How are neural circuits created

How brain researchers approach the circuits in the brain

In the nine months in which a fetus grows up in the mother's womb, an unbelievable program takes place: the roughly 86 billion brain cells that make up the human brain are put together according to a precisely orchestrated blueprint that is laid down in the genome. "Between the third and sixth month, the stem cells develop up to 1000 neurons per second," says Simon Hippenmeyer, neurobiologist at the Institute of Science and Technology (IST) Austria in Klosterneuburg.

"It is the most fascinating and beautiful organ in our body," says neuroscientist Gaia Novarino, also a group leader at IST Austria. Looked at soberly, one can say that the adult brain weighs about one and a half kilograms and uses 20 percent of the oxygen in our blood. But what exactly is going on in this constantly changing thicket of neurons that communicate with each other via electrical surges over 100 trillion synapses is still one of the greatest puzzles of mankind.

What is certain is that the DNA "only" provides the basic building blocks and implements a system that is designed to rewire its circuits with every piece of information. The world that surrounds us, our experiences and sensory impressions permanently reshape our brain and its connections - in a highly efficient way. The US psychologist William James called this dynamic and adaptive system "plasticity". The neuroscientist David Eagleman from Stanford University calls it "Livewired" in his latest book "Livewired" (2020), analogous to hardware and software.

Black box in your head

It is true that brain research has achieved considerable success in recent years, particularly with regard to the understanding of motor and sensory abilities. The measurement of brain activity and imaging processes make it possible to track down feelings, ideas and thought patterns. And yet there is a huge gap between what we know about our thinking organ and what makes us who we are. The so-called higher cognitive abilities, the consciousness, the memory, the thinking in itself, are still a black box.

German brain researcher Wolf Singer is convinced that the results of modern neurosciences and a multitude of new technologies make completely different research approaches possible today. In contrast to previous ideas, we now know that the brain is "an extremely distributively organized system", "that finds its way around without a conductor, but organizes itself," said the emeritus director of the Max Planck Institute for Brain Research in Frankfurt Year at a lecture in Vienna. "We have replaced linear systems theory, we are now more in the world of complexity theory," says Singer.

How to learn synapses

So, thanks to new technologies, is brain research at the beginning of a new era, as Singer postulates? "Technology is the key," says neuroscientist Peter Jonas from IST Austria. In 2016 he was awarded the Wittgenstein Prize by the Ministry of Science and Science Fund (FWF) for research into the basic mechanisms of synapses in the hippocampus.

He describes as one of the "breakthrough events of the last few years" that synaptic structures and their processes can be measured directly in tissue - spatially in nanometer resolution and temporally in microseconds. One of Jonas' primary goals: "We want to find out how synapses behave in order to cause higher brain functions, for example how incomplete information is supplemented with the help of the network in the hippocampus - one of the basic functions of memory."

It is important on the one hand to measure the activity in the nerve cells and on the other hand to see how they change. Jonas has taken a leading role in developing a number of technologies: The "subcellular patch-clamp technique" enables electrical signals that are transmitted via the synaptic gap to be intercepted directly at the starting cell and at the same time to observe the response at the target cell. "We can currently measure the processes in an ensemble of eight cells at the same time. That is a nice way to study the interactions in such a mini-network," explains Jonas.

Memory bubbles

With the "Flash and Freeze" method developed in 2013 and refined by Jonas, neurons are stimulated with light and snap-frozen within a thousandth of a second so that they can be examined under the electron microscope. Because in order for brain functions such as remembering to be able to be exercised, something in the structures of the neural network has to change.

Jonas and his team were recently able to show what such a change looks like: They recorded vesicles with neurotransmitters, the displacement of which at the synapse is linked to short-term memory. "This suggests that information is stored in the form of such vesicle pools," says Jonas. A step to get on the track of the brain's learning process.

In addition to technical advances in electron microscopy and the measurement of signals directly on the individual cell, further developments in genetics have also been groundbreaking in recent years - above all high-speed gene sequencing and the Crispr gene scissors. Using optogenetics, of which the Austrian neurophysiologist Gero Miesenböck is one of the pioneers, the activity of neurons can be controlled with light. Neural circuits can thus be examined on a whole new level.

The Mosaic Analysis with Double Markers method (MADM method), which Simon Hippenmeyer decisively further developed, is based on genetically modified cells that are provided with color markers. He and his team can use the mouse model to follow brightly colored neurons on their way from the stem cell to the cerebral cortex. In this way, it is possible to visualize how cells with built-in genetic defects that lead to serious brain diseases in humans behave in comparison to healthy cells.

"It's an exciting time for brain research," says Hippenmeyer. "With today's technologies, we can observe a million cells developing at the same time. That was not possible five years ago." The MADM method will in future be used to systematically examine how all genes influence brain development at the cellular level. Even if the human genome differs only slightly from that of the mouse - which codes make the human brain so extraordinary is still in the dark. Other approaches are needed for this, says Hippenmeyer.

Autism research on protograins

Gaia Novarino, who is dedicated to researching the still unknown causes of autism, is pursuing one of these. She and her team reprogram brain and blood cells from children with autism into stem cells. The neurons developed from this are grown into organoids, i.e. tiny proto-brains, in culture dishes. "This allows us to examine the neurons of an individual without being invasive," says Novarino in the IST Austria Science Talk. "This enables us to examine processes that take place very early in the development of the brain in the laboratory and to see what happens in connection with a mutation."

For the theoretical neuroscientist Tim Vogels, who moved from Oxford University to IST last August, this is a "revolutionary technique", since with this method "the simplest possible circuit diagram can be developed" which can then be expanded step by step. Since such experiments can also be simulated in the model, a direct comparison between theory and practice is possible.

From cell to behavior

While neurobiologists feel their way down to the smallest structures within the brain cells in order to then work their way to larger networks, psychologists and other cognitive researchers focus on what the brain produces, i.e. primarily behavior. Here, too, in the past few years there has been a move not only to measuring the activity of individual brain regions of test subjects when performing a wide variety of tasks, but also to looking at connectivity, i.e. the connections and networks in the brain.

Imaging processes, algorithms and machine learning methods with ever better resolution enable ever more precise analyzes of higher brain functions, especially learning and memory - and give hope to better understand diseases such as Alzheimer's and dementia.

After all, one of the great potential for breakthroughs in brain research lies in computer science. The huge amounts of data that are generated in experiments can be used to reproduce at least small areas of the brain on the computer and to analyze them "in silico" - in order to achieve a better understanding of the thought processes. Conversely, artificial intelligence research benefits from the results of neurobiologists. It is no coincidence that artificial neural networks are based on processes in the brain.

In order to understand the incredibly complex network in our heads, different disciplines have to be networked further. Because even if we are still far from an overall understanding of the complex miracle of the brain and there are still great rifts to be overcome between the nerve cell and human behavior - the new technologies that are prevailing at all levels suggest that a new, profound one The epoch of brain research has long since begun. (Karin Krichmayr, January 31, 2021)