Gene mutation in children with microcephaly reveals an essential ingredient for brain development

This electron microscope image shows two multivesicular bodies in the dendrites of neighboring neurons in the cerebellum of a normal mouse. Each contains vesicles bearing sonic hedgehog. (Michael Coulter/Boston Children’s Hospital)

In 2012, researchers in the Boston Children’s Hospital lab of Christopher Walsh, MD, PhD, reported a study of three unrelated families that had children with microcephaly. All had smaller-than-normal brains — both the cerebrum and the cerebellum were reduced in size— and all had mutations that knocked out the function of a gene called CHMP1A.

It was clear that CHMP1A is needed for the brain to grow to its proper dimensions. But the study stopped there.

“Then I came along, and my goal was to figure out what this gene is doing in brain during development, and why, when you lose it, you have a small brain,” says Michael Coulter, MD, PhD, who joined the Walsh lab as a student in 2012.

The new study, published recently in Cell Reports, reveals a previously unrecognized mechanism for brain development: the release of small extracellular vesicles, carrying an essential growth factor to different locations in the brain.

A dearth of sonic hedgehog

Coulter and colleagues started by knocking out the CHMP1A gene in developing mice. Like the human patients, the animals were born with microcephaly, particularly affecting the cerebellum. The team then looked at the brain tissue to see what was happening.

Going in, they knew two important things. First, they knew that a growth factor called sonic hedgehog is important in the growth of the cerebellum. So they looked for it specifically.

“We found only about half as much sonic hedgehog protein as normal in the cerebrospinal fluid of developing mice that lacked CHMP1A,” says Coulter.

They also knew that CHMP1A is part of a group of proteins involved in forming extracellular vesicles, small sacs cells use to package up cargo and release it to the outside. “We wanted to see if there was problem secreting sonic hedgehog in these small extracellular vesicles if CHMP1A was mutated.”

A missing ‘escort’

Indeed, when CHMP1A was knocked out, neurons released fewer vesicles bearing sonic hedgehog. Inside the neurons, there were fewer intraluminal vesicles, which gather in organelles called multivesicular bodies before being released as extracellular vesicles.

Multivesicular bodies containing vesicles are more sparse in mice lacking CHMP1A. The Chmp1a protein, shown in red at left, appears to be important in formation or budding of vesicles. (Michael Coulter) Below, multivesicular bodies in a cortical neuron (Wei Lee/Boston Children’s Hospital)

Extracellular vesicles, commonly known as exosomes, have emerged in the past decade as important vehicles for cell-to-cell communication. They come in many flavors, and multiple companies are now exploiting them as diagnostic aids or therapeutic delivery vehicles.

And without them, sonic hedgehog wouldn’t get far in the developing brain. Chemically, it’s hydrophobic, so it doesn’t diffuse well in a watery environment. “It can’t be released directly,” says Coulter. “It needs a carrier.”

But riding on a vesicle, sonic hedgehog is buoyed along and can dock at distant ports, spurring brain growth.

“We think this is the first example of extracellular vesicles’ role in mammalian brain development,” says Coulter.

A vesicular model of brain disease?

The researchers think a similar vesicle-based mechanism may be active in the full-grown brain, disseminating other factors that maintain and regulate neurons. Testing cerebrospinal fluid from an adult human patient, they found that sonic hedgehog and its carrier vesicle are still present. Electron microscopy data, provided by coauthor Wei-Chung Allen Lee, PhD, of Boston Children’s F.M. Kirby Neurobiology Center, showed multivesicular bodies throughout adult neurons:

(Wei Lee)

Coulter and colleagues even speculate that vesicles could be hijacked in some diseases. For example, there is evidence that Tau protein, part of the Alzheimer’s disease process, may be carried in extracellular vesicles. Now, with a mouse model, it may be possible to test whether vesicles are helping to spread Tau in the brain. If so, disrupting the release of Tau-bearing vesicles could be a potential therapeutic strategy.

“Our work showing how vesicle release happens in the mammalian brain could help provide the foundation for such an approach,” says Coulter.


Coulter was co-first author on the study with Cristina Dorobantu of Universite ́ de Strasbourg, France, and Gerrald Lodewijk of the University of Amsterdam, Netherlands. Frank M.J.Jacobs (University of Amsterdam), Raphael Gaudin (Universite ́ de Strasbourg) and Christopher Walsh (Boston Children’s Hospital) were co-senior authors. Walsh is also a principal faculty member of the Harvard Stem Cell Institute.

Funders included the National Institutes of Health (R01 NS088566, U19AI109740, GM075252), the NIH National Institute of Neurological Disorders and Stroke (R01 NS35129, F30 MH102909), the Howard Hughes Medical Institute, IdEx Universite ́ de Strasbourg via the Agence Nationale de la Recherche and the Nancy Lurie Marks Family Foundation.

See the paper for a full list of authors and funders.