The role of disinhibition in sensory processing

The brain is composed of billions of neurons that make trillions of synaptic connections. So how can we come to understand such a complicated organ? One way to deal with this complexity, is to study small components of the brain. Just like a computer, the brain is built of basic circuit components that are used for multiple purposes. If we understand the role of these circuits we will be closer to understanding the entire brain. When information from our senses, such as light and sound, enters our brain it gets routed around in the form of electrical impulses. When this storm of electrical activity is out of control this would lead to an epileptic seizure. Therefore, the brain has regulatory mechanisms that balance electrical activity. The neurons that route information around are called excitatory neurons, the neurons that keep these neurons from going out of control are called inhibitory neurons. But inhibitory neurons perform a lot more complicated functions too. In fact, there are many different kin ds of inhibitory neurons, possibly hundreds, and they may all have different functions, most of which are not understood. I will study the function of inhibitory neurons that are called Vasoactive-Intestinal-Peptide neurons (VIP cells). These cells have an unusual feature; Instead of inhibiting excitatory neurons, they inhibit other inhibitory neurons. I hypothesize that these cells can make us more sensitive to sensory stimuli. This could be a mechanism for how attention works, and how alertness is reg ulated. It would be difficult to study these cells in humans, but fortunately the brains of rodents have a remarkably similar architecture, and also have VIP cells. In this project, I will train mice to discriminate sensory stimuli while I manipulate t he activity of VIP cells and monitor how it affects their behavior. This will reveal the function of these cells in normal healthy brains, and will bring us closer to understanding how our brain underlies behavior. In the first phase of this project, we have developed a new behaviour task for mice that tests their sensitivity to tactile stimuli. Mice are trained to use their whiskers to discriminate how far an object is located from their snout. Mice learn this in about 1-2 weeks. Currently we are manipulating the activity of VIP neutrons to test whether the sensitivity of their whiskers has increased. Furthermore, a device is being fabricated, in collaboration with the Institute of Computer Science, that monitors the movements of the whiskers and excites VIP neutrons only when they whisk. In the second year of the project, the behavior setup is completed, so is the advanced whisker tracking system that can capture pictures of whisker movements at a 1000 Hz and at high spatial resolution. Since we will record neural activity in awake mice we have fabricated a device that can measure fine electrode movements at nanometer resolution. The simple optical device to measure displacements is ~10 times cheaper than any other commercially available device and is therefore submitted as a declaration of invention (DOFI) with Invent2 and is currently under review. The fabrication of this device will be published in 2017. Finally, everything is now in place to proceed to our full-scale planned experiments. In the third year of the project we have obtained recordings of hundreds of neutrons in the barrel cortex. We have used optogenetics to manipulate the activity of neutrons during behaviour. Currently the data is being analysed. We are identifying VIP neurons with immunolabeling techniques. Advanced analysis methods are in place to identify the role of inhibitory cells in the barrel cortex during behaviour. Finally, we have developed network models to provide a mechanistic understanding of our experimental observations and to make predictions that can be tested with new experiments. In the fourth year, we have further developed these computer network models. We discovered that the dendrites of inhibitory neurons are important determinants of network rhythms in the brain. We found that a subset of inhibitory neurons have dendritic properties that makes brain oscillations more robust against heterogeneities such as difference in synaptic connectivity strength and synaptic drive.

See final Results report attached.

One of the most exciting problems in science is to understand how the brain converts sensory information into perceptions and actions. This is truly challenging because the brain is composed of billions of neurons and trillions of synaptic connections. On e way to conquer this complexity is to study small clusters of neurons that connect to each other in a repeated pattern, such as the connections between excitatory and inhibitory neurons. These circuit motifs are simple and tractable and may serve as a ba sic building block used for multiple functions throughout the brain. However, because our understanding of such circuits is mostly based on in-vitro data, their role in behaving animals is poorly understood. In this project I propose to study a circuit c omposed of inhibitory neurons that inhibit another type of inhibitory cells, which is expected to increase network excitability by disinhibition. I will focus on inhibitory neurons that express vasoactive-intestinal peptide (VIP), and test the hypothesis that these neurons are excited by long-range connections to disinhibit their local circuits. This could create permissive windows of increased excitability that could be an attentional mechanism, and may be important for learning. This hypothesis is in pa rt based on preliminary data that I obtained as a Junior Fellow at Janelia Farm. To test this, I will use optogenetics to manipulate the activity of VIP neurons in the mouse barrel cortex during a perceptual learning task. I will study how this affects b ehavior, and I will use large-scale extracellular recording to detect changes in the network activity. These data will be used to build network models to provide a more mechanistic understanding of disinhibition. These results will provide two major adva nces: First, they will add to our understanding of sensory processing, and hence will help to understand mental diseases. Second, the methodology could be a testbed to reveal the functions of other cortical motifs.