How do nerve signals morph into meaning? asks S.Ananthanarayanan.

Do you and I perceive the same image when we see an object? We both call it with the same name, but what in fact to we see? By what process does the brain interpret signals from the senses?

Researcher Jerry N Chen, from Boston University, with colleagues, Cameron Condylis, Abed Ghanbari, Mikita Manjrekar and Karina Bistrong, and Sehnqin Yao, Zizhen Yao, Thuc Nghi Nguyen, Hongkui Zeng and Bosiljka Tasic from Allen Institute for Brain Science, Seattle, describe, in the journal, Science, their sally into the unknown territory of the cells in the brain, to unravel a part of the mystery. And they identify a kind of brain cell that plays a role in routing nerve signals from a first layer to specified cells in the next layer.

Nerve cells, or neurons, are different from other cells in that they are channels of communication. What this means is that they can receive signals from another cell and then transmit signals to yet another cell. Unlike a muscle cell, for instance, which reacts by contracting, when triggered, not by passing on the message to another cell.

How nerve cells go about this action is through electrical activity in the cell and parts of the cell that communicate with other cells. In the normal state, the nerve cell has an excess of positive, potassium ions and some large, negatively charged protein molecules. And a deficiency of positive, sodium ions and negative, chlorine ions. And the relative concentration of the charged ‘ions’ is the reverse in the medium outside the cell. The cell is hence some 70-80 mV negative compared to the outside and there is electrical ‘tension’ on the two sides of the cell wall.

The arrival of a signal, from the environment or a neighbouring cell to a receptor of a cell, however, causes gaps, called gateways, in the material of the cell wall to open. This allows positively charged sodium ions outside the cell to rush in, and the net change is reversed to about +40 mV. When this level is reached, the cell ‘fires’, a signal to the next cell, and the charge level drops. The change opens another set of gateways that allow positive potassium ions to rush out, to the potassium deficient exterior, and the negative polarity of the cell is restored.

There is, however, great diversity in the types of neurons, for different purposes, and hundreds of thousands of kinds within the brain. “The diversity of cell types is a defining feature of the neuronal circuitry that makes up the areas and layers of the mammalian cortex,” say the authors, about the structure of the brain. There are now methods to identify the genetic clippings that cells generate, and identify single cells. Using these methods, the cells in layers in the cortex, or outer surface of the brain have been ‘profiled’ and classified, the paper says, into some 300 different groups.

A question that the group was seeking to answer was how nerve signals that are received from peripheral sources, in the present case, the whiskers of experimental mice, are passed from layer to layer in the brain, to be integrated with records of earlier encounters, stored in deeper layers of the brain. The task, the paper observes, calls for measuring the activity of neurons while the animal is sensing things, which is to say, is live and active. The existing methods, although they can deal with individual neurons, can only identify the type of the neuron, not its function, the paper says.

The team has then developed a technique, Comprehensive Readout of Activity and Cell Type Markers, with the acronym, CRACK, which evokes the idea that this may help ‘crack’ the code the brain employs to combine nerve signals with memory and arrive at perception. CRACK has two components - the first is two-photon calcium imaging, followed by multiplexed fluorescent in situ hybridization (also known as FISH).

The first component works with charged calcium ions, which form a signalling mechanism, in addition to the sodium ion channel. The experimenters detect activity of calcium ions by a sensitive and accurate imaging technique called two photon microscopy. Imaging of cells usually employs fluorescence, where certain atoms absorb light of one frequency, and de-excite by emission, but at a lower frequency. An example is the domestic tube lamp, where electric discharge creates UV light. The coating of the tube absorbs UV and emits at frequencies in the visible range. In two photon imaging, the excitation is not by one high energy photon, but by a pair of lower energy photons, generated in quick succession by a laser. This method permits more focussed imaging, with less ‘background’, and imaging a little deeper within a tissue sample.

Calcium ion detection is combined with imaging markers that attach to specific parts of the DNA of cells, the technique called FISH, referred to earlier. With both components working, Crack is able to image the action of cells and to identify the type of the nature of the cells.

The team used CRACK to observe the working of specific cell types in layers 1 and 2 of the brain cortex of experimental mice. As the imaging is done non-invasively, the team could record neural excitation when the mice were active, feeling around themselves with their whiskers, which are their sensitive, tactile equipment.

The team started with an existing catalogue of neuron types found in the mouse brain, created by the Allen Institute of Brain Science, which listed the cell types, but only the types, without their function. The work of the team hence adds a layer of information, the activity pattern of the different cell types – the reason for ‘Comprehensive,’ the first ‘C’ in the acronym CRACK.

The team profiled the activity of eleven kinds of neural cells in the mouse cortex, in terms of changes in task-related properties, along with differences in molecular type. One particular cell type, Bazla, was found to be “a highly active detector of tactile features.” Simultaneous imaging of activity of more cells then revealed relationships between groups of cells. And it was found that Bazla acts as a kind of ‘hub’, to “orchestrate local neural activity patterns,” the paper says. And other tests, with mice where the whiskers had been removed, showed that Bazla was able to adapt and compensate for the change.

“The ability to map functional and transcriptional relationships across neuronal populations provides insight into how the organizing principles of the cortex give rise to the computations it performs,” the paper says. The exponential growth of capacity for computing has promoted Artificial Intelligence, which find its inspiration in theories of how the brain functions. The relationships between the software ‘neurons’ in an AI system are programme defined, but this is not so in the animal brain. The current work is a step towards understanding how it works

------------------------------------------------------------------------------------------ Do respond to : response@simplescience.in

-------------------------------------------