The New York Times - January 28th, 1992
A Brain Cell Surprise: Genes Don't Set Function
Job of neurons depends on location, a new study shows
By Natalie Angier
Young nerve cells can wander far greater distances across the growing brain
than anyone previously imagined, two scientists have found. And once the
neurons have settled into a particular neighborhood, they learn what they
are meant to do from signals that surround them, rather than from an innate
genetic program, the researchers said.
The new research strongly supports the idea that the cerebral cortex, the
most advanced component of the mammalian brain, is extremely supple and
responsive to its environment, instead of being fixed early on by a
blueprint in the genes, as some scientists had believed.
By the latest theory, a neuron that ends up in the visual region of the
brain becomes a visual neuron not because its genes so instructed it, but
because the cell is exposed to the appropriate signals from nearby cells and
from the optic nerve. If that same fledgling neuron had landed in the
auditory region of the brain, the scientists said, it would have matured
into a neuron fit for hearing.
In the last several years many biologists have come to favor the idea of
neuronal plasticity, but the latest report offers the most detailed and
powerful confirmation yet of the theory.
Dr. Constance L. Cepko and Dr. Christopher Walsh of Harvard
Medical School in Boston demonstrated that otherwise identical neurons
descended from the same mother nerve cell will, during brain development,
journey to wildly different regions of the cortex and assume very different
careers once settled.
The report appears in the current issue of the journal Science.
"We're normally thought of as nihilists, and so we didn't think anything
could surprise us," said Dr. Walsh, who is also a neurologist at
Massachusetts General Hospital, referring to his philosophy that many prior
conceptions turn out to be wrong.
"But even we were surprised by how far these guys would travel" from the
point where the dividing parent neuron had spawned them, he said.
The researchers do not yet know what determines when a young neuron will
cease migrating across the cortex and join a particular spot of the brain,
but they believe external cues like a tempting vacancy are probably far more
important than any genetic instructions the neurons carry with them.
"The mother cells do not impart specific information to their daughters
about what they become," Dr. Cepko said. "They do not say to one cell, you
go to the visual cortex and to another, you go the motor cortex."
The researchers tracked the patterns of cell migration by using a
technically daunting procedure, first injecting telltale molecular tags into
the developing brains of fetal rats, and later screening brain cells to
determine which neurons had landed where.
"I think this is one of the coolest pieces of work I've ever read," said Dr.
Carla J. Shatz, a professor neurobiology at the University of California at
Berkeley. "Doing this kind of experiment is like climbing Mount Everest.
It's not the sort of thing many people are willing or able to undertake.
They did a beautiful job."
"A Technical Tour de Force"
Dr. Dennis D. M. O'Leary, a neurobiologist at the Salk Institute in La
Jolla, California, who has also shown neural adaptability by transplanting
brain tissue in fetal rodents, said, "It was a technical tour de force, and
a very seminal study."
Neurobiologists have been arguing for decades over whether embryonic neurons
are blank slates or prefabricated units destined for a particular fate. Some
studies suggested that during development, related neurons move in lockstep
along set paths to reach one region of the cortex or another, as though
under genetic command, but until now scientists could not follow cell
migrations precisely enough to understand unequivocally what they were
Conversely, other experiments seemed to indicate that neurons are enormously
flexible. Researchers demonstrated that if they moved a tiny piece of tissue
from a fetal rat's visual cortex - the region of the brain that controls
sight - and grafted it onto the rat's somatosensory cortex, which rules over
body sensations, the transplant would take on all the features of its new
location, becoming a swatch of the somatosensory cortex.
In an effort to settle the debate, Dr. Walsh and Dr. Cepko devised an
exceedingly sophisticated technique. They knew that the brain develops as a
tube, with precursor cells within the tube dividing to give rise to
offspring neurons, which then migrate outward in all directions along
underlying connective fibers. "It's designed like spokes on a wheel," Dr.
But the precursor cells, or mother neurons, remain within the tube,
providing a steady pool of daughter cells to flesh out the brain. The Boston
researchers decided to try marking different mother cells with distinctive
tags, which would then be passed on to their children and could be followed
wherever the daughter cells should roam.
A Molecular Signature
The scientists began with 100 different strains of viruses, intended to
serve as easily identified fingerprints. Anesthetizing a pregnant rat at the
15th day of gestation, while the embryonic brains within her were at an
early stage of development, the researchers injected the viruses deep into
each pup's neural tube, where they infected some of the precursor neurons.
Because so many different types of viruses had been introduced into the
growing brains, the odds greatly favored each neuron's being injected with a
unique viral strain that then became its unmistakable molecular signature.
The operation had no apparent effect on the fetal rats, and the animals
developed and were born normally.
A week or so later, the scientists removed the rodents' brains and then
isolated cortical tissue. Using a technique called polymerase chain reaction
to discriminate between viral markers in individual neurons, the biologists
saw astonishing patterns of cell distribution. Daughter cells from many
different mother neurons were sitting side by side in the same cortical
neighborhoods. Equally impressive, siblings of the same mother cell had
meandered to all points on the globe of the cortex, taking up residence in
many different functional regions.
Judging by the relatively casual and widespread distribution of neurons, the
researchers concluded that the cells were not likely to be obeying strict
maternal advice inscribed into their DNA. Instead, the scientists propose
that when the neurons arrive at their final location, they are hailed by
other cell components - long feathery fibers called axons - which hook up to
the neurons and instruct them on their assignment.
"The neurons are greeted by axons coming from other parts of the brain, and
the axons tell them precisely what to do," Dr. Walsh said. "One set of axons
carries visual information, another set carries information about
In the visual cortex, for example, axons may extend backward into the brain
from the optic nerve, relaying visual pulses inward that in turn influence
the local neurons.
2 Competing Theories
"Development can either proceed by the British plan, where one's fate is
determined by one's ancestors, or by the American plan, where one's fate is
determined by one's neighborhood," said Dr. Walsh, using an analogy first
suggested by the British geneticist Sydney Brenner. "The cortex turns out be
on the American plan in the extreme."
The researchers admit that their engenders more questions than it resolves,
and that in a sense it is a negative result: the experiment says what the
neurons of the cortex are not. They are not prefixed units outfitted with a
complete code of how to behave, and they almost surely do not provide the
organizing principle of the mammalian brain. Other researchers are
struggling to find out what is then the master plan behind the body's gray
matter. If the cortical neurons are clued into their assignment by axons
that are already in place, what signals told the axons where to go and what
to do? What tells an axon that it is meant to help the brain see, or smell
Many researchers suspect that brain development proceeds in a great and
exquisitely complex feedback loop, partly determined by the oldest regions
of the brain, like the hindbrain that controls breathing, and partly by
signals coursing in from distant regions of the flowering nervous system.
They also believe that the thalamus, a region deep in the midbrain where
many sensory signals converge, develops before the cerebral cortex and then
helps shape the cortical layers that grow up around it. But the basis of the
architecture of the higher brain remains a subject of intense investigation.
Whatever the lingering mysteries, the new results do suggest why the
mammalian brain was able to evolve so quickly and to such an impressive
degree of complexity. Because neurons of the cortex seem to be remarkably
malleable, able to adjust their performance depending on the input they
receive, they can swiftly respond to changes in the rest of the animal's
body without having to bother going through cumbersome genetic alterations
For example, the paw of a primitive rodent can evolve into the wing of a
bat, but the neurons that interpret sensory signals from the limb need not
be reprogrammed to manage the change in body plan. That degree of elasticity
in the brain may have had all sorts of spinoffs, including, eventually, the
evolution of consciousness.