Scientists at the NIH are mapping the activity of thousands of individual neurons inside the brain of a zebrafish as the animal hunts for food.
In a small, windowless room that houses two powerful electron microscopes, a scientist is searching for the perfect fish brain.
As the massive machines hum nearby, two gigantic fish eyes loom large, taking up most of a computer screen. The magnified perspective is misleading. The zebrafish is a larva, a newborn, just one week old, and barely six millimeters long. On the screen, it looks grumpy, like it’s frowning.
Chris Harris, a postdoctoral researcher at the lab, is scrolling through the image. As he zooms in, the eyes become even larger and then disappear altogether, replaced by a glimpse of what lies within and behind them in its brain: a jungle of axons and dendrites and cell bodies — all the stuff that makes up individual neurons.
He traces the outer edge of one of the cells with a gloved finger. “This layer is the nuclear membrane,” he says. “And just outside of that is the cell body membrane itself.” He points out the mitochondria, the individual axons, which send nerve impulses from one neuron to the next; the branching dendrites, which receive signals; and thick black dots that represent synaptic vesicles — pouches that hold neurotransmitters, or brain chemicals.
“What I’m looking at is the quality of the tissue,” Harris says. “Are the cell walls distinct? Can I see synaptic vesicles? Can I see myelin?”
Harris and his team, led by Kevin Briggman at the National Institute of Neurological Disorders and Stroke in Bethesda, Maryland, are recording the activity of tens of thousands of individual neurons and preparing to create a sort of wiring diagram, a map that shows each cellular connection at the synapse. Such a map is known as a connectome. The goal is to ultimately create a comprehensive atlas of the fish brain — in unprecedented detail. This requires a full reconstruction of every synapse and every vesicle of every individual neuron, along with a recording of the brain’s activity as the fish hunts for food.
In February 2013, news hummed with the announcement that the Obama administration was planning a decade-long effort to map the human brain. Originally priced at $100 million, the cost estimate was recently revised upward to $4.5 billion by the National Institutes of Health. The ambitious effort has been compared to the Human Genome Project. And also to walking on the moon.
But in labs across America dizzyingly complex efforts were already underway to map the brains of animals. Briggman’s is one such lab.
In 1986, a map of a brain of the first living organism was completed – the C. elegans, a tiny, millimeter long roundworm that feeds on bacteria that thrives in rotting fruit, flowers and animals. The effort took more than a decade and required studying 20,000 electron microscope cross sections of the worm. The C. elegans brain contains about 300 neurons.
The fish brain, on the other hand, has 100,000 neurons. For perspective, the mouse brain is made up of roughly 100 million neurons, and the human brain likely ranges from 86 billion to 100 billion neurons.
“We’re trying to take a similar approach, but to scale it up to vertebrate organisms,” says Briggman, one of 22 researchers to be named a Pew Scholar in the Biomedical Sciences last week. He adds that they would be the first team to reconstruct a full vertebrate brain at this scale. An organism, in other words, with a spinal cord and the same basic brain architecture, Briggman says, as that of a human.
It’s hard to believe that a fish brain could be anything like the brains of thinking people who ride bicycles and balance checkbooks and invent driverless cars. But compared to a worm or a snail, they share many of our brain regions. Fish also have a brain stem that controls movement, a cerebellum that modulates those movements, a big olfactory bulb and a forebrain. “Some argue that they even have a hippocampus, which has a role in memory formation and fear,” Harris says. “Everything is there.”
These brain regions are sized differently in fish. The fish brain is dominated by the visual areas, because vision is so important for hunting food. The olfactory bulb is well developed too, allowing fish to smell chemicals dissolved in water. What fish lack is a developed cerebral cortex, where higher-order thinking takes place in primates and humans.
Yet while a fish darting after a dot of light may seem to be little more than instinct, the connections underlying the behavior are fairly complex, requiring the use of dozens of brain regions. Briggman describes it this way:
“It sees a goal, and adapts its behavior. It turns a certain direction, rotates its eyes so its eyes converge on a target, and eventually has to make a decision of whether it’s going to eat this thing or not. There’s some degree of decision-making going on in this behavior.”
Also, the larva are transparent, which means scientists can see clearly through the young fish skin and into the brain tissue without having to dissect it.
Of particular importance to Briggman’s team is how these neurons are organized into circuits.
“A circuit would be just like an electrical circuit,” Briggmann says. “It’s a cluster of neurons that are connected to each other, and the circuits in the brain, in our brains or in the brain of fish, perform certain computations.”
Harris, 30, is tall with a Swedish accent and a close-cut beard, and he chooses his words thoughtfully when talking about his research. On a recent afternoon in the team’s “wet lab,” he is preparing another week-old zebrafish fish for examination. He adjusts it with tweezers on a small tray and closes it into a black cardboard box, hewn together with duct tape.
To study the zebrafish brain in action, Harris explains, they must first immobilize it by affixing a light gelatin substance to its neck. That stabilizes the head, but allows the eyes and tail to dart back and forth.
Later, he will secure it onto a Lego-sized plastic mount, which was built by threading spools of plastic onto a heated nozzle in a 3-D printer that sits on Briggman’s desk. Then he will slide it under a two-photon microscope where he’ll project a stimulus, in this case an L.E.D. micro-projector light that simulates the movement of fish prey like plankton. The neurons in Briggman’s fish have been genetically engineered to glow green each time they fire. So in real time, they can watch the brain networks at work as the fish turns left, turns right and approaches the light. They can watch as it makes the decision to lunge at something and try to capture it.
Normally, the eyes of the fish lie parallel against the head, but while hunting, they bend forward, endowing the animal with binocular vision. Each flick of the tail brings it closer to its prey.
“We try to basically fake out the fish, to trick the fish into thinking it’s looking at a tiny paramecium moving in front of it,” Briggman says.
In this way, they are effectively watching the fish brain make a decision and studying the neural mechanisms at work as it does so.
“When the light turns on, you see a pulse of activity — neurons lighting up on the screen,” Harris says.
On a computer screen near the microscope, some nerve cells light up alone against a black backdrop. Others appear tangled in a mess of axons and dendrites. Using a computer program, the team will go on to identify each neuron and circuit that responds to the light.
For these fish, the final destination is the electron microscope, where scientists plunge into a dense, three-dimensional model of the brain. Since axons can be as small as 50 nanometers — that’s 2,000 times thinner than the width of a human hair — imaging them requires a higher resolution than any optical microscope can provide. But a living brain can’t withstand the power of an electron beam, so the fish must be killed and its brain preserved in formaldehyde, embedded in plastic for structural stability, stained with heavy metals and then sliced into tens of thousands of sections with a sharp-edged diamond knife and a device not unlike a miniature deli slicer.
The heavy metals stain the fatty tissues, which provide enough contrast to see details, while protecting the brain from the powerful blast of high voltage electron beams. “You need conductive material so the electrons don’t get stuck in the plastic surrounding the brain,” Harris says.
The zebrafish is a good model for studying the vast depths of a brain and how it’s wired, said Scott Emmons, a professor of molecular genetics and neuroscience at the Albert Einstein College of Medicine who is not involved in the research. Emmons and his team recently published a more complete reconstruction of the C. elegans nervous system.
“The zebrafish is the right size in the sense that it’s really small, but it’s still big in terms of connectomics,” Emmons said. “It’s one of the biggest ones you can reach.”
And it could help unlock an outstanding mystery in genetics, Emmons said: how brain connections are coded in the genes.
“There’s still a whole class of genes we don’t know,” Emmons said. “Those are genes that determine connections in the nervous system. That’s a major outstanding problem in all of neuroscience.”
Once Briggman’s team finds the perfect fish brain — a healthy brain with strong contrast for visualizing the neurons and their networks — they will prepare it for the electron microscope, where it will spend as long as six months collecting 10 terabytes of structural data.
And then there’s the analysis, which requires following each neuron three-dimensionally through a maze of other neurons.
“When you’re moving through a data set, you have the perception that you’re flying through this data,” Briggman said.
Zebrafish and their transparent brains aren’t only useful for mapping. They are increasingly being used as models for certain neurological disorders. Believe it or not, fish can be used to model depression, Alzheimer’s and Parkinson’s-like symptoms.
“That becomes very relevant to human disease,” Briggman said. “If we see part of the brain wired in a different way in a depressed fish, in a fish that doesn’t eat, or one that overeats or that likes alcohol in the water too much, that gives us something that can be tested in humans.”
Briggman’s main question is how circuits work within the brain to perform certain computations. For example, to detect motion or encode memories.
“All of those happen at the level of individual neurons and individual synapses,” he said. “What I hope is that we can learn the general principles of how circuits in the brain do these things. And then test the ideas in higher-order vertebrates and mammals.”
For Harris, the value of the research also extends beyond the basic mechanics of the fish brain. The fish is a means to understanding something much more conceptual: how the brain can “want” something, he says. As the fish flicks his tail on a screen behind him, he launches into an explanation about dopamine, a natural neurotransmitter involved in the basic experience of pleasure. Chocolate, sex, gambling and nicotine all cause a rush of dopamine in the brain’s reward center.
“A friend of mine once came to me who was trying to quit smoking, and he just said that he didn’t understand why he kept doing it, because every time he smoked afterwards he felt miserable,” Harris says. “And I told him about dopamine — the fact that dopamine tends to be released before and just when you’re about to get what you want and then for a few moments when you’re getting what you want, but then it drops.”
It explains, he says, why people chase that initial rush, even though it doesn’t match the lousy feeling that results. Knowing how the brain and its chemicals work might help people tackle self-destructive behavior, for example.
“Understanding why we fail to follow through on our long-term goals, that’s really a huge thing for me,” he says.
So what does that have to do with the fish?
“The machinery that allows us to identify something in the world as important and pursue it…,” Harris says. “All that is there in the fish, and it’s working perfectly.”
Video edited by Rebecca Jacobson.