U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Gilbert SF. Developmental Biology. 6th edition. Sunderland (MA): Sinauer Associates; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Developmental Biology

Developmental Biology. 6th edition.

Show details

Tissue Architecture of the Central Nervous System

The neurons of the brain are organized into layers (cortices) and clusters (nuclei), each having different functions and connections. The original neural tube is composed of a germinal neuroepithelium that is one cell layer thick. This is a layer of rapidly dividing neural stem cells. Sauer (1935) and others have shown that all the cells of the germinal epithelium are continuous from the luminal surface of the neural tube to the outside surface, but that the nuclei of these cells are at different heights, thereby giving the superficial impression that the neural tube has numerous cell layers. The nuclei move within their cells as they go through the cell cycle. DNA synthesis (S phase) occurs while the nucleus is at the outside edge of the neural tube, and the nucleus migrates luminally as the cell cycle proceeds (Figure 12.15). Mitosis occurs on the luminal side of the cell layer. If mammalian neural tube cells are labeled with radioactive thymidine during early development, 100% of them will incorporate this base into their DNA (Fujita 1964). Shortly thereafter, certain cells stop incorporating these DNA precursors, thereby indicating that they are no longer participating in DNA synthesis and mitosis. These neuronal and glial cells then migrate and differentiate outside the neural tube (Fujita 1966; Jacobson 1968).

Figure 12.15. (A) Schematic section of a chick embryo neural tube, showing the position of the nucleus in a neuroepithelial cell as a function of the cell cycle.

Figure 12.15

(A) Schematic section of a chick embryo neural tube, showing the position of the nucleus in a neuroepithelial cell as a function of the cell cycle. Mitotic cells are found near the center of the neural tube, adjacent to the lumen. (B) Scanning electron (more...)

If dividing cells in the germinal neuroepithelium are labeled with radioactive thymidine at a single point in their development, and their progeny are found in the outer cortex in the adult brain, then those neurons must have migrated to their cortical positions from the germinal neuroepithelium. This happens because a neuroepithelial stem cell divides “vertically” instead of “horizontally.” Thus, the daughter cell adjacent to the lumen remains connected to the ventricular surface (and usually remains a stem cell), while the other daughter cell migrates away (Chenn and McConnell 1995). The time of this vertical division is the last time the latter cell will divide, and is called that neuron's birthday. Different types of neurons and glial cells have their birthdays at different times. Labeling at different times during development shows that cells with the earliest birthdays migrate the shortest distances. The cells with later birthdays migrate through these layers to form the more superficial regions of the cortex. Subsequent differentiation depends on the positions these neurons occupy once outside the germinal neuroepithelium (Letourneau 1977; Jacobson 1991).

Spinal chord and medulla organization

As the cells adjacent to the lumen continue to divide, the migrating cells form a second layer around the original neural tube. This layer becomes progressively thicker as more cells are added to it from the germinal neuroepithelium. This new layer is called the mantle (or intermediate) zone, and the germinal epithelium is now called the ventricular zone (and, later, the ependyma) (Figure 12.16). The mantle zone cells differentiate into both neurons and glia. The neurons make connections among themselves and send forth axons away from the lumen, thereby creating a cell-poor marginal zone. Eventually, glial cells cover many of the axons in the marginal zone in myelin sheaths, giving them a whitish appearance. Hence, the mantle zone, containing the neuronal cell bodies, is often referred to as the gray matter; the axonal, marginal layer is often called the white matter.

Figure 12.16. Differentiation of the walls of the neural tube.

Figure 12.16

Differentiation of the walls of the neural tube. A section of a 5-week human neural tube (left) reveals three zones: ventricular (ependymal), intermediate (mantle), and marginal. In the spinal cord and medulla (top row), the ependymal zone remains the (more...)

In the spinal cord and medulla, this basic three-zone pattern of ependymal, mantle, and marginal layers is retained throughout development. The gray matter (mantle) gradually becomes a butterfly-shaped structure surrounded by white matter; and both become encased in connective tissue. As the neural tube matures, a longitudinal groove—the sulcus limitans— divides it into dorsal and ventral halves. The dorsal portion receives input from sensory neurons, whereas the ventral portion is involved in effecting various motor functions (Figure 12.17).

Figure 12.17. Development of the human spinal cord.

Figure 12.17

Development of the human spinal cord. (A-D) The neural tube is functionally divided into dorsal (alar) and ventral (basal) regions, separated by the sulcus limitans. As cells from the adjacent somites form the spinal vertebrae, the neural tube differentiates (more...)

Cerebellar organization

In the brain, cell migration, differential neuronal proliferation, and selective cell death produce modifications of the three-zone pattern (Figure 12.16). In the cerebellum, some neuronal precursors enter the marginal zone to form clusters of neurons called nuclei. Each nucleus works as a functional unit, serving as a relay station between the outer layers of the cerebellum and other parts of the brain. In the cerebellum, some neuronal precursors can also migrate away from the germinal epithelium. These precursor cells, called neuroblasts, migrate to the outer surface of the developing cerebellum and form a new germinal zone, the external granule layer, near the outer boundary of the neural tube. At the outer boundary of the external granule layer (which is one to two cells thick), neuroblasts proliferate. The inner compartment of the external granule layer contains postmitotic neuroblasts that are the precursors of the major neurons of the cerebellar cortex, the granule neurons. These granule neurons migrate back into the developing cerebellar white matter to produce a region called the internal granule layer. Meanwhile, the original ependymal layer of the cerebellum generates a wide variety of neurons and glial cells, including the distinctive and large Purkinje neurons. Purkinje neurons are not only critical in the electrical pathway of the cerebellum, they also support the granule neurons. The Purkinje cell secretes Sonic hedgehog, which sustains the division of granule neuron precursors in the external granule layer (Wallace 1999). Each Purkinje neuron has an enormous dendritic arbor, which spreads like a tree above a bulblike cell body. A typical Purkinje neuron may form as many as 100,000 connections (synapses) with other neurons, more than any other neuron studied. Each Purkinje neuron also emits a slender axon, which connects to neurons in the deep cerebellar nuclei.

The development of spatial organization is critical for the proper functioning of the cerebellum. All electrical impulses eventually regulate the activity of the Purkinje cells, which are the only output neurons of the cerebellar cortex. For this to happen, the proper cells must differentiate at the appropriate place and time. How is this accomplished?

One mechanism thought to be important for positioning young neurons within the developing mammalian brain is glial guidance (Rakic 1972; Hatten 1990). Throughout the cortex, neurons are seen to ride “the glial monorail” to their respective destinations. In the cerebellum, the granule cell precursors travel on the long processes of the Bergmann glia (Figure 12.18; Rakic and Sidman 1973; Rakic 1975). This neural-glial interaction is a complex and fascinating series of events, involving reciprocal recognition between glia and neuroblasts (Hatten 1990; Komuro and Rakic 1992). The neuron maintains its adhesion to the glial cell through a number of proteins, one of them an adhesion protein called astrotactin. If the astrotactin on a neuron is masked by antibodies to that protein, the neuron will fail to adhere to the glial processes (Edmondson et al. 1988; Fishell and Hatten 1991).

Figure 12.18. Neuronal migration on glial cell processes.

Figure 12.18

Neuronal migration on glial cell processes. (A) Diagram of a cortical neuron migrating on a glial cell process. (B) Electron micrograph of the region where the neuronal cell body adheres to the glial process. (C) Sequential photographs of a neuron migrating (more...)

Much insight into the mechanisms of spatial ordering in the brain has come from the analysis of neurological mutations in mice. Over 30 mutations are known to affect the arrangement of cerebellar neurons. Many of these cerebellar mutants have been found because the phenotype of such mutants—namely, the inability to keep balance while walking—can be easily recognized. For obvious reasons, these mutations are given names such as weaver, reeler, staggerer, and waltzer.

WEBSITE

12.5 Cerebellar mutations of the mouse. The mouse mutations affecting cerebellar function have given us remarkable insights into the ways in which the cerebellum is constructed. The reeler mutation, in particular, has been extremely important in our knowledge of how cerebellar neurons migrate. http://www.devbio.com/chap12/link1205.shtml

Cerebral organization

The three-zone arrangement of the neural tube is also modified in the cerebrum. The cerebrum is organized in two distinct ways. First, like the cerebellum, it is organized vertically into layers that interact with one another (see Figure 12.16). Certain neuroblasts from the mantle zone migrate on glial processes through the white matter to generate a second zone of neurons at the outer surface of the brain. This new layer of gray matter is called the neocortex. The neocortex eventually stratifies into six layers of neuronal cell bodies; the adult forms of these layers are not completed until the middle of childhood. Each layer of the neocortex differs from the others in its functional properties, the types of neurons found there, and the sets of connections that they make. For instance, neurons in layer 4 receive their major input from the thalamus (a region that forms from the diencephalon), while neurons in layer 6 send their major output back to the thalamus.

Second, the cerebral cortex is organized horizontally into over 40 regions that regulate anatomically and functionally distinct processes. For instance, neurons in cortical layer 6 of the “visual cortex” project axons to the lateral geniculate nucleus of the thalamus (see Chapter 13), while layer 6 neurons of the auditory cortex (located more anteriorly than the visual cortex) project axons to the medial geniculate nucleus of the thalamus (for hearing).

Neither the vertical nor the horizontal organization of the cerebral cortex is clonally specified. Rather, the developing cortex forms from the mixing of cells derived from numerous stem cells. After their final mitosis, most of the neuronal precursors generated in the ventricular (ependymal) zone migrate outward along glial processes to form the cortical plate at the outer surface of the brain. As in the rest of the brain, those neuronal precursors with the earliest “birthdays” form the layer closest to the ventricle. Subsequent neurons travel greater distances to form the more superficial layers of the cortex. This process forms an “inside-out” gradient of development (Figure 12.19; Rakic 1974). A single stem cell in the ventricular layer can give rise to neurons (and glial cells) in any of the cortical layers (Walsh and Cepko 1988). But how do the cells know which layer to enter?

Figure 12.19. “Inside-out” gradient of cerebral cortex formation in the rhesus monkey.

Figure 12.19

“Inside-out” gradient of cerebral cortex formation in the rhesus monkey. The “birthdays” of the cortical neurons were determined by injecting pregnant animals intravenously with [3H]-thymidine at certain times of gestation. (more...)

McConnell and Kaznowski (1991) have shown that the determination of laminar identity (i.e., which layer a cell migrates to) is made during the final cell division. Newly generated neuronal precursors transplanted after this last division from young brains (where they would form layer 6) into older brains whose migratory neurons are forming layer 2 are committed to their fate, and migrate only to layer 6. However, if these cells are transplanted prior to their final division (during mid-S phase), they are uncommitted, and can migrate to layer 2 (Figure 12.20). The fates of neuronal precursors from older brains are more fixed. While the neuronal precursor cells formed early in development have the potential to become any neuron (at layers 2 or 6, for instance), later precursor cells give rise only to upper-level (layer 2) neurons (Frantz and McConnell 1996). We still do not know the nature of the information given to the cell as it becomes committed.

Figure 12.20. Determination of laminar identity in the ferret cerebrum.

Figure 12.20

Determination of laminar identity in the ferret cerebrum. (A) “Early” neuronal precursors (birthdays on embryonic day 29) migrate to layer 6. (B) “Late” neuronal precursors (birthdays on postnatal day 1) migrate farther, (more...)

Not all neurons, however, migrate radially. O'Rourke and her colleagues (1992) labeled young ferret neurons with fluorescent dye and followed their migration through the brain. While a great majority of the young neurons migrated radially on glial processes from the ventricular zone into the cortical plate, about 12% of them migrated laterally from one functional region of the cerebral cortex into another. These observations meshed well with those of Walsh and Cepko (1992), who infected ventricular stem cells with a retrovirus and were able to stain these cells and their progeny after birth. They found that the neural descendants of a single ventricular stem cell were dispersed across the functional regions of the cortex. Thus, the specification of the cortical areas into specific functional domains occurs after neurogenesis. Once the cells arrive at their final destination, it is thought that they produce particular adhesion molecules that organize them together as brain nuclei (Matsunami and Takeichi 1995).

The cerebrum is quite plastic. The development of the human neocortex is particularly striking in this regard. The human brain continues to develop at fetal rates even after birth (Holt et al. 1975). Based on morphological and behavioral criteria and on comparisons with other primates, Portmann (1941,1945) suggested that human gestation should really last 21 months instead of 9. However, no woman could deliver a 21-month-old fetus because the head would not pass through the birth canal; thus, humans give birth at the end of 9 months. Montagu (1962) and Gould (1977) have suggested that during our first year of life, we are essentially extrauterine fetuses, and they speculate that much of human intelligence comes from the stimulation of the nervous system as it is forming during that first year.*

WEBSITE

12.6 Constructing the cerebral cortex. Three genes have recently been shown to be necessary for the proper lamination of the mammalian brain. They appear to be important for cortical neural migration, and when mutated in humans can produce profound mental retardation. http://www.devbio.com/chap12/link1206.shtml

WEBSITE

12.7 What makes us human. Human brain size is off the scale when we compare ourselves with other apes. Our retention of fetal neural stem cell division rates for a year after birth may be the critical step in our becoming human. http://www.devbio.com/chap12/link1207.shtml

Adult neural stem cells

Until recently, it had been generally thought that once the nervous system was mature, no new neurons were “born.” The neurons we formed in utero and during the first few years of life were all we could ever expect to have. However, the good news from recent studies is that environmental stimulation can increase the number of new neurons in the mammalian brain (Kemperman et al. 1997a,b; Gould et al. 1999a,b; Praag et al. 1999). To do these experiments, researchers injected adult mice, rats, or marmosets with bromodeoxyuridine (BrdU), a nucleoside that resembles thymidine and which will be incorporated into a cell's DNA only if the cell is undergoing DNA replication. Thus, any cell labeled with BrdU must have been undergoing mitosis during the time when it was exposed to BrdU. This technique showed that thousands of new neurons were being made each day in adult mammals. Injecting humans with BrdU is usually unethical, since large doses of BrdU are often lethal. However, in certain cancer patients, the progress of chemotherapy is monitored by transfusing the patient with a small amount of BrdU. Gage and colleagues (Erikkson et al. 1998) took postmortem samples from the brains of five such patients who had died between 16 and 781 days after the BrdU infusion. In all five subjects, they saw new neurons in the granular cell layer of the hippocampal dentate gyrus (a part of the brain where memories may be formed). The BrdU-labeled cells also stained for neuron-specific markers (Figure 12.21). It appears that the stem cells producing these neurons are located in the ependyma, the former ventricular layer in which the embryonic neural stem cells once resided (Doetsch et al. 1999; Johansson et al. 1999). These results are surprising, since the ependyma consists of differentiated glial cells whose ciliated surface keeps the cerebral spinal fluid flowing. Indeed, one author (Barres 1999) described them as “the most boring of all glial subtypes.” It appears that these glial cells (or perhaps only some of them) can dedifferentiate and become neural stem cells. Thus, although the rate of new neuron formation in adulthood may be relatively small, the human brain is not an anatomical fait accompli at birth, or even after childhood.

Figure 12.21. Evidence of adult neural stem cells.

Figure 12.21

Evidence of adult neural stem cells. (A) Newly generated neurons in the adult human brain. This cell is located in the dentate gyrus of the hippocampus. The green staining, which indicates newly divided cells, is from a fluorescent antibody against BrdU (more...)

Footnotes

*

Contrary to the claims of a widely circulated anti-abortion film, the human cerebral cortex has no neuronal connections at 12 weeks' gestation (and therefore cannot move in response to a thought, nor experience consciousness or fear). Measurable electrical activity characteristic of neural cells (the electroencephalogram, or EEG, pattern) is first seen at 7 months' gestation. Morowitz and Trefil (1992) put forth the provocative opinion that since society in the United States has defined death as the loss of the EEG pattern, perhaps it should accept the acquisition of the EEG pattern as the start of human life.

The use of cultured neural stem cells to regenerate or repair parts of the brain will be considered in the next chapter.

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, Sinauer Associates.
Bookshelf ID: NBK10047

Views

  • Cite this Page

Related Items in Bookshelf

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...