones in the neural tube of early embryos is related to the distribution of acetylcholinesterase activities (Burt, 1975; Miki & Mizoguti, 1982). Acid phosphatase is mainly localized in the sensory neurones in the dorsal horn of the rat spinal cord (Knyihar & Csillik, 1977) and this group of enzyme is present in different forms in the substantia gelatinosa and in the motor neurones (Sanyal & Rustioni, 1974). The activity of alkaline phosphatase has been localized in the developing cord of the human foetuses. At an early stage when the cord is entirely composed of the ventricular layer, the enzymatic activity is present in every cell but cells in the ventricular layer show a stronger activity than those in the intermediate and marginal layers. With development the enzymatic activity progressively diminishes and by 4−5 months the enzymatic activity becomes confined to the blood vessels of choroid plexuses and meninges of the brain and spinal cord (Rossi & Reale, 1957).
Alkaline phosphatase activity has been found in many tissues of the adult spinal cord and brain of the mouse and the rat, primarily on the wall of arterial vessels (Shimizu, 1950; Leduc & Wislocki, 1952). Adult neurones generally show a moderate enzymatic activity (Sanyal & Rustioni, 1974; Sood & Mulchandani, 1977); however, no information is available for early embryonic mouse neural tube. In other embryonic tissues, alkaline phosphatase is initially present in a high concentration during the early stage of cell differentiation, decreases during differentiation and stabilizes at a low adult level which usually shows no regional pattern of distribution. A high enzymatic activity is present in embryonal carcinoma cells (Bluthman et al. 1983) and in embryonic ectoderm (Damjanov, Solter & Skreb, 1971); in both cases, the differentiated derivatives exhibit a low enzyme level (Bernstine, Hooper, Grandchamp & Ephrussi, 1973; Solter, Damjanov & Skreb, 1973). A strong enzymatic activity has been found at the early organogenetic stage in the somitic mesoderm, intestinal epithelium and migrating primordial germ cells (Rossi & Reale, 1957; Tam & Snow, 1981). The present study is to investigate the developmental changes of alkaline phosphatase activity in the foetal spinal cord. This enzyme pattern is interpreted in relation to the migration of neuroblasts and the formation of the neuronal pattern.
MATERIALS AND METHODS
ICR strain mouse embryos at 9·5−17·5 days of gestation were used and the morning when the vaginal plug was detected was designated as 0·5 days post coitus. The pregnant mice were killed by cervical dislocation and the embryos were dissected out of the uterus into cold phosphate-buffered saline. Whole embryos at 9·5−13·5 days, and the lumbosacral region of the embryos at 15·5−17·5 days, which was identified as the region corresponding to the hindlimb, and of the adult spinal cord were processed according to the following methods. For general histology of the spinal cord, the specimens were fixed in Bouin’s fluid, and serial transverse sections were obtained and stained by Mallory’s acid fuschin-aniline blue-orange G triple staining method. To stain neuronal processes, the specimens were fixed with DeCastro’s fixative (100 ml distilled water, 100 ml 95% ethanol, 6gm chloral hydrate, 3 ml concentrated nitric acid) and then processed with a modified silver method based on Cajal and DeCastro (Levi-Montalcini, 1949). For the demonstration of alkaline phosphatase, serial paraffin sections of ethanol-fixed tissues were processed with the azo-coupling technique (Gabe, 1975), using sodium a-naphthyl phosphate as the substrate and Fast TR Red salt for the coupling-staining reaction. Control for the histochemical reaction was performed either by adding tetramisole hydrochloride (1 mM; Borgers, 1973) or EDTA (10 mM) to the incubation medium to inhibit the enzyme, or by removing the substrate from the medium.
RESULTS
When tetramisole hydrochloride was added to the incubation medium, no enzymatic reaction was detected in any part of the spinal cord at all stages. Removal of the substrate from the incubation medium or addition of EDTA to the medium also drastically reduced the intensity of the reaction. The control thus indicated that the activity of alkaline phosphatase was responsible for the staining reaction.
The neural tube of the 9·5-day embryo was made up of a tall pseudostratified neuroepithelium, which had not differentiated into the definitive layers. A strong alkaline phosphatase activity was found uniformly distributed throughout the transverse section of the tube (Fig. 1). No stained processes were evident in the silver preparations.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The differentiation of neuroblasts occurred primarily in the basal plate of the neural tube so that by 11·5 days, the marginal layer and the intermediate layer containing the lateral motor column had developed (Fig. 7). A band with a strong alkaline phosphatase activity was observed in the ventricular layer of the basal plate (Fig. 2). It extended from the luminal surface to the vicinity of the dorsal half of the lateral motor column. The lateral motor column itself showed little activity. The band consisted of enzyme-positive ventricular cells and radially aligned processes (Figs 8, 9). In the marginal layer, a high enzymatic activity was found in the craniocaudally oriented processes (Fig. 9), which appeared to be continuous with the radial processes. In addition, a few strands were seen along the border between the ventricular layer and the lateral motor column. The remaining part of the spinal cord did not show any enzymatic activity
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
The pattern of distribution of alkaline phosphatase activity in the spinal cord at different stages: 9·5-day (Fig. 1), 11·5-day (Fig. 2), 13·5-day (Fig. 3), 15·5-day (Fig. 4) and 17·5-day (Fig. 5). In the adult spinal cord, the blood capillaries (arrow heads) showed a strong activity (Fig. 6). The boundary between the ventricular (ependymal) and the intermediate layers was demarcated by a broken line, ml = marginal layer. de = dorsal columns, sg = substantia gelatinosa. Bar equals 100 μm.
Silver staining revealed the presence of many processes which were organized into five groups according to their location and orientation (Fig. 10). It can be seen that the pattern of silver-positive processes differed notably in two aspects from tie pattern of alkaline phosphatase activity. Firstly, no silver-positive processes were present in the strong enzyme-positive band of the ventricular layer, and secondly, in the lateral motor column, the silver-positive processes were de rsoventrally oriented whereas the enzyme-positive processes were radially oriented.
Triple-stained transverse section of the 11·5-day cord. The ventricular layer with mediolaterally aligned cells was demarcated (broken line) from the intermed ate layer. The marginal layer (ml) was well developed on the periphery of the prominent lateral motor column (Imc). Bar equals 50 μm.
Alkaline phosphatase activity in transverse (Fig. 8) and horizontal (Fig. 9) sections of the 11·5-day cord. A few processes (arrow) arose from the band of enzyme-positive cells in the ventricular layer and extended across the lateral motor column (Imc) to the marginal layer (ml). Bar equals 50 μm.
Alkaline phosphatase activity in transverse (Fig. 8) and horizontal (Fig. 9) sections of the 11·5-day cord. A few processes (arrow) arose from the band of enzyme-positive cells in the ventricular layer and extended across the lateral motor column (Imc) to the marginal layer (ml). Bar equals 50 μm.
Silver-stained transverse section of the 11·5-day cord, showing five groups of neuronal processes: (A) Dorsolateral group, consisting of craniocaudally oriented processes at the dorsolateral margin of the spinal cord. (B) Ventrolateral group, cons sting of craniocaudally oriented processes on the ventrolateral margin of the motor column. (C) Lateral motor column processes, which were dorsoventrally orier .ted and were aggregated on the lateral part of the motor column. (D) Circumferer .tial group, which intervened between the ventricular layer and the lateral motor column, and crossed to the opposite side ventral to the neural canal; some of these processes diverged from the circumferential pathway to enter the lateral motor column as Group C processes. (E) Ventral root axons. Bar equals 50 μm.
Silver-stained transverse section of the 11·5-day cord, showing five groups of neuronal processes: (A) Dorsolateral group, consisting of craniocaudally oriented processes at the dorsolateral margin of the spinal cord. (B) Ventrolateral group, cons sting of craniocaudally oriented processes on the ventrolateral margin of the motor column. (C) Lateral motor column processes, which were dorsoventrally orier .ted and were aggregated on the lateral part of the motor column. (D) Circumferer .tial group, which intervened between the ventricular layer and the lateral motor column, and crossed to the opposite side ventral to the neural canal; some of these processes diverged from the circumferential pathway to enter the lateral motor column as Group C processes. (E) Ventral root axons. Bar equals 50 μm.
In the 13·5-day embryo, both the intermediate and the marginal layers were well developed on both the ventral and dorsal parts of the cord. The ventricular layer remained thick dorsally, but had become much narrower on its ventral part. The cord contained a transverse band of strong alkaline phosphatase activity (Fig. 3) in a similar location as the 11·5-day cord. However, this band extended from the luminal surface through the dorsal part of the lateral motor column and reached the marginal layer (Figs 11−13). The same five groups of silver-stained processes as those in the 11·5-day cord were observed in the 13·5-day cord (Fig. 14). In the marginal layer, the dorsolateral and ventrolateral groups of processes had thickened. The originally dorso ventrally oriented processes of the lateral motor column had become obliquely oriented and spread out to the whole motor column.
A transverse section of the 13·5-day cord, showing the enzyme-positive transverse band and its radial processes (arrow). lmc = lateral motor column. ml = marginal layer. Bar equals 50 μm.
A horizontal section of the 13 ·5-day cord. Numerous enzyme-positive radial processes (arrow) of the transverse band joined the craniocaudally oriented processes (arrow head) of the marginal layer (ml). Bar equals 50μm.
A horizontal section of the 13·5-day cord, showing enzyme-positive processes traversing through stained and unstained cells from the luminal side (lu) to th2 marginal layer (ml). The cells with high enzymatic activity were especially dens< fly packed near the luminal side. The processes appeared in close apposition with adjat ent cells where the cell surface often appeared flattened. Bar equals 25 μm.
A horizontal section of the 13·5-day cord, showing enzyme-positive processes traversing through stained and unstained cells from the luminal side (lu) to th2 marginal layer (ml). The cells with high enzymatic activity were especially dens< fly packed near the luminal side. The processes appeared in close apposition with adjat ent cells where the cell surface often appeared flattened. Bar equals 25 μm.
A silver-stained half-section of the 13·5-day cord, and a camera-lucida drawing to illustrate the stained fine processes. Groups of processes are labelled similarly to those in the 11·5-day cord. Bar equals 50 gm.
Further differentiation resulted in an increase in thickness of the intermediate layer in the 15·5-day cord, while the ventricular layer became a thin ependymal layer. The cord at this stage differed markedly from that of the 13·5-day embryo in both the distribution and intensity of the alkaline phosphatase activity. Instead of being confined to a transverse band, the activity was distributed throughout the ventral two-thirds of the cord (Fig. 4). The overall activity was substantially lower than that of the transverse band in the previous stage and only a few cells, mostly ependymal cells, were stained. From the ependyma, faintly to moderately stained processes extended laterally and ventrally (Fig. 15). The remaining dorsal part of the cord and the floor plate area did not have any detectable activity.
Alkaline phosphatase activity in the ventral part of the 15·5-day cord, which showed radial processes (arrow head) and mediolaterally oriented processes (arrow). Some of the processes extended to the marginal layer (ml) and formed partirions (pt) between groups of longitudinal processes. Bar equals 50 μm.
Alkaline phosphatase activity in the ventral part of the 15·5-day cord, which showed radial processes (arrow head) and mediolaterally oriented processes (arrow). Some of the processes extended to the marginal layer (ml) and formed partirions (pt) between groups of longitudinal processes. Bar equals 50 μm.
In the silver preparation, processes of various orientations were organized in the same basic pattern as those in the 13·5-day cord, except that a well-defined circumferential group was no longer recognizable. Like the earlier stages, the organization of the stained processes did not correspond at all to that of the enzyme-positive processes.
By 17·5 days, the neural canal and the ependymal layer had become restricted to the ventral part of the cord. The alkaline phosphatase activity at this stage was located in a meshwork of neuropil in the ventral half of the cord and in the longitudinal processes of the marginal layer (Fig. 5). A stronger activity was found on either side of the ventral white commissure and in the marginal layer. The organization of silver-stained processes followed the same pattern as that in the 15·5-day cord.
The alkaline phosphatase activity in the adult spinal cord was low, except for the substantia gelatinosa which was weakly stained (Fig. 6).
DISCUSSION
The distribution of the alkaline phosphatase activity changed during the development of the embryonic neural tube. At the earliest stage, the enzymatic activity was found in the neuroepithelial cells. With further development, the activity became confined to a band of cells and processes in the ventricular layer of 11·5- and 13·5-day spinal cord. The enzymatic activity remained evident in the ventral grey matter and the adjacent marginal layer of the late foetal spinal cord but the overall activity had diminished. The enzymatic activity is generally low in the adult mouse spinal cord except for the cells of blood capillaries. The association of the enzyme with the ependyma of the neural canal and the blood vessels in the grey horns, the choroid plexus and the meninges of the adult central nervous system (Shimizu, 1950; Leduc & Wislocki, 1952; Sood & Mulchandani, 1977) has led to the suggestion that the enzymatic activity is related to the trophic function of the blood-neural tissue barrier and that the enzyme pattern in the embryonic neural tube reflects the process of angiogenesis (Ciani, Contestabile, Minelli & Quaglia, 1973). However, in the foetal spinal cord there were no significant variations in the vascular density between the dorsal and ventral areas of the spinal cord (Sturrock, 1981b; Simon-Marin, Vilanova, Aguinagalde & Barbera-Guillem, 1983) that could account for a regional localization of the enzymatic activity described in the present study. Furthermore, the enzymatic activity diminished during the late foetal stage when the vascular pattern is maximally developed (Sturrock, 1981b).
A comparison of the pattern of enzyme-positive processes with that of silver-stained processes suggests that the enzyme-positive processes were not neuronal processes. Observations of the present study suggested that the differentiating neuroblasts were in intimate apposition with these enzyme-positive processes. Similar radially oriented processes, which were termed radial glial processes or ependymoglial processes, have been described in the spinal cord and telencephalon of the mouse embryo (Henrikson & Vaughn, 1974; Hinds & Ruffett, 1971; Sturrock, 1981a, 1982; Holley, Nornes & Morita, 1982). Many of these ependymoglial fibres later degenerate or become astrocytes, but some may persist in the dorsal median septum and in the floor plate of the late foetal cord (Sturrock, 1981a). In the lateral marginal layer of the embryonic mouse spinal cord, the growing dendrites from the lateral motor column make a close association with the radial glial processes (Henrikson & Vaughn, 1974). A similar association between the radial processes and the migrating neuroblasts has been observed in the developing cortex of the cerebrum (Sidman & Rakic, 1973). This structural specialization which seems to favour an orderly radial migration of the neuroblasts to a distant position is crucial to the organization of neurones in a cortical column (Rakic, 1978). The exact mechanism for such a contact guidance of neuronal patterning is not known. The ependymoglial processes may provide a mechanical and/or chemical pathway for neurites to follow (Singer, Norlander & Egar, 1979). These glycogen-rich glial processes may also serve as a trophic medium from which the neuroblasts obtain the energy for migration and growth (Ciani, Contestabile, Minelli & Quaglia, 1973; Sturrock, 1981a,b). The elevated level of alkaline phosphatase activity may reflect the active metabolic state of these processes.
During the development of the lumbosacral segment of the spinal cord, a high alkaline phosphatase activity occurred concomitantly with the formation of the ventral grey matter from the basal plate. The ventricular cells of the basal plate always show a lower mitotic activity than the dorsal counterpart (Smart, 1972) and the newly formed ventricular cells spend a short time in the ventricular layer before migrating laterally to form the lateral motor column (Nornes & Carry, 1978 ; McConnell, 1981 ; Sturrock, 1981 a). In the dorsal portion of the neural tube, the thickness of the ventricular layer increases as a result of the accumulation of daughter cells. Transformation of these cells to those of the intermediate layer does not occur progressively as in the ventral portion but takes place rapidly (Smart, 1972; Sturrock, 1981a). The lower enzymatic activity in the dorsal portion may therefore not be due to fewer ependymoglial processes but to a lower rate of neuroblast migration. It may not be coincidental that the band of highest alkaline phosphatase activity was found in the ventral region of the ventricular layer where the mitotic activity is declining (Smart, 1972) and the intermediate layer neuroblasts begin to emerge in recognizable numbers (Sturrock, 1981a). Many radial processes are already existing in the ventricular layer prior to the birth of motor neurones (Nornes & Das, 1974; Wentworth & Hinds, 1978) and they seem to enhance the dispersion of the neuroblasts to the intermediate layer (Holley, Nornes & Morita, 1982). A high metabolic activity of the radial process may be heralding an accelerated exodus of neuroblasts out of the ventricular layer.