|  An
unexpected form of plasticity in the adult brain is the constant generation
of new neurons throughout adult life. The addition of thousands of new
neurons each day may be a means of sculpting brain circuitry. This turnover
has important implications not only for brain function but also for brain
repair and pathological conditions such as tumors. This review will cover
neurogenesis and the identity of stem cells in the adult brain.
The
adult brain contains two main types of cells, neurons and macroglia. Macroglia
comprise oligodendrocytes and astrocytes. Oligodendrocytes are the cells
that form myelin sheaths around the axons of neurons and allow the rapid
conduction of electrical impulses. Astrocytes are mysterious cells that
have multiple functions, including maintaining homeostasis, absorbing
neurotransmitters released by neurons, and providing trophic support.
Astrocytes are also induced to divide in response to lesions of the brain
resulting in the formation of glial scars.
Almost
all neurons and glia in the brain are derived from multipotent stem cells
located in the germinal layers next to the ventricular cavities of the
developing brain. As development proceeds, precursors become progressively
more restricted into neuronal or glial lineages. After their birth, neurons
use two kinds of migration to reach the appropriate layers of the brain.
In radial migration, neurons migrate along the processes of radial glial
cells, which span the width of the developing brain. In nonradial migration,
cells disperse without the need for radial glia. Around the time of birth,
radial glia lose their long processes and are transformed into astrocytes
with their characteristic branched star-shaped morphology. Although neurogenesis
is mostly restricted to embryonic development, glial cells continue to
be generated throughout adulthood.
It
has been known for about a century that the subventricular zone (SVZ),
also called the subependymal zone, which lies next to the brain ventricle
cavities of adult mammals, contains dividing cells. Various theories have
raged about the fate of these dividing cells, with the most prominent
being abortive cell death or the generation of glial cells. Thirty years
ago, by labeling dividing cells with radioactive precursors, Joseph Altman
first showed that dividing cells in the adult give rise to new neurons.
However, the alternative hypotheses of cell death and gliogenesis dominated,
and this idea was largely buried. Over the past decade, it has finally
become widely accepted that new neurons are generated in two regions of
the brains of adult mammals: the SVZ, which generates neurons destined
for the olfactory bulb, and the dentate gyrus of the hippocampus. This
required the advent of new methods for labeling dividing cells with permanent
markers, such as retroviruses that integrate into the genome of a dividing
cell and therefore tag all of its progeny, and microsurgery that allows
focal labeling of cells in the brain.
Neurogenesis
has been most convincingly demonstrated in adult rodents and birds, but
the phenomenon has been documented in all vertebrates examined spanning
the phylogenetic tree from lizards, fish, and cats to primates, including
humans. It is interesting that neurons in different species integrate
into different brain regions. These differences are likely due to the
availability of distinct migration routes in each species. In adult birds
and lizards, radial glia are maintained throughout adulthood and are used
by newly generated neurons to migrate and settle throughout the forebrain.
In contrast, in mammals, radial glia transform into astrocytes and this
substrate for migration is lost. Consequently, the dispersal of newly
generated neurons is more restricted in mammals. However, a recent report
suggested that neurons are found throughout the neocortex of adult primates.
The
SVZ is a thin layer of dividing cells that lies along the length of the
lateral wall of the lateral ventricles (Fig.
1). It is a region dedicated to the production and trafficking of
thousands of newly generated neurons each day. In adult mice, about 12,000
new cells reach the olfactory bulb daily, after migrating up to 8 mm from
their origin in the SVZ (more than half the length of the forebrain).
The extent of trafficking is dramatically revealed in whole-mount preparations
that expose the entire surface of the ventricle (Fig.
1A). A network of newly generated neurons extends from the back of
the ventricle to its anterior tip. New neurons within this network primarily
stream forward to converge on a restricted path that leads into the olfactory
bulb. The new neurons do not use radial glia or axonal guides, but migrate
in association with one another in long chains. This novel form of translocation
is called chain migration, and cells move rapidly through the brain by
crawling over each other. The chains of neurons travel through glial tunnels
formed by the processes of astrocytes. The role of these tunnels is unknown,
but they may confine the newly generated neurons to their restricted path,
or may provide them with, or protect them from, cues in the surrounding
brain tissue. Once the new neurons reach the core of the olfactory bulb,
they escape from the chains and migrate into two layers of the olfactory
bulb, where they differentiate into granule and periglomerular neurons,
two kinds of inhibitory interneurons. Many of the newly generated neurons
are culled 2 weeks to 1 month after their arrival. It is not yet known
whether those that survive replace other neurons in the olfactory bulb
that have died, or whether they are being integrated into novel circuits
as the need arises. The SVZ network of streaming neurons is present in
all vertebrates examined so far, including primates.
In
contrast to the extensive migration undertaken by olfactory bulb neurons,
new neurons in the hippocampus are generated locally in the subgranular
layer immediately adjacent to their final destination. They differentiate
into granule neurons that project to the CA3 region of the hippocampus.
The cell types in this region have not yet been characterized in detail.
The newly generated granule neurons are also transitory cells and are
eliminated about 2 weeks after their birth. Various factors have been
proposed to regulate adult neurogenesis including stress, hormones, and
physical exercise. The role of newly generated neurons and their potential
involvement in learning and memory formation remain fascinating and unanswered
questions.
The
adult brain contains neural stem cells that have been proposed to be important
for maintaining neurogenesis in vivo throughout adult life. These cells
can be cultured in vitro and exhibit the two fundamental properties of
stem cells. First, they demonstrate self-renewal, that is, they divide
to give rise to another cell that is identical with themselves. Second,
they are multipotential and can differentiate into neurons, astrocytes,
and oligodendrocytes. These cells can be propagated over years in culture
by different methods.
Which
cells in the adult brain are the in vivo stem cells? Stem cells can be
isolated from both the SVZ and the hippocampus, yet thus far only the
stem cells from the SVZ have been identified. When the SVZ is dissociated
to single cells, some divide in response to growth factors to generate
floating balls of cells called neurospheres. These neurospheres can differentiate
into both neurons and glia and can be passaged to generate new neurospheres,
thereby satisfying the criteria of multipotentiality and self-renewal.
The
SVZ contains four main cell types: newly generated neurons, astrocytes,
rapidly dividing precursors, and ependymal cells (Fig.
1C). The rapidly dividing immature precursors are closely associated
with the chains of newly generated neurons that migrate through the glial
tubes formed by the processes of SVZ astrocytes. They are scattered in
focal clusters along the network of chains. The multiciliated ependymal
cells line the ventricular cavity.
Surprisingly,
slowly dividing SVZ astrocytes act as stem cells in this region. After
specific labeling in vivo with a fluorescent marker, they generate fluorescent
neurospheres that can be passaged and give rise to both neurons and glia.
Furthermore, when they are specifically infected in vivo with a retroviral
tracer, they generate new neurons that migrate to the olfactory bulb via
the rapidly dividing type C cells. SVZ astrocytes are also capable of
tissue regeneration, a feature often exhibited by stem cells. We have
shown that when the SVZ is destroyed by administration of the antimitotic
drug Ara-C that kills dividing cells, only SVZ astrocytes and ependymal
cells remain. The SVZ network is rapidly regenerated within 10 days. This
regeneration occurs simultaneously over the entire wall of the ventricle
in small focal spots, suggesting that the stem cells are widely distributed.
By tracing which cells divide and following their offspring with time-lapse
photography, we showed that SVZ astrocytes, but not ependymal cells, begin
to divide rapidly after the antimitotic treatment is stopped and fully
regenerate the SVZ network of chains. When pulsed with a tracer, labeled
SVZ astrocytes give rise to the rapidly dividing type C cells that in
turn divide to generate the neurons that form the network of chains. This
pattern of genesis follows that of other stem cell systems, namely, a
relatively quiescent cell type divides to give rise to a rapidly dividing
amplifying cell population that in turn gives rise to more differentiated
cells. The multiciliated ependymal cells have also been postulated to
be stem cells in the adult brain; however, this finding has not been confirmed
by other groups.
The
finding that SVZ astrocytes are stem cells in the adult brain was most
unexpected, given that astrocytes and neurons are believed to arise from
different lineages during development. However, it raises the hopeful
possibility that astrocytes throughout the brain may have a latent capacity
to give rise to neurons. Multipotential precursors can be cultured even
from non-neurogenic regions of the brain, including the spinal cord, septum,
and striatum, as well as from the neurogenic hippocampal formation. It
is tempting to speculate that these as yet unidentified cells are similar
to SVZ astrocytes.
Recent
work hints that neural stem cells might not be restricted to the generation
of brain cells, but may be able to cross lineage boundaries after culturing
in vitro. Experiments suggest that neural stem cells may be able to generate
blood and muscle cells when transplanted into the adult and to contribute
to various lineages when grafted into a developing embryo. However, the
precise identity of the transplanted cells needs to be elucidated to substantiate
these potentially exciting results.
For
stem cells to persist in the adult brain, they must reside in a molecular
niche that permits their maintenance but also prevents unrestricted proliferation.
As such, strict control must be exerted over their division and differentiation.
When such control goes awry, unchecked proliferation could result in the
formation of brain tumors, including gliomas. The glial nature of the
stem cells in the adult brain is consistent with such a possibility. As
we begin to understand how neurogenesis is regulated in the adult brain,
it may be possible to manipulate SVZ cells to arrest unchecked proliferation
or to stimulate these cells to be used for brain repair. Purification
of adult stem cells and identification of factors that induce their differentiation
along distinct lineages are paths currently being examined in the hope
of using these cells for therapeutic purposes. Endogenous SVZ precursors
can be expanded in vivo by administering known growth factors and can
also remyelinate axons.
We
are left with the concept of a dynamic brain, one in which memories are
perhaps formed by the addition of new cells, and possibility of a brain
with a latent potential for self-repair. The adult brain is no longer
the static entity it once was thought to be.
top of page
|
