Focus - January 23rd, 1992
Cell Milieu Directs Cortex Development
Drs. Cepko and Walsh Trace Progeny of Developing Brain Neurons
By Kenneth B. Chiacchia
The development of growing cells in the cortex of the brain is influenced
more by interaction with the cells' environment than by intrinsic genetic
cues, conclude Harvard Medical School researchers. The scientists, Drs.
Constance Cepko and Christopher Walsh, have been studying for several years
how cells learn what to become during development. They have shown that the
descendants of single progenitor cells in the developing brain can end up in
distant structures of the cerebral cortex, performing widely different
functions*. In deciding where to migrate and how to differentiate, the
developing cortical nerve cells must be responding to external signals, they
"One thing is certain," notes Dr. Cepko, HMS associate professor genetics,
"a mother cell does not tell all of its daughters to go to a particular
functional domain. Presumably, while newly generated cells are migrating, or
after they arrive at their destination, they are getting information about
how to differentiate by interacting with neighboring cells, or other
A central question in developmental biology, says Dr. Cepko, has been
whether neighboring cells in the brain's structures are descended from a
common progenitor, or "mother" cell, or whether they migrate and develop in
response to interactions with their neighbors. "It is a question of
environment versus lineage," she says. Adds Dr. Walsh, HMS research fellow
in genetics, "It is a general problem in developmental biology but also
might have eventual relevance to such disorders as brain malformations and
epilepsy." Dr. Walsh is also a neurologist at Mass. General Hospital.
Nature vs. Nurture
The question of "nature versus nurture" - whether the mind's psychological
structure is more a product of inborn qualities or of life experiences - has
fascinated thinkers since ancient times. But as scientists have begun to
uncover how the single cell of a zygote leads to the many complex structures
in an organism, a similar controversy has arisen over how the physical
structure of the brain is determined. Do the neurons of the developing brain
have inborn genetic instructions that dictate where they will go and what
functions they will carry out? Or do signals from other cells orient them,
causing them to grow in a way that gives rise to the intricate networks of
connections that comprise the brain?
Previous studies using more primitive parts of the brain indicated that
descendants of neurons remained clustered together to form brain structures,
suggesting that internal genetic cues are more important for the development
of those parts of the brain. Work with lower organisms, such as the fruitfly
Drosophila melanogaster, had indicated that relatively tight genetic control
restricted developing neurons. "Segments of the Drosophila embryo, including
the compartments of the brain, correspond to cell lineage," says Dr. Walsh,
with single progenitor cells giving rise to clusters of nerve cells in the
same location and with the same structure and function. The hindbrain of
vertebrates - the most primitive structure of the brain and the part
responsible for basic functions like eating, breathing, and sleeping -
"shows a similar compartmentalization, with lineage compartments
corresponding to the structural segments" that give rise to such structures
as the cerebellum and the medulla oblongata.
Uniqueness of Cortex
The situation has not been as clear in the cerebral cortex, where different
experiments yielded contradictory results. The largest part of the human
brain, the cortex is the location of higher brain activity, the residence of
intelligence and memory, among other functions. "It is what we think of as
the most uniquely human part of the brain," states Dr. Cepko. For this
reason, and because fetal development of body structures often gives
scientists clues as to how they developed evolutionarily, developmental
biologists have been keenly interested in how the cells of the cortex know
how and where to specialize into their final functional forms.
The only way to address this question - whether descendants of progenitor
cells give rise to discrete functional units or whether they intermingle
more widely across the cortex - was to develop a new technology, says Dr.
Walsh. By infecting progenitor cells in the developing rat brain with a
number of slightly varied retroviruses, he explains, the scientists could
label them. Descendants of the labeled cells could be identified by the
differences in the viral DNA they had inherited from their ancestors.
The gene they engineered into the retroviruses was that for beta-galactosidase,
an enzyme which turns the labeled cells blue. By using a mixture of 100
retroviruses containing different DNA "markers" near the galactosidase gene,
the researchers were able to introduce a molecular fingerprint that could be
identified later on. "We could just cut out a chunk of tissue and amplify
DNA from a single labeled cell using PCR (the polymerase chain reaction
process by which trace amounts of DNA are multiplied and analyzed)," states
The researchers were surprised to observe where the descendants of the
labeled neurons wound up in the developing cortex: while offspring of a
single cell were often clustered together, a large proportion could be found
dispersed from one end of the cortex to the other, according to Dr. Walsh.
While clusters of cells descended from a single progenitor were all found to
reside in a single functional compartment in the hindbrain, such descendants
in the cerebral cortex usually end up in different functional domains.
"These descendants have a lot of different functions," he says.
With the apparent lack of internal instructions for how the neurons will
develop, developmental direction "has to arise from intracellular
interactions," states Dr. Walsh. "A piece of the cortex doesn't know whether
to be part of the auditory or visual structure unless it is told what to
do." Dr. Cepko expands, "This means that the instructions they are receiving
are coming from other cells. The cells that give the signals probably arose
much earlier in the evolution of the brain and most likely some are
non-cortical cells. The progenitors that make up the cortex are a population
of cells responsive to such signals." The next step, says Dr. Walsh, will be
to identify the cells that send out these signals, and how their message is
carried to the developing cortical neurons.
Developmental and evolutionary biologists have derived many important
lessons from each others' disciplines. The finding of the relative
importance of environmental signals in the development of cortical neurons
provides tantalizing hints as to how the cortex evolved, says Dr. Cepko.
Both the shape and size of the cortex evolved very recently, she adds, and
the question of how the cortex - both that of the rats examined in these
studies and that of humans - arose and from whence came higher brain
function may be one and the same. These results, she believes, "could imply
something about its evolution. We found that the progenitor cells are not
programmed to provide a certain number of cells to go to a certain place.
The plan is not hard-wired into the genome.
The absence of a genetically-determined plan "may have provided flexibility
in terms of development," adds Dr. Walsh. The cortex, he says, may have
constituted a neurological structure that could be "adapted to the
particular needs of the organism. " The researchers postulate that the
plasticity of the cortex may have aided evolution somehow, allowing
relatively fast physical development to meet the challenges faced by mammals
as they evolved. This, in turn, may prove an important piece of the
evolutionary puzzle of how and why human intelligence developed.
*"Widespread Dispersion of Neuronal Clones Across Functional Regions in the
Cerebral Cortex." Christopher Walsh and Constance L. Cepko. Science, January