September 18, 2024

The fine art of brain development

photo by Steve McAlister/The Image Bank

From rocks that took nature eons to build, some of the world’s most notable sculptures—Michelangelo’s “David,” Rodin’s “The Thinker,” Mount Rushmore—have emerged, their basic shapes roughed out with chisel and mallet, and their fine detail and subtle textures carefully carved and refined with more delicate tools and a lighter hand.

Perhaps the grandest sculpture of all, the human brain, is shaped by a combination of these basic processes—an early “building up” of the brain’s bulk, followed by a “roughing out” of the major brain regions, and later, a meticulous refinement of the detail that imparts its unique functions.

Sculpting a brain—or indeed, an entire nervous system—takes a lot more than a hammer and chisel, requiring at least one half of the entire human genome. The end product, a grayish-pink, 3-pound gelatinous mass, may be the most complex structure in the known universe— containing at least 100 billion nerve cells (neurons) and one trillion support cells (glia), which can make at least one quadrillion connections between them. The perhaps hundreds of different chemicals (neurotransmitters) that relay information between these neurons further increase the complexity.

It’s no wonder that many questions about how the brain develops—both normally and abnormally—remain unanswered. How is the incredible diversity of brain cells and connections generated from our finite genome? How do the maturing neurons know where to go and which neighbors to “hook up” with? And how do events during development affect the brain’s ability later in life to acquire and store new information through rewiring (plasticity)?

Using a range of animal models from fly to mouse, Vanderbilt researchers across a number of disciplines are probing the many mysteries of brain development and are providing insights into how it may go awry in neurological disease.

“Normal brain development is a staggeringly beautiful and wondrous thing,” says Kendal Broadie, Ph.D., Stevenson Professor of Neurobiology and professor of Biological Sciences and Pharmacology at Vanderbilt University. “It gives rise to this structure that’s beyond our comprehension—a structure that allows you to see, think, run and sing.”

This remarkable structure begins to emerge from a single layer of neural stem cells lining a tube in the early embryo at around the third to fourth week of gestation in humans.

Kendal Broadie, Ph.D., with an image of an adult fruit fly (Drosophila) brain.

Photo by Anne Rayner

Thus begins the “build up” phase of brain sculpting. The cells lining the wall of this neural tube begin dividing rapidly—by some estimates, at the rate of 50,000 cells per second— and the walls progressively thicken. Soon, decisions are made as to whether these primitive cells go on to become neurons, the cells that process and transmit information, or glial cells, the supportive “partner” cells that provide nutrients, oxygen and other necessities to neurons.

As primitive nerve cells become neurons, they develop extensions from their cell bodies—many short projections called dendrites that receive incoming signals, and a single, long axon that transmits those signals to the next neuron.

Glial cells, though they have many similar features to neurons, do not develop these specialized appendages. Instead, some of them go on to form the protective sheath called myelin that wraps and insulates the axons of many neurons and enhances the speed with which nerve impulses can travel from cell to cell.

Glowing genes

Bruce Appel, Ph.D., associate professor of Biological Sciences at Vanderbilt, is studying the development and specification of oligodendrocytes, the glial cells that form myelin in the central nervous system (CNS), which includes the brain and spinal cord.

In humans, myelination begins shortly before birth and continues into adolescence. In Appel’s research subject, the zebrafish, myelination starts around the third day after fertilization.

The zebrafish is a great model system for studying nervous system development, Appel says, because the embryo is transparent and develops entirely outside the mother. And it develops in two days. By comparison, the mouse embryo takes about 10 times longer to mature.

By engineering certain zebrafish genes to glow green, Appel can easily view specific sets of neural progenitor cells—immature nerve cells—and in particular, the cells that go on to produce oligodendrocytes.

“We’ve found that oligodendrocytes, which were always considered to be really boring cells, actually turn out to be incredibly dynamic,” he says. The cells send out fine processes, called filopodia, and appear to use these membranous “arms” to explore their surroundings, sampling the environment.

“They zip around, all over, until they finally arrive at their target axons. They continue to explore their area and move around and settle into a fairly regular distribution. That’s really fascinating to me, and we don’t understand it at all.”

They also appear to be very flexible, he notes. “We’re finding that the oligodendrocyte is very plastic … We’re beginning to get the sense that there are different kinds of oligodendrocytes. There are certain mutations that result in the absence of one kind of oligodendrocyte, but these may be rapidly replaced by another kind.”

After moving his lab to the University of Colorado Denver School of Medicine this summer, Appel plans to continue his search for genes that guide the developmental stages of oligodendrocyte progenitor cells (OPCs)—the immature cells that develop into mature oligodendrocytes—and the genes that determine their unusual behaviors.

To find them, Appel’s team uses a traditional genetics approach—causing random mutations in zebrafish embryos and screening these “mutants” to find ones with disrupted oligodendrocyte development.

He hopes that these mutants will point to genes that influence their specification (whether they go on to become an oligodendrocyte or another type of neuron), how fast they divide, and how they recognize and insulate their “target” axon and not other axons.

“We’re picking up mutations that affect all of those things,” he says. One mutant, called pescadillo, or “little fish,” produces an excess of OPCs, perhaps due to a genetic defect that causes their multipotent precursor cells (even more primitive cells than the OPCs that can produce oligodendrocytes or motor neurons) to continually divide. Another mutant, which Appel has aptly dubbed Peter Pan, has OPCs that “never grow up” —they don’t mature into myelinating cells.

His lab will try to identify the genes affected in these mutants—not an easy task, to be sure. But he says, “It’s going to be a lot of fun to work through.”

Using another approach, a screen for chemicals that disrupt oligodendrocyte development, Appel has found compounds that cause an excess formation of oligodendrocyte lineage cells.

“This was far beyond my wildest dreams because I thought we’d find things that would block oligodendrocyte development,” he says. “There are far more ways to block something than promote it.”

A chemical that promotes the development of myelin-forming oligodendrocytes may point the way toward therapies for remyelination—which could be beneficial for diseases like multiple sclerosis in which myelin abnormally degrades and results in nervous system dysfunction. Appel is hoping to pursue this lead with a biotech company to determine whether this compound or others like it might be feasible therapeutic targets.

“We need to determine whether this (compound) can direct differentiation of multipotent stem cells into the oligodendrocyte pathway,” he says. If so, Appel predicts this compound might become a “super-wonder-drug.”

Death signal

Motor neuron expressing different colored fluorescent proteins in a C. elegans larva.

Image courtesy of David Miller, Ph.D., Vanderbilt Department of Cell & Developmental Biology

One of the more curious aspects of nervous system sculpting is the natural overbuilding that occurs. More neurons are produced than we will ever need or use—and thus are eventually “chipped away.” About half of all neurons born will die through a pre-programmed “suicide” mechanism called apoptosis, says Bruce Carter, Ph.D., professor of Biochemistry at Vanderbilt.

“In some places, you lose all the neurons,” he explains. “In other places you lose 10 percent. It varies, but about half of the neurons generated die—it’s a normal pruning process.”

The delicate balance between life and death of brain cells is centered on a family of molecules called neurotrophins. This family includes nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF) and neurotrophins 3 and 4 (NT3 and NT4), and has been an intense area of focus for Carter and colleagues.

The neurotrophins can bind to two different classes of receptors embedded within the neuronal membrane: the Trk family of receptors, which usually promotes survival, and the p75 receptor, which can promote either survival or cell death.

The dual role of p75 had baffled researchers. “It was already known for 50 years that NGF promotes survival,” Carter says. “So the idea that somehow the receptor for NGF could cause cell death didn’t really make any sense.

“However, what we’ve learned is that neurons that get the ‘right’ neurotrophin first will survive through a combined Trk-p75 signal, and they start producing a different neurotrophin, which acts through p75 alone to cause death of their neighbors. Thus, there is a beautifully regulated competition set up so that the proper connections are efficiently established.”

How these factors produced such opposing signals was still a mystery. Several years ago, while investigating how these conflicting signals are generated by p75, Carter and colleagues discovered a protein, called NRIF (neurotrophin receptor interacting factor), that binds to part of the receptor and appears to be required for p75-induced cell death. NRIF resembled well-known transcription factors that alter gene expression within the nucleus. The Vanderbilt researchers and others had also determined that NRIF entry into the nucleus induced apoptotic cell death.

“It was kind of puzzling that we found a putative transcription factor 'out there' at the cell surface of the neuron (instead of in the nucleus),” Carter said.

Cross-section through the midbrain of a 4-day-old zebrafish larva. Axons coming from both sides of the forebrain have been labeled with green fluorescent protein, while only those coming from the left have been tagged, in red, with an antibody. Cell nuclei have been stained blue.

Epifluorescence microscopy image by Robert Taylor, graduate student in the Vanderbilt Department of Biological Sciences, courtesy of Josh Gamse, Ph.D.

Carter and research instructor Rajappa Kenchappa, Ph.D., have since determined that an enzyme cleaves p75, liberating NRIF from the cell surface and allowing it to travel to the nucleus to affect its “pro-death” signal.

This mechanism may explain some of the naturally occurring neuron death during development. Mice lacking p75 have an overabundance of neurons because the cells cannot die, Carter says.

Knowing these 'death signals' could also allow researchers to develop therapies that prevent the undesirable cell death that occurs in neurodegenerative diseases like Alzheimer's as well as after spinal cord injury and stroke.

Structure vs. function

It’s not enough just to have the appropriate complement of neurons; they must also connect with other neurons. The formation of these connections, or synapses, sets up communication links between neurons. The ability to alter the strength and number of these connections—a property known as “plasticity”—throughout the entire lifespan of an organism drives behavioral changes and underlies learning.

“Synapse formation … is the end of building structure and the beginning of building function,” Broadie says.

In humans, synapse formation begins during late embryonic development (around the beginning of the third trimester) after the bulk of brain “building” is complete. And, unlike the earlier steps of brain development—differentiation, migration and axon guidance to their targets—synapse formation and later plasticity are dependent on neural activity, particularly on sensory activity.

The complex process seems to require an almost inconceivable number of “coincidences.”

“You have to have a signaling cell and a receptive cell in register at the same time, the same place, and also of the same ‘flavors,’” Broadie notes. These “flavors” are the neurotransmitter systems expressed by the cells. A neuron that produces dopamine, for example, needs to connect with cells that possess a receptor for the neurochemical.

Broadie uses the fruit fly Drosophila to dissect all aspects of the life cycle of the synapse: how it’s made, how it works, and how it changes throughout the organism’s lifespan. One way he does so is in the context of a disease called fragile X syndrome, in which synaptic development and/or function goes awry.

Fragile X disease, the most common inherited form of mental retardation, causes a structural overgrowth of dendrites and axons during development, as well as functional abnormalities in synaptic plasticity later in life.

“There’s no question in my mind that fragile X is a disease of development,” says Broadie. “But there is a real split in the field whether it is primarily a disease of development, a disease of plasticity, or both.”

The answer is vital to developing intervention strategies, Broadie says. “If you want to fix the problem, you absolutely have to know where the problem is—or when the problem is.”

Fragile X in humans is caused by altered expression of a gene called FMR1 (fragile X mental retardation 1) resulting in the loss of its protein product, FMRP. Broadie and colleagues have developed a Drosophila model of the disease and have used the fly model to examine the developmental roles of FMRP.

They’ve found that FRMP is most highly expressed during a brief window of time during late brain development, and that the protein’s expression is increased by sensory input. Their work shows that FMRP plays a critical role in limiting axon and dendrite growth, in particular the activity-dependent “pruning” of neuronal branching, which is vividly illustrated in the overgrowth of neuronal processes and abnormal synapse formation in flies lacking the protein.

“If you compare a fragile X mutant brain to a normal brain, there are fairly severe problems in things like nerve cell structure and synapse formation,” Broadie says. “But—and here’s the crux of the problem—most of those defects go away.” In mouse models of fragile X, he says, after the first month following birth, their brains look fairly normal.

Even though the structural abnormalities appear to go away, the functional problems associated with fragile X persist. Even though the synapses look normal, he notes, it is unclear whether they function properly.

The dynamic brain

So while the link between altered brain development and the later problems associated with fragile X is being resolved, Broadie and others are already finding factors that might be exploited to improve the symptoms of fragile X.

Because FMRP is a protein that regulates the expression of other proteins, Broadie and his colleagues are looking for genes and proteins that might be affected by FMRP.

Only a handful—about eight—have been proven so far. One protein found by the Broadie team regulates the internal scaffolding, or cytoskeleton, of neurons, which, he says, “makes perfect sense in that the main defect you see is the change in the structure of nerve cells (and) the cytoskeleton determines the structure.”

Another prospect is the involvement of a neurotransmitter receptor called the metabotropic glutamate receptor (mGluR). The receptor—which is activated by glutamate, the main excitatory neurotransmitter in the central nervous system— is important for neuronal plasticity throughout life, and FMRP acts downstream of mGluR activity. Broadie is using the fly model to study the interactions between mGluR and FMRP by manipulating the expression of their corresponding genes in combination.

Studies in mice suggest that excessive signaling through mGluR5 may be responsible for the neurological and psychiatric consequences of fragile X syndrome. Even though FMRP is missing in humans with fragile X, Broadie notes, it may be possible to find ways to manipulate signaling through the mGluR and circumvent some of the later problems of fragile X.

Researchers at Vanderbilt, for example, have identified more than 400 “negative allosteric modulators,” compounds that selectively “turn down” the activation of mGluR when glutamate binds to it. With support from Seaside Therapeutics of Cambridge, Mass., they are developing compounds with drug-like properties for further study.

“It's a really innovative idea,” says Jeffrey Conn, Ph.D., director of the Vanderbilt Program in Drug Discovery, who is leading the project in collaboration with Craig Lindsley, Ph.D., Alice Rodriguez, Ph.D., and David Weaver, Ph.D. “If it works, it could be transformative … It could totally change the way people view developmental disorders.”

Unlike sculpture released from stone by the human hand, the brain never achieves a final form. The biological “thinker” is constantly in motion. Throughout life, synapses rearrange and become stronger or weaker, neurons die and (in a few cases) new neurons are born.

Though invisible to the naked eye, this dynamic, continual process of brain sculpting is what gives brain researchers hope that we can find ways to not only treat or prevent diseases like fragile X, but also to just improve the function of the normal brain.

“The brain is not static, it constantly changes itself in response to its environment,” says Broadie. “Your heart doesn’t do that. Your liver doesn’t do that. That’s the property that makes it so special.

“That’s what makes the brain, the brain.”