Adapted from a story that originally appeared on the Alzheimer Research Forum.
26 February 2013. Despite decades of research, scientists still do not understand how neuronal firing encodes thought and behavior. To decipher this, a group of scientists proposes an ambitious project to map the activity of every neuron in the brain, and correlate the data with behavior and disease states. The White House has embraced the idea, with President Obama mentioning the plan in his 2013 State of the Union speech (see New York Times story). The project is expected to appear in the president’s forthcoming federal budget proposal. Proponents suggest that this Brain Activity Map (BAM) Project, like the Human Genome Project, represents an investment in future technology and basic research that will pay tremendous dividends later. The findings may eventually lead to better treatments for conditions such as Parkinson’s disease, epilepsy, and schizophrenia, and perhaps even Alzheimer’s disease, said Rafael Yuste at Columbia University in New York City, who helped develop the proposal. BAM plans to start with animal models while developing non-invasive technology to eventually study human brains. Human studies will only come much later, despite the emphasis in popular press reports on human brain mapping.
“This is a thoughtful and provocative set of ideas that will accelerate our efforts to decipher the fantastic complexities of brains,” David Van Essen at Washington University in St. Louis, Missouri, told Alzforum. He is not involved with the BAM proposal. However, he co-leads the Human Connectome Project, another large brain mapping initiative. This ongoing study maps the structural wiring of the human brain. It also gathers whole-brain activity data using functional magnetic resonance imaging (fMRI). Van Essen noted that the proposed Brain Activity Map Project will provide an immense number of data that will be complementary to the Human Connectome Project. By measuring the activity of large ensembles of individual neurons in animal models, the BAM Project could help researchers develop computational models of how brain circuits operate. “That would give us a much better basis for interpreting and making sense of the vast numbers of fMRI data being obtained on whole-brain activity in humans,” Van Essen said.
With the federal budget still in limbo, it is unclear how much funding the BAM Project will receive. Story Landis, who directs the National Institute for Neurological Disorders and Stroke (NINDS), said that the project is still in the planning stages. At the moment, Landis and Thomas Insel, who directs the National Institute of Mental Health (NIMH), manage the effort. Strategy meetings are ongoing. One possibility is that the project will be funded under the National Institute of Health’s Blueprint for Neuroscience Research, which tackles large projects beyond the scope of a single institution, Landis said. Scientists involved in the project claimed that private foundations will also commit funds to the project, although no public announcements have been made. At the request of the White House, scientists stayed mum on funding details, such as how much money the project might need. For comparison, government spending on the Human Genome Project amounted to almost $4 billion over 13 years, while the National Alzheimer’s Project Act has been granted about $130 million so far (see government press release).
The BAM proposal gestated at workshops supported by the Kavli Foundation, the Gatsby Charitable Foundation, and the Allen Institute for Brain Science. Six of the scientists involved published their thoughts in a June 2012 Neuron paper (see Alivisatos et al., 2012). They pointed to a need for data to bridge the gap between traditional electrophysiology, which samples from only a handful of neurons at a time, and whole-brain imaging methods like fMRI, which lack fine resolution. The solution is to record from thousands or millions of neurons at once to reveal the detailed behavior of entire circuits, the authors suggest. They plan to start with simple organisms such as C. elegans or Drosophila, which have manageable numbers of neurons. Initial circuit mapping could use the well-established technique of calcium imaging, since calcium currents reflect neuronal activation. Ultimately, however, the researchers will need to record voltage, said senior author Yuste. “We don’t yet have a good voltage indicator. Some of the most promising leads for future development come from nanotechnology.”
The project may draw upon advances and new technology generated by the National Nanotechnology Initiative. For example, some inorganic nanoparticles are sensitive to the surrounding electrical field and can emit light in response to excitation, making them promising voltage indicators. Even more useful would be nanoprobes, small silicon arrays stuffed with electrodes. Nanoprobes with dozens of electrodes are already available, and the devices could potentially hold thousands, allowing massive parallel recording from neurons in a circuit, the authors note.
These methods would be too invasive to use in people. The researchers plan to develop non-invasive methods for human studies. These might include optical imaging of electrical and chemical activity, or using engineered DNA molecules to store action potential spike data, Yuste said. In this latter approach, synthetic cells would contain the DNA and serve as local reporters of brain activity. Because the error rates of DNA polymerases depend on cation concentration, patterns of errors would correspond to spike frequencies. This method would take advantage of the vast information storage capacity of DNA. Yuste noted that these approaches are still speculative.
Despite the technical hurdles, commentators agreed that the proposal is likely to further knowledge of the brain’s workings. “Many of the methods are likely to yield fruit, especially if there is a substantial investment in new technology,” Van Essen said.
In fact, the initiative may well spur major advances in neuroscience techniques. “MRI methodology has improved tremendously in the last decade, but there has not yet been the same kind of revolution in the way that we record activity in neural circuits,” Landis pointed out. This effort may provide those tools, she suggested. Marcus Raichle, also at WashU, said, “This represents a big-time investment in technology development. You can imagine that a lot of interesting things will come out of it that may or may not be relevant to our understanding of the human brain.”
The authors of the Neuron paper draw a parallel to the Human Genome Project, which launched a revolution in sequencing methods that dramatically dropped the price of sequencing and stimulated the genomics industry. An analysis by the research firm Battelle Technology Partnership Practice, which helps forge public-private partnerships, estimated that every dollar invested in the Genome Project returned $141 to the economy. This includes job creation and tax revenue from genomics companies. Moreover, the scientific impact continues to grow, the report notes. The BAM Project has the potential to do the same thing, supporters suggest. The Human Genome Project has not translated into a treatment for AD, though it has helped researchers pinpoint genetic variants that increase risk for the disease.
Researchers involved in the BAM Project believe that the work may one day lead to a number of therapeutic applications. “The deepest motivation to do this is a clinical one,” Yuste told Alzforum. For example, eventually he hopes to be able to compare brain activity in people with schizophrenia to that in healthy controls to find abnormal patterns of activity. “If we can do that, we may be able to devise strategies to alter those abnormal patterns and re-channel the activity,” he said. The findings might help treat other circuit disorders like epilepsy or even autism, he suggested. Landis sees additional therapeutic possibilities. Currently, deep-brain stimulation (DBS), which involves placing electrodes in specific regions of the brain, alleviates symptoms in people with Parkinson’s disease, depression, and is showing promise in other diseases. With a better understanding of how the circuits work, “We would be able to make much more sophisticated changes,” Landis said. Such knowledge would also allow researchers to develop better brain-machine interfaces for controlling prosthetic devices, already a subject of intense investigation.
Direct applications for Alzheimer’s disease are less clear, since the disease is neurodegenerative rather than one of faulty wiring. However, brain connections also falter in AD. Yuste sees another possibility. Nanoprobes could be modified to serve as chemical sensors, detecting molecules such as β amyloid or tau. This technology could allow researchers to study the distribution and concentration of these molecules in real time in animal models of AD, Yuste suggested. In this way, nanotechnology being developed for the BAM Project could advance AD research.
Raichle noted, however, that given the length of time before such methods are ready, other therapeutic approaches to AD may well provide answers first. He also stressed that the BAM Project complements, but does not replace, other methods of studying the brain. “There are very important metabolic processes that don’t necessarily track with the firing of an action potential,” he said. One example is brain glucose use and metabolism, which can be visualized with FDG-PET.—Madolyn Bowman Rogers.