When The New York Times reported last month that the Obama administration was planning to launch a Brain Activity Map (BAM) project at a projected cost of $1 billion or more, there was much criticism within the scientific community. Aside from funding concerns and administrative issues, many researchers questioned whether researchers possessed the tools and technical capabilities necessary for such a tremendous undertaking. Would the BAM project develop the technology and data processing capabilities needed to fully map the connections in the human brain? And what types of new techniques might emerge as a result of this initiative?
At its core, the BAM project is really a proposal to "help catalyze action," say the authors of a Science paper published in March outlining the project yet again. Initially, new tools for imaging or recording the individual activity of most if not all neurons in a brain circuit would need to be built.. From there, because the aim is to understand the function of the circuits in the brain, techniques for controlling the activity of each individual neuron in the circuit need to be developed. The authors see the project as an "open, international collaboration of scientists, engineers, and theoreticians, throughout academia and industry." They set a target of observing 1 million neurons in 15 years. How might they achieve such a lofty goal?
The Power of Light
Several promising light-based technologies have emerged in recent years for mapping brain activity. One such approach is multiphoton microscopy, where two or more photons are used to excite chromophores in vivo or in brain slices from mice. Using light at the infrared end of the spectrum and moving toward three photons instead of two, which has been the standard since the technique was developed in the 1990s, allows researchers to penetrate deeper into the brain.
Yuste says his lab, in collaboration with Karl Deisseroth at Stanford University and others, is working to combine two-photon microscopy with optogenetics in order to selectively excite neurons at different depths within living tissues. In a paper published in Nature Methods, they used this two-photon optogenetic approach, along with a newly developed light-activated cation channel from Deisseroth’s lab called C1V1, to generate action potentials in neurons with single-cell precision and to map neural circuits in mouse brain slices. In addition, Yuste’s group was also able to split the laser beam, which allowed simultaneous activation of neurons in three dimensions.
"With optogenetics you can really determine the functional connectivity of cells in a neural network," says Ed Boyden, a neuroscientist at MIT. "You can aim light at the exact set of cells you want and thereby activate those cells and not the other ones."
Similar to optogenetics, optochemical genetics is another light-based approach that holds promise for mapping brain activity. Optochemical genetics takes advantage of caged ligands—compounds that are synthesized in the lab by taking any neurotransmitters containing nitrogen (glutamate, GABA, serotonin, nicotine, among others) and attaching an inactivating group that is photolabile. The caged ligand remains inactive until exposed to light, at which time the inactivating group is photocleaved and releases the active neurotransmitter. In this way, researchers are able to tease apart the functions of individual neurotransmitters in specific circuits within the brain.
Shoot for the Moon
Even the most ardent proponents of the BAM project agree that mapping the activity of the brain is not something that is going to happen overnight, or even in the next few years. And whether it could be done with optogenetics, optochemical genetics, or some other as yet undiscovered technique remains to be determined.
"I think it will be similar in timeframe to the human genome," says Yuste, meaning BAM could require 15 years or more to complete.
And yet, some researchers are plunging headfirst into mapping brain activity already. Just last week a paper was published in Nature Methods by Phillipp Keller and others at the Howard Hughes Medical Institute's Janelia Farm Research Campus in Ashburn, Virginia, detailing the mapping of 80 percent of the activity of the zebrafish brain using calcium imaging and a new microscopy technique known as light-sheet microscopy. Although the approach generated images very slowly—about one per second—the results were still "quite remarkable," says Yuste.
Even if the ultimate goal of mapping the complete set of neuronal connections within the human brain is never achieved, each small part of the map that is generated can yield fundamental insights into how neurons compute, or fail to compute in the case of brain diseases, perhaps leading to new treatments for circuits that have gone awry. And, Boyden points out, these type of "moon-shot" projects with heroic goals often spin off technologies that have powerful applications—applications that may have not yet even been imagined yet.
"The idea that we can bring lots of technology ideas into neuroscience is a very powerful one," says Boyden. "Most fields of engineering have not been applied to the brain. If we analyze these potential connections between different engineering fields and neuroscience, it might foster the development of fundamentally new technologies for revealing the underpinnings of brain operation and for clinical targets, too. The science can drive the technology as much as the technology can drive the science," says Boyden. "And that interplay is very exciting."