Decoding the BRAIN Initiative
The Obama BRAIN Initiative and technologies that make it possible.
BRAIN: America, The Ambitious
Folks who worry about America’s competitiveness in a globalized economy, and its capacity to do big things, need to look no further than the 21st Century Grand Challenges issued by the White House. From identifying all asteroids that threaten the earth to making electrical vehicles affordable, these challenges symbolize America at its innovative and entrepreneurial best. Last year, President Obama unveiled, perhaps, the most ambitious of all challenges, the BRAIN Initiative. The human brain, with its 100 billion neurons and their 100 trillion connections, is the most complex entity in the universe. The BRAIN (Brain Research through Advancing Innovative Neurotechnologies) Initiative aims to revolutionize our understanding of the human brain by accelerating technological innovation to give scientists “the tools they need to get a dynamic picture of the brain in action and better understand how we think, and how we learn, and how we remember”. By wedding technological innovation to the understanding of the brain, the administration has shown remarkable astuteness in addressing multiple objectives, from advancing human knowledge, to creating the jobs and industries of the future, to forging a high-impact partnership between the public and private sectors. In short, the BRAIN Initiative is exactly the kind of project that would allow Thomas Friedman to get a good night’s sleep!
Focus On Neural Circuits
So, what exactly is the BRAIN Initiative? The interim report of the BRAIN Initiative Working Group, co-chaired by neuroscientist rockstars, Cornelia Bargmann (HHMI/Rockefeller) and William Newsome (HHMI/Stanford), is a good starting point to demystify it. I absolutely urge you to read its inspiring preamble and executive summary, if not all 58 pages. The interim report provides a robust conceptual framework for the initiative by firmly anchoring it in technological innovation for investigating the brain at the level of neural circuits. So, why is circuit level investigation important? And, why is further technological innovation needed?
The brain is like a digital painting. Focus on a few pixels, and you get an incomplete view. Observe from too far, and you lose detail. But, stand at just the right distance, and you see that "cohesive whole".
Modern understanding of the brain lacks this coveted perspective. That is because today's technologies inherently limit researchers to studying the brain at the level of a few individual neurons (poor scale), or at the level of a million neurons as a whole (poor resolution). But, the brain really operates at the level of neural circuits. Neural circuits are comprised of populations of interconnected neurons that work in concert to perform specific functions (for example, vision, feeding, or decision-making). So, the hope is that by studying large populations of neurons at single cell resolution, we will understand how neural circuits function, how they interact with each other, and how they change with time, to ultimately obtain a cohesive picture of the brain in action. This will, however, require technological innovation on an unprecedented scale.
The Foundational Technologies
Acquiring an authentic and comprehensive understanding of neural circuits is essentially a four-pronged effort. It involves: 1) Identifying and profiling all the neuronal and glial cell types of a circuit (see here); 2) Mapping their connections in the nervous system (long-distance, local, synaptic); 3) Monitoring the dynamic electrical and chemical activities in these neuronal and glial cell types during behavior, and extracting meaningful patterns i.e. "neural codes" or "circuit signatures" from them; and 4) Manipulating these activity codes in specific cell types to probe their causal role in those behaviors. These are all identified as high priority research areas in the interim report.
In recent years, several disruptive technologies have not only accelerated this four-pronged effort, they have emboldened researchers and policymakers to conceptualize this aspirational BRAIN Initiative. The following is a small sampling of these "foundational technologies" to illustrate the kind of interdisciplinary approaches that will be required to understand the brain at the level of neural circuits.
Making Brains Transparent
Genetic engineering and fluorescence microscopy have been the two great allies of researchers in unraveling the mysteries of the brain. While genetic engineering allows brain cells of interest to be labeled with (genetically encoded) fluorescent markers, fluorescence microscopy facilitates their selective imaging at high resolution. One of the challenges in mapping neural circuitry is that the brains of most commonly studied animals (e.g. mice), and humans, are opaque. Several important structures are buried deep within the brain and the only way to access them (with light and molecules) has been to physically take the brain apart by sectioning it into thin slices. This causes breaks, and results in the loss of 3D perspective while mapping neural connections. Karl Deisseroth’s lab (HHMI/Stanford) circumvented this problem by transforming intact post-mortem brains into transparent hydrogels, thereby enabling 3D, high resolution, brain-wide mapping of neural circuits. The clever chemical and material engineering tricks used in their method, CLARITY, are explained in this video:
Imaging Whole Brains
So, how many neurons does one have to image to get an authentic picture of the brain in action? Thousands? Millions? Every single neuron in the brain? Now, that's a raging debate in systems neuroscience. Philip Keller's and Misha Ahrens's labs (HHMI/Janelia) made an audacious attempt to find out by performing the first ever whole-brain imaging of dynamic neural activity at single neuron resolution in a vertebrate animal. The researchers employed high speed light sheet fluorescence microscopy to image immobilized zebrafish that were genetically engineered to express a protein (GCaMP) that reports neural activity by giving off a fluorescent signal (see video below). By analyzing the flickering patterns of fluorescence in individual neurons across the entire volume of the brain, the researchers identified two neuronal networks in the hindbrain that may be involved in movement. Truly, a landmark first glimpse into what whole brain imaging can reveal.
The World's Smallest Brain Observatory
Sometimes, to think big, you have to build small. A major goal of the BRAIN Initiative is to monitor and link neural activity in a thinking, working brain to the behaviors that it is producing. One obstacle is that standard fluorescence microscopes are bulky and non-portable. Enter microoptics. Mark Schnitzer’s lab (Stanford/HHMI), and later Inscopix, built the world's smallest fluoresence microscope to spy on diverse brain regions of mice engaged in natural behaviors. For example, with these head mounted fingertip microscopes, the researchers repeatedly imaged activity in thousands of hippocampal neurons as mice were exploring a familiar maze (see video below). They observed something truly remarkable: It was actually possible to deduce the exact location of the mice in the maze by simply looking at which neurons were firing! This fascinating insight into how large hippocampal neuronal ensembles code for a point-by-point spatial map of the external world is additionally important for understanding spatial memory impairment in Alzheimer’s Disease.
Controlling Behavior With Light
If there is one technology from the last decade that exemplifies brain research at its game-changing best, it is Optogenetics. Also developed in the Deisseroth lab, optogenetics allows researchers to manipulate neural activity in freely behaving animals, at will, with beams of light. In this technique, brain cells of interest are genetically engineered to express light sensitive proteins (ion channels called opsins) that are normally found in algae. These "optogenetic actuators" can either trigger neural activity (e.g. channelrhodopsin) or suppress neural activity (e.g. halo or archaerhodopsin) when light of specific wavelengths is shined on cells expressing them. Because optogenetics enables rapid, reversible, and precise control of neural activity, with the animal's behavior serving as an immediate read-out, researchers are now able to investigate "causality" with unparalleled sophistication. The video below explains this powerful circuit manipulation technique, in detail.
To Eat Or Not To Eat
And finally, here’s a brilliant example of how optogenetics helped solve a 50 year old brain mystery. Researchers have known since the 1950’s that electrical stimulation of a brain region called the lateral hypothalamus (LH) induces mice to eat, even if they are well fed. However, the exact neural circuitry involved was unknown. Garret Stuber's lab (UNC Chapel Hill) investigated whether connections between the brain’s emotion center, the amygdala, and the LH, were involved. They genetically engineered a specific amygdaloidal neuronal population (BNST GABAergic), which communicates with another distinct neuronal population (glutamatergic) in the LH, to express either channelrhodopsin or archaerhodopsin. When well-fed channelrohdopsin expressing mice were stimulated by shining blue light on the amygdaloidal axons in the LH, they exhibited voracious feeding (see video below); Conversely, when food-deprived archaerhodopsin expressing mice were stimulated with yellow light, they exhibited reduced feeding, despite hunger. This study is additionally important for understanding how dysregulated neuronal activity can lead to eating disorders such as anorexia and binge-eating, and obesity.
Of Things To Come
The technologies described above, although transformative, are far from perfect. They need to be made cheaper, faster, non-toxic, scalable, smaller, simpler, lighter, non-invasive, or human-relevant. The hope is that by the end of the BRAIN Initiative, these and other tour de force technologies will be refined by orders-of-magnitude enabling their routine use in brain research. So, for example, we can certainly expect even smaller microscopes with larger fields of view, neural activity reporters with higher signal-to-noise ratio, a bigger arsenal of optogenetic actuators for concurrent manipulation of multiple neural circuits, multimodal technologies for simultaneous circuit imaging and manipulation, and superior algorithms for extracting neural codes for even our most esoteric cognitive behaviors. And, then of course, there will be the technologies that we cannot even imagine today.
All these technologies, really means to an end, will vastly advance our understanding of human behavior and cognition. This new body of knowledge will build the foundations for a more sophisticated analysis of brain disorders, from Autism to PTSD to Alzheimer's, and perhaps, even act as a catalyst for entirely new kinds of therapeutic interventions in the decades to come.
The future is knocking at our doors and the time to seize it is now.
The author’s views are entirely his or her own and may not reflect the views of Inscopix.