Brain in a test tube 04.06.2015

We can learn about what is happening inside the brain with the help of functional magnetic resonance imaging (fMRI) - it allows you to see activity in certain parts of the nervous tissue and quite accurately compare this activity with the performance of a particular task. But we will not be able to learn everything about the brain if we do not penetrate to the cellular level, to the level of neurons and interneuronal contacts - synapses, to the level of auxiliary glial cells, which not only nourish neurons, but also interfere with the conduction of a neurochemical signal. And it should be remembered that there are many neural varieties. For example, if we carefully examine the cerebral cortex, we will find six layers in it, differing from each other in the ratio of neurons of different types. To understand how the higher cognitive functions are realized at the molecular-cellular level (namely, the cortex is engaged in them), we need to understand the structure and the relationship of its layers to each other to the subtleties.

Something, of course, can be studied on the brains of rodents and primates. In addition, the interaction of neurons is often studied in cell culture: cells live in a nutrient medium at the bottom of some laboratory vessel, and neuroscientists monitor how, for example, the strength of their synapses changes in response to certain stimuli. As a result, some conclusions can be drawn about the causes of schizophrenia, autism, and other cognitive impairments - after all, in the case of such pathologies, it is the neural architecture, the interconnection of neurons with each other, that is violated. But a flat layer of cell culture is still not a bark with its six layers. Another way is to analyze samples taken from deceased people. Needless to say, here one must always remember about post-mortem changes in the cellular structure, and it is impossible to study signal conduction in such samples. Ideally, we would like to have in our hands a three-dimensional cellular model that completely recreates one or another element of the brain structure, if not the entire brain. The experiments of researchers from Stanford University bring us closer to this ideal.

Of course, the matter was not without stem cells - Sergiu Pasca (Sergiu Pasca) and his colleagues received induced stem cells from human skin and then turned them into neurons. Now this is almost a standard procedure: differentiated cells are forced to "remember their youth", when they were stem cells and could not do anything but divide. But they can be turned into any other cell type, you just need to direct them along the right path using molecular signals. At first, everything went as usual: artificial stem cells grew flat in a culture dish. But then they were separated from the bottom and transplanted into a special new "place of residence", where they could no longer firmly attach to the walls or to the bottom. Within a few hours, the cells united into microballoons, in which they continued to divide. And here they started turning into cells of nervous tissue.

After seven weeks, 80% of the cells, by molecular and other characteristics, became similar to nerve cells. Moreover, 7% turned not into neurons, but into glial astrocytes, which support and nourish neurons, protect them from the penetration of harmful substances from the blood, and also regulate neuronal activity. Until now, it was not possible to grow both neurons and the cells that support them from the same stem material, you had to use third-party astrocytes obtained from a different stem cell line, which meant that genetically both turned out to be different - whereas in the brain all cells carry the same genes . Now, apparently, this difficulty will disappear.

But the most important thing became clear when they analyzed the structure of cell complexes (they were called cortical spheroids) - it turns out that their architecture was similar to that which is in the cerebral cortex. Moreover, 80% of neurons responded to an external stimulus, and 86% demonstrated spontaneous activity and formed neural chains with each other, transmitting a signal to each other. In other words, it was possible to obtain a fairly plausible three-dimensional model of the cerebral cortex.