Brain shot

Neuroscientists scramble to take on ambitious presidential challenge

BRAIN POWER  Deciphering how the brain’s circuitry produces thought and behavior is an ambitious and enticing goal on the scale of the Apollo Program or the Human Genome Project. But the neuroscientists involved in a new federal effort have many challenges ahead.

Michael Morgenstern

When the president of the United States makes a request, scientists usually listen. Physicists created the atomic bomb for President Roosevelt. NASA engineers put men on the moon for President Kennedy. Biologists presented their first draft of the human genetic catalog to an appreciative President Clinton.

So when President Obama announced an ambitious plan to understand the brain in April 2013, people were quick to view it as the next Manhattan Project, or Human Genome Project, or moon shot.

But these analogies may not be so apt. Compared with understanding the mysterious inner workings of the brain, those other endeavors started with an end in sight.

In a human brain, 85 billion nerve cells communicate via trillions of connections using complex patterns of electrical jolts and more than 100 different chemicals. A pea-sized lump of brain tissue contains more information than the Library of Congress.

A central part of the initiative focuses on the need to improve existing technologies and to develop new ones that probe the brain’s inner workings on smaller and smaller scales. Learn more about those technologies below. <a href=”#slideshow”>View the slideshow</a> Y. Kaneonke et al/PLOS One 2012

But unlike those orderly shelved and cataloged books, the organization of the brain remains mostly indecipherable, concealing the mysteries underlying thought, learning, emotion and memory.

Still, as with other challenging enterprises prompted by presidential initiatives, success would change the world. A deep understanding of how the brain works, and what goes wrong when it doesn’t, could lead to a dazzling array of treatments for brain disorders — from autism and Alzheimer’s disease to depression and drug addiction — that afflict millions of people around the world.

That’s why President Obama threw his weight behind the BRAIN Initiative, short for Brain Research through Advancing Innovative Neurotechnologies (SN: 5/4/13, p. 22). The premise is simple: Before doctors can fix the brain, scientists must first understand how it works. And to understand how it works, scientists need tools to study it. With $110 million of federal funding in its first year, the BRAIN Initiative is intended to spur scientists to develop new technologies to measure and manipulate the brain. Eventually, if it is to join the list of presidential science successes, the project will catalog all the brain’s parts and processes, explore how cells and molecules create thought and behavior, and build powerful new weapons for neutralizing the pathological enemies of the brain and mind.

Yet even aside from those scientific challenges, which are all huge in their own right, the project faces many major logistical hurdles.

It’s not clear, for instance, how various govern­ment agencies and private institutions involved in the project will coordinate their efforts. Nor is it clear how the BRAIN Initiative will relate to the European Union’s $1.3 billion Human Brain Project. Some scientists say the BRAIN Initiative’s initial funding is too paltry to make real progress and that future funding is a political uncertainty.

Perhaps most unsettling, the BRAIN Initiative has no definitive goal. Unlike mushroom clouds, a collection of moon rocks or the software for a human being, the BRAIN project envisions no tangible result, many scientists say. “It isn’t clear what victory will look like on this project,” says Thomas Insel, director of the National Institute of Mental Health in Bethesda, Md. “I think people have to be comfortable with that.”

Despite these caveats, though, many neuroscientists appreciate that President Obama’s announcement elevated the status of brain research and captured the attention of their community. “When the president says it, people listen up,” says Christof Koch of the Allen Institute for Brain Science in Seattle. “I think that, by itself, is a very important thing. It really shows that neuro­science has come of age.”

Ambitious goals

While the BRAIN Initiative’s objectives are hard to express in concrete terms, the project is full of visionary promise. “The ultimate goal is to understand who we are,” says Terry Sejnowski of the Salk Institute for Biological Studies in La Jolla, Calif. “How is it that our brain is able to look out into the world and see things? How is it that we are able to make decisions? How is it that we’re able to coordinate enormous amounts of knowledge?”

The people charged with translating these esoteric goals into concrete action are beginning to define their task more precisely. But there is no consensus on how to proceed. With no central organizing entity, the three government agencies participating — the National Institutes of Health, the Defense Advanced Research Projects Agency and the National Science Foundation—interpret the BRAIN Initiative’s mission in their own ways. So do the private organizations involved, including the Salk Institute, the Allen Institute, the Kavli Foundation and the Howard Hughes Medical Institute’s Janelia Farm campus in Ashburn, Va.

The NIH, which is putting up $40 million in funding for fiscal year 2014, has taken a methodical approach by first appointing a committee of 16 neuroscientists. That group, headed by William Newsome of Stanford University and Cornelia Bargmann of the Rockefeller University in New York, spent the summer of 2013 in four workshops talking with neuroscientists about what ought to be included in the initiative.

In September, the committee released a preliminary report describing nine priorities. They were extremely ambitious. For instance, the NIH panel wants a census of all the different types of brain cells and a description of how nerve cells, or neurons, collectively give rise to behavior. Powerful new technology also features prominently on the wish list. One of the goals calls for developing tools capable of producing anatomical maps of the human brain with unprecedented clarity.

Further priorities include developing techniques that can eavesdrop on many neurons at the same time and allow scientists not just to listen in, but to change how those neurons behave. To make sense of all of this data, the report calls for improved theoretical, statistical and modeling approaches.

Each of these goals on its own could easily take years or even decades to accomplish. For the final report, the goals, milestones and timeline will be sharpened, says Sejnowski, a member of the working group.

Like NIH, DARPA also emphasizes new tools, but with a much more targeted goal in mind for its $50 million investment in 2014: healing soldiers. “We serve a constituency” — the active duty service member — says Geoffrey Ling, deputy director of the Defense Sciences Office. “And the active duty service member right now has got a lot of issues medically.”

Currently, more military members die from suicide than from direct combat injury, says Ling. “And that’s the tip of the iceberg.” DARPA interprets its role in the BRAIN Initiative as alleviating some of the pernicious mental health problems that plague service members.

In October and November, DARPA announced two projects. One, called SUBNETS (for Systems-Based Neurotechnology for Emerging Therapies), seeks new ways to record neural activity from and stimulate the brains of people with post-traumatic stress disorder, anxiety disorder, traumatic brain injury and other diseases. DARPA envisions a device that both diagnoses and treats mental health problems, first by listening for abnormal electrical signals and then correcting them. For success, the project will need engineers to build new medical devices, computational neuroscientists to develop theories about how neurons transmit information and clinicians to test the prototype in people.

The second project, called RAM (for Restoring Active Memory), aims to develop an implantable brain device that will help restore lost memories to soldiers or veterans. Devised by Ling, the project plans to move from idea to device quickly. “Our timeline is four to five years, and we’re not joking,” he says.

Seeking new tools

Achieving such ambitious goals — figuring out how neurons create behavior, curing mental illness and restoring lost memory — is not possible with today’s technology. Even though scientists have made huge leaps in their ability to listen to and manipulate neurons, current methods are still far from where they need to be.

“The tools have to be the focus,” says Sejnowski. “We have to get those tools in place. There’s no way of even getting off the ground until we have those tools.”

In optogenetics, scientists use light to manipulate brain cells that have been genetically modified to respond to light. While already producing useful findings in animals such as mice, optogenetics would not be practical in people. Inbal Goshen & Karl DeisSeroth
NIH’s initial plans reflect that belief. On December 17, the agency released six calls for projects to fund as part of the BRAIN Initiative. Each describes a tool-building plan. “What we’re trying to do is get the tools and infrastructure in place so we can get a much deeper understanding of how the brain works in both health and disease,” Insel says.

Scientists want to monitor the electrical and chemical behavior of many neurons — thousands or even millions — at the same time, while being able to zoom in to see and even manipulate those cells. One of today’s common ways to eavesdrop on neuron behavior relies on electrodes designed decades ago. Neuroscience is still stuck using technology from the 1950s, Insel says, “while the rest of the world has learned how to go wireless and miniaturized.”

Magnetic resonance imaging, or MRI, allows scientists to get good anatomical maps of the whole brain and broad activity patterns (SN: 12/19/09, p. 16). But as good as it is, MRI technology still misses lots of detail. A million neurons can reside in a single voxel, the smallest unit that functional MRI can detect. “MRI shows you wonderful neuroanatomical details, fantastic, but it does have a resolution limit,” Ling says. “How can we increase it? Easy — build a bigger magnet. Oh, good, let’s have a 50 Tesla magnet. What city are you going to put this in? Because you have to wipe out about seven blocks to do it.” Supersizing existing technology won’t work, Ling says. Fundamentally new ideas are needed.

Radically new approaches will also be needed to enable precise control of neurons’ behavior. One powerful new technique, called optogenetics, lets researchers use light to control certain brain cells in animals. But it’s not feasible in people, because it requires genetic alterations to make neurons

produce specialized light-sensitive proteins. For now, scientists are forced to rely on less precise methods to change the activity of human neurons.

Psychiatric drugs can alter neuron behavior, for instance, but the results are imprecise, like dousing the entire engine of a car with oil. Every cell in the brain gets dosed, when only a select few actually need the drugs. What’s more, in most cases scientists still don’t understand how psychiatric drugs work.

A different, drug-free approach may work better: newer technology called deep brain stimulation. It serves as a brain pacemaker. By zapping neural highways with electrodes implanted in the brain, deep brain stimulation has shown promise for treating people with Parkinson’s disease and severe depression. And yet, as with psychiatric drugs, the technique is still imprecise and not well understood.

“I think people understand that you’re not going to be able to fix something as complicated as the brain with the current tools. They’re too crude,” Sejnowski says, like “trying to fix a computer with a wrench.”

Instead of more wrenches, scientists need powerful precision tools, some of which are in the works. The Allen Institute, HHMI and others have partnered with a nanoelectronics research center called imec to build tiny but powerful electrodes that can record the behavior of hundreds of neurons with great accuracy. “That’s going to be pretty awesome technology,” Koch says.

Scientists at Janelia Farm are devising intricate ways to illuminate neural behavior in zebrafish, flies and mice. By genetically engineering neurons to produce a protein called GCaMP, the researchers can watch each individual neuron fire off a message. The concept of using proteins to detect neuronal activity isn’t new, but the sensitivity of this latest version is much better than previous attempts, says Gerald Rubin, Janelia Farm’s executive director. It awed many in the field in March 2013 when the Janelia Farm researchers circulated a movie of zebrafish brain activity.

Scientists are also making progress on mapping the connections between neurons. Most attempts rely on powerful microscopy (SN: 6/15/13, p. 20). But there might be a better way, says Anthony Zador of Cold Spring Harbor Laboratory in New York.

Instead of using microscopes, Zador and colleagues are attempting to attack the problem with DNA sequencing, which is cheap and reliable. Their method relies on using genetic tricks to tag neurons with unique chains of DNA. By analyzing how those DNA tags mingle at synapses — the communication connections between neurons — a computer could reconstruct all of the physical connections in a brain. So far, the team has had success only in cells in a dish. If the technique works in animals, figuring out every single synapse in the entire mouse cortex will cost just a few thousand dollars, Zador estimates.

Making the vision real

If the BRAIN Initiative serves as an incubator for the next great brain technology, the payoff would be huge, Newsome says. The ability to record the behavior of tens of thousands of neurons simultaneously, map the connections between those neurons and then manipulate those neurons, all in a fully awake behaving animal, or person, “is not something that neuroscience has ever been able to contemplate in its history,” says Newsome. “It’s kind of a breathtaking vision.”

Success in realizing that vision, though, will mean that neuro­scientists have to face a big shift in how they do their work.

As neurotechnology improves and produces ever more massive piles of complicated data, neuroscientists will need to know more about statistics, engineering and computational biology. To succeed, the BRAIN Initiative will need to bring together experts in all these different fields. It will also need neuroscientists to coordinate their efforts and share their data freely. “We are going to have to reorganize the way people do their research,” Sejnowski says. The same goes for the nongovernmental agencies working on the BRAIN Initiative.

Currently no one agency or person is in charge of the initiative. Such decentralization might make swift progress, without duplication of effort, difficult. Although the three government agencies involved — DARPA, NIH and NSF — have been aware of what the others were doing, they haven’t merged their differing priorities. “We’re all so busy getting these projects launched, I don’t think we’ve had a lot of time to think about how they’re going to be integrated,” Insel says.

The private groups, referred to by the White House as BRAIN Initiative “partners,” set their own course and spend their own money as they see fit. Because the goals of the Allen Institute, the Salk Institute and Janelia Farm align closely with federal goals, these institutes don’t anticipate changing their previously set research agendas.

One organization working with the BRAIN Initiative, the Allen Institute for Brain Science, has been developing an atlas of connections in the mouse brain (top). Scientists at the Howard Hughes Medical Institute’s Janelia Farm campus have been probing the brain of fruit flies (bottom). Courtesy of the Allen Institute for Brain Science; Courtesy of the Simpson lab/Janelia Farm Research Campus

“Partnership is a funny word,” says Janelia Farm’s Rubin. In a sense, Janelia Farm is a partner, he says, because it is putting up a big chunk of money — $30 million annually — toward projects that dovetail with the BRAIN Initiative. Much of that research focuses on how information in the fruit fly brain is stored and processed. “But we’re not a partner in the sense that we’re getting together at a big table with NIH folks and NSF and DARPA and deciding what the goals are and then collectively deciding what we’re doing,” Rubin says.

Likewise, Koch says that the Allen Institute has been working on most of the questions the NIH group came up with. “We are fulfilling already a part of the BRAIN Initiative mission,” he says. Allen Institute scientists have been building detailed maps of the mouse brain, which could help efforts to understand how the human brain works.

Aside from the privately funded research that is already under way, major new BRAIN Initiative research will require new federal money. And so far, the new money has been scant. In 2014, NIH is contributing $40 million to the BRAIN Initiative, less than 1 percent of the $5.7 billion that the agency spends on neuroscience research each year. In 2014, DARPA will spend $50 million on its newly announced projects, and NSF will spend $20 million on relevant projects already in progress.

Although to many scientists, the BRAIN Initiative’s preliminary pot seems small, that money is a down payment, Insel says, meant to get the ball rolling in the hope that Congress will provide more support for the project.

Of course, to help sell the project to politicians who vote on budgets (or as has recently been the case, don’t vote on budgets), scientists need a tagline, a slogan, a simple way to encapsulate the importance of the BRAIN Initiative. So far, there is no such thing.

“We’ve talked and thought about that a lot in the committee,” Newsome says. But the group couldn’t come up with a pithy way to capture the project’s essence. “That question, it’s a real tension. To what extent do we try to depict this as an Apollo-like project or a genome-like project, in which case you have a tagline and a particular deliverable you’re looking for? Or to what extent do you acknowledge that this is a more open-ended kind of project?”

The sell is important, says Zador. In the current economic squeeze, funding must be justified to politicians and the public. If the BRAIN Initiative underdelivers, people will be disappointed.

Instead of a promise to cure brain disease, or unlock the mysteries of memory, emotion and thought, perhaps the brain project is best described as a wedge, says Sean Eddy of Janelia Farm. With its emphasis on sophisticated tools, the project promises to pry open an entirely new realm of neuroscience research, Eddy wrote last April in Current Biology, enabling countless labs around the world to make discoveries in their small corner of the brain.

This vision of the initiative’s success, in which thousands of neuroscientists storm the inhospitable terrain of the brain armed with an awe-inspiring new arsenal of tools, is staggering. Compared with any other project President Obama could have backed, the brain is the most worthy, Eddy says. “This is it.”

If the project flops, scientists will still learn a lot in the attempt. If the project succeeds, the benefits are almost unimaginable. Clinicians might be able to neutralize — or even prevent — devastating disorders such as autism, Alzheimer’s disease and traumatic brain injury. Computers could become ever more powerful by cribbing from the brain’s operating system. Classrooms, military training camps and courtrooms could be optimized to play to the strengths of the human brain and protect it from its weaknesses. These are among the motives that drive neuroscientists, and for all its shortcomings, the BRAIN Initiative has already succeeded in getting people dreaming about what might be possible.

“I can’t tell you where we’re going to be two years from now,” Sejnowski says, “but I can tell you that it’ll be far ahead.”


Priorities for the BRAIN Initiative

In response to President Obama’s call for a major new initiative to understand the brain, the National Institutes of Health organized a committee to articulate a list of priorities. The group identified nine, briefly summarized here:

1. Generate a census of cell types
Characterize all cell types in the nervous system and develop tools to observe and manipulate them.

2. Create structural maps of the brain
Map connected neurons in local and distributed circuits in order to understand the relationship between neuron structure and function. Create tools to reconstruct anatomy of neural circuits at all scales and identify circuit inputs and outputs.

3. Develop new large-scale network recording capabilities
Record neuron activity from complete networks over long time periods in all brain areas with the use of improved existing technologies and entirely new technologies.

4. Develop a suite of tools for circuit manipulation
Activate or inhibit activity in neuronal circuits, using opto­genetic, pharmacological, biochemical and electromagnetic tools.

5. Link neural activity to behavior
Understand the neural basis for cognition and behavior, using technologies that can precisely observe behavior under a broad set of conditions while measuring and manipulating neuron activity.

6. Integrate theory, modeling, statistics and computation with experimentation
Apply theory and statistics to understand complex brain functions where human intuition fails, using new methods for mining, analyzing and interpreting data.

7. Delineate mechanisms underlying human imaging technologies
Improve spatial and time resolution of human brain imaging techniques while developing a better understanding of the cellular mechanisms underlying the signals used to compose images.

8. Create mechanisms to enable collection of human data
Develop high ethical standards for clinical care research while developing methods to maximize the collection of beneficial information from humans undergoing diagnostic brain monitoring or receiving neurotechnology for clinical applications.

9. Disseminate knowledge and training
Accelerate progress by the rapid dissemination of newly developed skills and tools across the scientific and medical communities.


Brain Technology

President Obama’s BRAIN Initiative seeks a comprehensive understanding of how molecular and electrical processes orchestrated by nerve cells produce the brain’s ability to think, learn and control behavior. A central part of the initiative focuses on the need to improve existing technologies and to develop new ones that probe the brain’s inner workings on smaller and smaller scales.

OLD TECH  Two techniques using magnetic resonance imaging, or MRI, have already made progress in mapping the physical connections (fibers of white matter) linking various parts of the brain and in identifying regions that are active during specific tasks.

NEW TECH  Several novel technologies have been developed or proposed to acquire even more detailed data about how the regions and even individual cells of the brain do their jobs.

OLD: Functional MRI This process records activity in different brain areas by tracking the flow of blood and its oxygen content. Active brain cells require more blood to provide oxygen. fMRI signals show which parts of the brain are at work during certain tasks or behaviors. Y. Kaneonke et al/PLOS One 2012
OLD: Diffusion tensor imaging This process maps white matter fibers by measuring the diffusion of water. The diffusion rate is faster parallel to fibers than perpendicular; mathematical analysis of the flow rate in various directions (using quantities called tensors) allows reconstruction of fibers’ locations. Courtesy of the Laboratory of Neuro Imaging and Martinos Center for Biomedical Imaging, Consortium of the Human Connectome Project – www.humanconnectomeproject.org
NEW: CLARITY Short for Clear Lipid-exchanged Anatomically Rigid Imaging/immunostaining-compatible Tissue Hydrogel, this replaces fat cells in the brain with a clear gel, allowing neurons to be illuminated. By infusing a brain with a liquid that eventually hardens into a gel, scientists can see previously hidden structures deep within the brain. Already, CLARITY has been used to see nerve fibers and connections in a mouse brain, and even individual neurons in a human brain that had been preserved in formaldehyde for six years. Kwanghun Chung & Karl Deisseroth/HHMI, Stanford Univ.
NEW: GCaMP This is a protein that labels active neurons in the whole brain. Designer proteins can sense when nerve cells are active and change color as a result. Calcium ions that are present when a neuron fires change the shape of GCaMP proteins, causing them to fluoresce and creating a beacon that can be detected by sensitive microscopes. Scientists have used these proteins to watch the behavior of more than 80 percent of all neurons in the brain of a zebrafish larva. Philipp Keller & Kristin Branson/Janelia Farm Research Campus
NEW: DNA tagging By labeling each neuron with a DNA bar code, scientists can reconstruct the physical connections between neurons quickly and cheaply. So far, this method has been used to reconstruct all the connections between cells in a dish. Otwell
NEW: Mini electrodes For decades, scientists have used electrodes to eavesdrop on neurons. But those electrodes are limited in their ability to record many neurons at one time. Scientists and technology companies are working to make smaller, more accurate electrode arrays that can both accurately detect and change the behavior of hundreds or thousands of neurons at once. Gyorgy Buzsaki Lab/ NYU

Laura Sanders is the neuroscience writer. She holds a Ph.D. in molecular biology from the University of Southern California.