October 21, 2005
Brains response to visual stimuli helps us to focus on what we should see, rather than all there is to see
La Jolla, CA - Delving ever deeper into the intricate architecture of the brain, researchers at The Salk Institute have now described how two different types of nerve cells, called neurons, work together in tiny sub-networks to pass on just the right amount and the right kind of sensory information.
Their study, published online by Nature Neuroscience, depicts how specific types of inhibitory neurons in the visual cortex of a rat brain are wired to, and "talk" with, discrete excitatory neurons. They also show how that "conversation," aimed at keeping the right balance of chemical signals, often excludes surrounding neurons."The inhibitory neurons are not just brakes, they can also be used to steer." said co-author Ed Callaway, Ph.D., associate professor in Salk's Systems Neurobiology Laboratories. For example, in vision, inhibitory responses in the visual cortex help people to focus on what they want to see, rather than all there is to see, he explained.
This new study is filling in the picture of how the brain is organized into "smart" efficient networks, and researchers hope that details of this complex design might, one day, uncover the roots of such neurological diseases as schizophrenia.
"We know already that schizophrenia is a problem with organization of inhibitory circuits of neurons, and now we are uncovering how these specialized nerve cells work together and with other neurons," Callaway explained.
"By understanding the brain in finer and finer resolution, we can then trace what happens when these neural circuits are mis-wired," he added.
The Nature Neuroscience report is the latest published study in a series by Callaway and first author Yumiko Yoshimura, of both Salk and Japan's Nagoya University, that reveal how neurons in the brain's cortex are finely wired to pass on thought and perception.
The brain cortex is the folded tissue that looks like the outside of cauliflower, but which in humans has a total surface area of about five feet, if stretched out. The cortex is separated into large areas of specialized function, such as the motor and visual cortex, but is also organized on a finer scale into vertical "functional columns" within the .05-inch thickness of the cortex.
Research in the 1960s showed that these columns contain brain cells with similar functions, suggesting that "like-minded" neurons needed to be networked together to perform a function. To account for the tasks carried out in the larger areas, scientists assumed that the function of these columns varied smoothly across the cortex surface. Later research demonstrated that each of the six layers in the cortex is also wired into distinct circuits.
But Callaway and Yoshimura tested the prevailing notion of columnar function by devising an experimental method that used glass microelectrodes to "listen" to two neurons at a time. They used rats, whose brain organization mirrors that of a human, and dissected tiny vertical slices to preserve the circuitry. With fine glass pipettes, they recorded tiny impulses in individual live neurons and listened to their responses when other neurons were optically stimulated. If impulses occurred at precisely the same times in both neurons, this indicated that they both shared the same inputs.
The Salk scientists found that the thousands of neighboring neurons that make up these columns are not the same, and they often don't communicate directly with each other. In February, they reported in Nature how excitatory neurons were organized into sub-networks that were found within the same column but which had nothing to do with other cells nearby.
That meant that the brain's circuitry is organized on a much finer scale than was previously suspected. And this makes sense, explained Callaway, because neuroscientists were puzzled as to why so many neurons were needed in the same part of the brain to carry out the same function.
"But if you realize that the brain has ten times as many subsets of neurons, it is doing ten times as much computation, and is that much smarter," he noted.
In this study, Callaway and Yoshimura sought to look at the networks that pair together excitatory and inhibitory neurons. These neurons work directly with each other to shape responses to stimuli. This "go-no" kind of interaction is necessary, Callaway says, or a positive feedback of chemical signals across neuronal synapses would result in a myriad of disorders, including epilepsy.
They simultaneously measured activity in both a specific type of inhibitory neuron, called fast-spiking, and a neighboring excitatory neuron. They then stimulated one or the other neuron and measured responses in the second neuron. They found that connections from the fast-spiking neurons were six times stronger when both cells were interconnected than if there was only a one-way inhibitory connection "This demonstrated that neurons primarily inhibited just the cells that excited it, and that tells us there is specificity in these fine-scale circuits." They then went on two show that these neuron pairs with two-way connections belonged to the same fine-scale subnetworks. "This means that inhibitory circuits can also precisely influence the activity of selected sub networks."
Callaway says the study demonstrates that neuroscientists "are really getting down to real nuts and bolts of cell type specificity, meaning that different types of neurons have different function, just like different blood cells perform different roles," he said. "We are not satisfied any more just to say what happens within a brain area. It is much more complicated _ and interesting _ than that."
On the World Wide Web: