Neuroscience got its start more than 100 years ago when Santiago Ramon y Cajal and Camillo Golgi began depicting what brain cells—neurons—actually looked like. The two shared the 1906 Nobel Prize for their efforts. Succeeding generations have taken advantage of technological advances to add depth and detail to our picture of neurons. This week, the team of investigators at the Allen Institute for Brain Science in Seattle released the first installment of human cells in its cell types database, thereby zooming in a little further to reveal a new level of intricacy and complexity in the tiny units that underlie your every thought, movement, and spoken word.
Neurons are the elementary units of the brain. Each consists of a cell body (which contains the nuclei), axons for sending signals, and dendrites for receiving them. Altogether, there are about 84 billion of neurons, which link together to form roughly 100 trillion connections. Those connections—electrical pulses traveling across synapses from one cell to the next to the next—make up the circuits that result in brain function.
Eighty-four billion is an enormous number, but all those neurons can be sorted into a much smaller number of categories, according to Ed Lein, an Allen Institute investigator. “We can reduce the complexity to some much smaller number of types that are simply repeated,” says Lein. “It’s looking like there are on the order of 80 to 100 types of cells based on the genes within just one part of the cortex.” The brain is, in fact, structured in layers like an onion and each layer is defined by the cell types that populate it, which have been given descriptive names like pyramidal, chandelier and basket cells.
Four years in the making, this new database is the latest in a series of big data brain resources made publicly available by the Allen Institute, a research organization devoted not just to making discoveries but to making resources available to other scientists so that they can make discoveries. Previous releases have included atlases of mouse and human brains, and a connectivity atlas of the mouse brain. All of it is available at brain-map.org.
Representing cells from just one part of the human brain, the neocortex, the new database reveals not just the structure, but also the electrical properties of different cell types. Neurons fire with different patterns according to their cell type. “When you put in a pulse of current,” explains Lein, “some of them fire a burst of action potentials. Some of them fire only a few. Some of them decrease over time with the same stimulus.”
Observing such electrical activity, which is at the heart of neural communication, is only possible in living brain tissue. To pull that off, Allen Institute investigators worked with Seattle-area neurosurgeons. Surgeries for epilepsy or tumor removal often require the removal of small amounts of healthy brain tissue, which is usually treated as medical waste. With patients’ approval, the surgeons instead handed over the live brain tissue to Allen Institute scientists in much the same way that organs are harvested for transplant. The scientists were pleased to discover that they could keep the tissue alive for up to two days (comparable mouse tissue only lives for a few hours) while they performed the experiments that led to the creation of the database.
The overall goal of the Allen Institute program is to truly understand how the human brain works. Most brain research in the United States has focused either on mouse brains, which serve as a model for human brains, or on the specifics of brain disease and disorder. By contrast, the cell types database gives us a picture of what individual healthy brain cells look like. “This can serve as a baseline for understanding what goes wrong,” says Lein. “Researchers can be looking at Alzheimer’s or autism at this very fine level and they might discover that certain types of these cells are missing or their properties are changed compared to normal cells.” He also suggests that the growing field of stem cell research, in which scientists try to create new healthy cells in the lab, might use the database to see if their stem cells look like the real thing.
It’s especially significant to have such a reference for human cells as opposed to mouse neurons. Historically, the development of drugs to battle diseases like Alzheimer’s has begun in mouse brains, because they are much easier to study. “The hope is that we can use model organisms like the mouse to replicate a human disease,” says Lein, “but in fact there are many, many differences. That whole endeavor is failing most of the time.” Indeed, the extent of the differences in size and complexity between mice and humans was one of the surprises for Lein and his colleagues in this project.
“If we really want to understand the details of what gives us our particular functions and cognitive abilities, or If we really want to understand human disease,” says Lein, “there’s no substitute but to actually understand the details of the real system itself.”